Relaxing whilst doing Competition Law is not an Oxymoron

Android meets Pronuptia, or why software licensing is like a franchising agreement

with 9 comments

Pronuptia and Android

Imagine the European Commission issued a Statement of Objections against McDonald’s saying:

The McDonald’s brand is commercially important for McDonald’s franchisees. In its franchising agreements, McDonald’s has made the licensing of its trade mark conditional on selling only its own products. As a result, competing products cannot become the default fast food served at McDonald’s restaurants. These clauses deprive consumers of more interesting, potentially superior, fast food combinations. Just think of how amazing it would be to have McDonald’s, Chipotle and Pizza Hut all in the same store. In addition, the Commission is concerned that McDonald’s insists on making money out of its licensing activities’.

The EU competition law community would unanimously disapprove such a Statement of Objections. Yet, if you think about it, this is pretty much what the Commission argues in its press release on Android. Arguably, the press release is more controversial, as it goes further (as if the above example did not go far enough).

It took me a while to realise, but the agreements concluded by Google and mobile phone manufacturers are essentially franchising agreements. Apple is a vertically-integrated operator that sells its own phones to create a unique ‘look and feel’. As is well known, it does not rely on third-party manufacturers. Google (like McDonald’s, Pizza Hut and Subway) uses various contractual mechanisms to mimic vertical integration and create a competing ‘look and feel’.

Android and the Pronuptia conditions

The Court of Justice realised early on, in Pronuptia, that franchising agreements are pro-competitive in the sense that they enhance inter-brand rivalry. It ruled that some clauses included in franchising agreements do not violate Article 101(1) TFEU at all. The Court did not say that they are not restrictive by object. It ruled that they are never restrictive of competition, irrespective of the market power enjoyed by the supplier (or franchisor).

In particular, the Court held in Pronuptia that any clauses that relate to the protection of the (i) know-how of the franchisor and (ii) the reputation and uniformity of the franchise are not caught by Article 101(1) TFEU. The restraints challenged by the Commission in Android could be examined in this light. Is the purpose of the alleged bundling of Google applications to create a uniform brand image across devices? Do the ‘anti-fragmentation’ clauses seek to preserve the reputation and integrity of the network?

Some people might react to the above by saying that the Android agreements are far less restrictive than the typical franchising agreement. Generally, a franchisor requires exclusivity from franchisees. Even then, the agreement does not fall within the scope of Article 101(1) TFEU. In Pronuptia, the Court accepted that (outright) exclusivity obligations may be necessary to protect the reputation of the franchise.

In Android, the Commission does not claim that Google requires (outright) exclusivity from third-party manufacturers. At most, the concerns relate to an alleged obligation to exclusively pre-install Google Search. This requirement, if established, would be considerably less stringent. If outright exclusivity would be acceptable in the context of a franchising agreement, why would a less stringent restraint be an issue?

The importance of the counterfactual

The Court understood in Pronuptia that, if a supplier were not able to protect its know-how and brand image, it would not use franchising. In the absence of some core clauses, franchising would simply not exist, which is why these core clauses are never caught by Article 101(1) TFEU. As other cases, Pronuptia is all about the analysis of the counterfactual: does the practice restrict competition that would have existed in its absence?

It is difficult to gather from the Android press release whether the Commission has assessed the conditions of competition with and without the restraints. This is (or should be) a fundamental step in the analysis. Google – it is not a secret to anyone – makes most of its money through advertising. By challenging, inter alia, its ability to bundle applications, the Commission is in fact challenging Google’s very business model.

Against this background, it would be necessary to consider what would have happened if Google had not been able to make money through bundling and/or by requiring that applications be pre-installed on third-party devices. Would Google have invested in the development of an ecosystem rivalling Apple’s if it had not been able to monetise its efforts?

If the answer to this question is no, then Google’s practices cannot be contrary to Article 102 TFEU. If the Android ecosystem would not have been created absent the clauses challenged by the Commission, the said clauses do not restrict competition that would otherwise have existed.

Now that Android is up and running the Commission may believe that the world could be a better place if some contractual restraints were removed. But looking at the matter from an ex post perspective alone is misleading. One cannot have it both ways. A world with the restraints may look imperfect to some, but it is definitely better than a world without Android. A proper analysis of the counterfactual should look at the practice from an ex ante and an ex post perspective.

Ex ante thinking transpires from Pronuptia and other Article 101(1) TFEU rulings such as Nungesser. An ex ante approach to the analysis of practices also explains why a refusal to license an intellectual property right is only abusive in ‘exceptional circumstances’. If the Court had looked at the matter solely from an ex post perspective in Magill, every refusal to license would be abusive: ex post, every single compulsory licensing obligation promotes competition and benefits consumers.

Yes, but Google is dominant: a ‘special responsibility’ not to make money?

One could argue that all of the above is irrelevant, because Google is allegedly dominant in one or more of the markets covered by its practices. Some people like to claim that, when a firm is dominant, anything goes. After all, it is argued, Article 102 TFEU cases are very different from each other and dominance is an exceptional occurrence.

I do not see things that way. It would be a mistake to stop thinking rationally simply because a case is about a dominant firm. The logic behind Pronuptia is not less compelling just because Article 102 TFEU is potentially applicable to an agreement. Arguably, the case for protecting the know-how, reputation and uniformity of a franchise is stronger when it has become really successful.

I am also uneasy with intervention that questions the business model of a company, even if the company is dominant. The ‘special responsibility’ under Article 102 TFEU is a nebulous concept. But I do not believe the concept can be stretched as far as to support the claim that dominant firms have a ‘special responsibility’ to give up their way of making money (and thus to subsidise rivals).

Written by Pablo Ibanez Colomo

25 April 2016 at 10:28 am

Posted in Uncategorized

9 Responses

Subscribe to comments with RSS.

  1. Great post Pablo. Thank you! I will certainly use your mcdonalds example!


    25 April 2016 at 11:16 am

  2. One could argue that in the counterfactual Google could have charged for the use of Android when adopted without its Apps, which might have been passed on, at least partly, to consumers in the form of higher smartphone prices. But then this would only have worked if charging occurred after Android established itself as the anti-Apple main platform, which would have arguably been very difficult unless Android was initially offered for free to manufacturers – hence hard to start charging licensing fees ex-post…?
    I am not sure that at the time Google had much choice really: at the time there where other alternatives vying for the title of anti-Apple platform in the smartphone Op System arena.
    I think the main point here is that Google has taken a big entrepreneurial risk.
    To me the main difference between Microsoft and this case is that in the former the tying product was the legacy monopolized one (the Op System) in order to pre-empt the next-generation threat (based on a web-browser). Whereas, here the tying product is a next-generation one, in order to maintain the prominence of the legacy monopolized (tied) products (mail, search and mapping).
    This is why I believe that Google’s tying is far less dodgy than Microsoft’s one to be honest.

    Paolo Siciliani

    25 April 2016 at 5:00 pm

  3. There is so much wrong with the McDonalds analogy that it is difficult to know where to start. In relation to the phone, Google is indeed supplying an important component but not the entire product, and Samsung and others would be surprised to see themselves described as mere franchisees of Google. But if we must pursue that indigestible burger analogy, try to think of Google as the supplier of the bun. Important indeed, but does that mean it should also be able to specify the supplier of the meat patty and the gherkins?

    Stephen Kinsella

    26 April 2016 at 8:50 am

  4. Dear Pablo,

    You write ´The Court did not say that they are not restrictive by object. It ruled that they are never restrictive of competition, irrespective of the market power enjoyed by the supplier (or franchisor).´. Where exactly in Pronuptia does the Court say the last ´irrespective of the martket power´bit? I must have missed that part in the judgment. In fact, wouldn’t you say that Google and Pronuptia can be differentiated because of the Court’s standard consideration that compatibility of such effects-restrictions depends on the ‘economic context (para. 27, point 1)?


    27 April 2016 at 8:53 am

  5. Thanks to all for your sharing your thoughts and comments. This is all very interesting.

    Dear Stephen: I am not sure that the relative importance we may wish to attach to a ‘component’ is relevant for the analogy. Apple only uses vertical integration (i.e. it only sells through ‘owned and operated’ products). As franchisors do, Google relies upon third parties. What matters for the analogy, in my view, is that Google needs contractual devices to address problems that Apple does not face. A franchisor needs to take measures to preserve the uniformity and reputation of its network, just as Google does. Similarly, the franchisor needs to take measures to ensure that its investments do not benefit competitors (‘even indirectly’, the Court held in Pronuptia, para 16). Free-riding concerns are crucial to understand Pronuptia, Cartes Bancaires and many vertical restraint cases. I believe your question goes to the heart of the problem. Typically, a franchisee is not able to cherry-pick which ‘components’ she gets from a franchisor (para 21, Pronuptia). She will not be able to say that she is happy to take the bun, but that a friend can find better patties at a lower price. Against this background, does it make legal and economic sense to take action allowing mobile phone manufacturers to cherry-pick which applications they want, even if this allows Google’s rivals to take a free ride on its investments?

    Dear Hans: I do not think Pronuptia leaves any scope for ambiguity. What needs to be established is whether the clauses are necessary to preserve the know-how of the franchisor and the reputation and uniformity of the franchise. The analysis of the nature of the agreement and the context of which it is part tells you whether the clauses of the agreement serve one of these two purposes. If they do, then the clauses fall outside Article 101(1) TFEU altogether. They do not restrict competition. If the clauses serve another purpose – for instance, market sharing among franchisees, which is the issue that Court discusses expressly – then a closer analysis may be necessary. Market power is nowhere to be found as a factor in Pronuptia. You may want to take a look at the Guidelines on vertical restraints – para 190 b) – where the Commission clarifies that it is not necessary to apply the market share thresholds to restraints that meet the Pronuptia conditions. The same is true for qualitative selective distribution (para 175), which under certain conditions also falls outside Article 101(1) TFEU altogether. This point is also very clearly explained by the Commission in the Guidelines on Article 101(3) – para 18(2).

    Pablo Ibanez Colomo

    27 April 2016 at 10:47 am

    • Hi Pablo,

      I’m aware of the Commission’s take. I’m just wondering where the Court says so in Pronuptia, so a reference would be greatly appreciated.

      Best wishes,


      27 April 2016 at 12:27 pm

  6. Dear Pablo, the argument you put forward is conceivable (I would even say persuasive) but if it holds true Microsoft should not have had any issues with its pre-installed default Media Player or Explorer.


    10 May 2016 at 4:25 pm

    • Hi Almos,

      Thanks for your comment! Please stay tuned, as I will be writing something else addressing the point you make! As a quick reaction, allow me to mention that there are differences between Microsoft and Android that make the latter more exciting intellectually.

      Pablo Ibanez Colomo

      11 May 2016 at 7:01 pm

      • Dear Pablo, looking forward to reading your follow-up on this! Keep on the good work:)


        25 May 2016 at 12:54 pm

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: