Therefore, to be possess’d with double pomp,
To guard a title that was rich before,
To gild refined gold, to paint the lily,
To throw a perfume on the violet,
To smooth the ice, or add another hue
Unto the rainbow, or with taper-light
To seek the beauteous eye of heaven to garnish,
Is wasteful and ridiculous excess.

Advocates can excel at this, both public and private; often not to the benefit of the positions they advocate.

A notifying party once wanted to define the relevant product market on one of my cases as “leisure time activities”.  It would have included everything from taking a nap, walking the dog, grabbing a beer, or watching television. Among other things. Unsurprisingly the market share of the notifying party would have been – on this definition – remarkably small. De minimis small. Which for a company regularly referred to as dominant, if not a monopolist, showed remarkable chutzpah on the part of the lawyers.

Now it’s true that at a certain point, everything becomes substitutable, but that’s not the basis on which markets work and companies compete. No competition authority could reasonably adopt that market definition. No CEO would use it to discuss with his board the strategic direction of the company.

This gilded market definition was easy to dismiss, and had the case come to court it would have fared badly. A market definition more narrowly focussed on a subset of “leisure time activities” would have been harder to dismiss. This gilding worked against them.

There’s always the temptation when you are trying to win an argument to throw everything at the wall and see what sticks. Junior lawyers are often told that it’s hard to predict what argument will convince a judge, so they should include even those arguments that they think are weak. Spotting where arguments stop being weak and start being whimsical can be hard, especially in the heat of drafting.

The consequences can be unpredictable.

Two contrasting examples from Commission decisions are worth noting.

For the first, in determining whether a set of anti-competitive actions constitutes a “single and continuous infringement” the Commission has, when assessing a complex set of facts, concluded that it would be:

artificial to split up continuous conduct [by the undertakings concerned], characterised by a single purpose, by treating it as consisting of several separate infringements, when what was involved was a single infringement.

This language was used, for example, in the Commission’s decision in Aalberts. There, the Court of Justice used this language  against the Commission. Finding that Aalberts had not participated in one aspect of the infringement, the Court cited the above wording – that it would be artificial to split up the infringement – to support a conclusion that if one element of the infringement was annulled, then it all should be:

65      Nevertheless, it must be noted that the contested decision complains only that the respondents participated in a single, complex and continuous infringement. Thus, that decision does not qualify the participation of Aquatis in the FNAS meetings as an infringement of Article 81 EC.. On the contrary, recital 546 to the contested decision, which lists the anti-competitive conduct which that decision covers, does not contain any reference to the FNAS meetings. Furthermore, recital 590 to the contested decision expressly confirms that the Commission took the view that it would be ‘artificial to split up continuous conduct [by the undertakings concerned], characterised by a single purpose, by treating it as consisting of several separate infringements, when what was involved was a single infringement’.

 66      In those circumstances, even if the FNAS meetings had had an anti-competitive purpose or effects, that constituent element of the single, complex and continuous infringement could not be severed from the remainder of the measure within the meaning of the case-law cited in paragraph 64 of the present judgment

The Court therefore annulled the decision in its entirety.

The Commission could have used more moderate language – for example that several aspects of the behaviour could be seen as standalone infringements of Article 101(1) but that on balance the infringements appeared to be linked. The Court might then have felt that the Commission’s interpretation was more defensible, or – if it still disagreed that this was a single infringement – it might have regarded the annulment of one aspect as not calling into question the entire decision.

Arguing that a case is clearly black and white, when many are really shades of grey, is a bad idea.  The belt and braces approach – “this is not only bad, it is incontrovertibly and indisputably bad” – is usually less convincing in practice than a more nuanced approach.  As Henry Fonda remarked in Once Upon a Time in the West – commenting on a man who wore both belts and braces – “How can you trust a man who doesn’t trust his own pants?”

Courts may think the same about reasoning that they see is overdoing it.

Unfortunately, the risk of gilding the lily is not just that the argument will be disbelieved. Believing it may be worse, at least in the long term.

The Commission’s Microsoft / Skype clearance decision was righty upheld by the Court. But the Commission went beyond the core of the case and used some  arguments that give the impression of throwing everything at the wall to see what sticks.

Paragraph 92 reads as follows:

Most consumers of communications services make the majority of their voice and video calls to the small number of family and friends that make up their so called “inner circle”. According to Facebook data, users engage in regular two-way interaction with four to six people. Therefore, it is not difficult for these groups to move between communications services. Moreover the Commission observes that consumers multi-home to a certain degree among various providers of consumer communications services.

There are several potential problems with this argument:

  • is Facebook data a relevant proxy for the video communication markets concerned in the Skype case?

  • is the “inner circle” relevant? Could a company sell a communication service to consumers based on an ability to communicate only with an “inner circle”?

  • Would consumers migrate if they could communicate with just that inner circle and not others?

These potential problems are – relatively – specific to the case. But it is the last argument argument used in the extract above that is not a potential problem only in the context of the case but is more generally a problem for networked communications markets.

Does the fact (let’s assume it is true) that people communicate mostly in groups of four to six lead to the conclusion that switching from one provider to another simply requires moving that group of six and so is easily possible? I’m afraid not.

My group of contacts will usually be different to the group of contacts of each of my contacts. There’ll be some overlap sure, but if each of my six contacts calls five of the same people as I do, and also calls just one different person each, then the group of people you need to migrate to a competing service, simplifying just a little bit, grows to infinity.

This is why network markets risk tipping (a term not used much today). And this is why telecoms regulation in both the EU and the US mandate interconnection between all voice telephony operators.

Without interconnection, switching between networks is hard – much harder than just moving your six contacts to another network.

(There may be other arguments that make switching easier – multihoming for example, where users routinely use multiple communications networks that serve the same purpose. Service provider switching is certainly a more complex issue in the era of applications communicating over data networks than it was in the era of the PSTN.)

So the argument that you only need to move four to six people onto a competing network seems potentially problematic. But the General Court didn’t agree, upholding the Commission’s decision (rightly), but also supporting all of the above arguments (more dubiously).

Is this evidence that my argument is wrong? That gilding the lily sometimes works? Perhaps. But, I suspect, only in the short term. The precedent would be a bad one if not confined to the particular case.