Archive for September 2013
Some thoughts on the new anti-Google (Android) complaint (2/3): Predatory pricing claims
This is the second post in a series; click here for Post 1 (on background and dominance)
According to FairSearch (see here) “Google’s predatory distribution of Android at below-cost makes it difficult for other providers of operating systems to recoup investments in competing with Google’s dominant mobile platform“.
Unsurprisingly, this claim has spurred very strong reactions from the FOSS community, which regards it as a direct attack to the Open Source/FreeSoftware development model (see notably here, here and here). Android is indeed FreeSoftware, meaning not only that it is distributed for free, but also that it adheres to the so-called 4 freedoms: (i) the freedom to run the program, for any purpose; (ii) the freedom to study how the program works, and to adapt it to the user’s needs; (iii) the freedom to redistribute copies; and (iv) the freedom to improve the program and release the improvement to the public. This means that asking Google to start charging for Android would be akin to force it to stop supporting FreeSoftware.
A quick look, however, would reveal that this is a non-issue. It is undisputable that given Android’s FreeSoftware/public good nature Google doesn’t have the ability to set a price. The price is 0.
There are certainly interesting pricing issues to be discussed in the software industry, but, in our view, they arise with respect to proprietary software, not free software.
This should be enough to end the discussion, but if this interests you, click on the hyperlink below for more developed thoughts (if you’re lazy you can just stick to the arguments in bold to get the general idea):
Some thoughts on the new anti-Google (Android) complaint (Post 1/3)
At Chillin’Competition we have paid considerable attention to a number of IT-related competition developments, and –like most other followers of these matters in Europe and elsewhere- we have shown predilection to comment on the pending EC investigation over Google’s search practices. Nicolas, Pablo Ibañez-Colomo and myself have devoted tenths of posts to offering our –often conflicting- views on a number of issues raised in that case.
We –or at least I- had until now not really paid attention to the more recent FairSearch complaint regarding Android, and this despite the repeated warnings of Enrique Colmenero (our new associate and a geek who knows a bit about Android (he says not sufficiently well, I say it’s unbelievable), who was also the real author of my Google ppt), and who kept on telling me that the allegations in this complaint merited some public discussion. I first looked into it last week while writing the post about Skype’s integration with Windows, and realized that he’s right.
Given that all things Google raise the number of visits to the blog and spur more debate than other topics, we’re decided to comment on this yet non-case. We devoted a weekend to writing our preliminary views, and since the result is fairly lengthy we’ll be breaking the discussion into three separate posts: Today we will provide some background and deal briefly with market definition issues. Tomorrow we’ll discuss the predation claims. And Monday we’ll address the bundling allegations.
Before getting into substance, four disclaimers are necessary. The first is that by myself I wouldn’t have had the required technical knowledge to comment about this, so I’m borrowing Enrique’s (any errors, however, are only mine). The second is that we are not working for any party interested in this case and therefore comment on the basis of publicly available info (for fuller disclosure, some time ago I had two chats with someone on the complainants side as well as with someone working for Google; in both cases they let me know their views on the complaint). The third is that since we don’t want this blog to be a place to discuss cases in a seemingly one-sided way (much less when they are ongoing, like this one), we’ll be happy to open this platform to anyone willing to reason any disagreement with the opinions provided below. We don’t intend to defend a given position, but to reflect on issues that interest the antitrust community, and we are more than open to be persuaded that what we say is wrong. The fourth is that even if now criticize a complaint lodged by Microsoft FairSearch in the past we’ve also heavily critized complaints targeting Microsoft, like this one.
Bored already? If you’re stil reading I guess not, so let’s get started:
Some background to the complaint
Back in April the anti-Google alliance FairSearch (in this case only two of its members Microsoft and Nokia [Note: after I was done writing this post I learnt the news that Microsoft is acquiring Nokia’s mobile business] seem to have a real interest in the case) lodged a complaint with DG Comp alleging: (a) that by giving Android to device-makers for “free” Google engages in predatory conduct (making it difficult for rivals to recoup the investments made in developing competing mobile operating systems; and (b) that “phone makers who want to include must-have Google apps such as Maps, Youtube or Play are required to pre-load an entire suite of Google mobile services, and to give them prominent default placement on the phone”. Click here for FairSearch’s Press Release.
Rumor has it that the Commission recently sent out requests for information in relation to this complaint.
A business problem model?
In our view, this complaint can only be properly understood once one is aware about the existence of essentially 3 different business models for mobile operating systems (OSs). One is Apple’s vertically integrated model (iPhones run on Apple’s own iOS), another is Microsoft’s licensing model (OEM’s wishing to have smartphones running on Windows have to pay for a license), and the third is Android’s free software model (Android is distributed for free under a an open source license which enables licensees to do whatever they wish with the code), which has also been the model adopted by all new market entrants (Ubuntu, Firefox OS, Jolla’s Sailfish or Tizen –backed among others by Samsung and Intel-); Nokia’s Symbian (the market leader until 2011, now maintained by Accenture) was always and is also open source.
Manufacturers that are not vertically integrated at the OS level like Apple or Blackberry had to find a competitive OS, there being, until now, essentially two reliable options: Microsoft’s Windows (which they had to pay for), and Android (which OEMs obtain on a free-license basis; even if they have to pay some royalties….to Microsoft! ; some even say that Microsoft makes more money from Android than from the Windows mobile OS). Not surprisingly, the market tends to favor the open source model and, quite logically, Microsoft doesn’t like that (you’ll recall that it also “had issues” with open source OS for PCs). It’s against this background that the complaint comes, in what some see as an attempt to reverse the course of the business model that is proving most successful.
On market power/dominance as a pre-requisite.
Every press-clip citing FairSearch’s allegations refer to the claim that Android enjoys a market share of 70%. This is a bit equivocal. In reality, the fact appears to be that 70% of smartphones (leaving tablets, led by Apple, aside on the assumption that they belong to a different market) shipped in the last quarter of 2012 had Android. And in reality, usage market shares appear to show a duopoly of iPhones and Android phones (see here or here) rather than an Android monopoly; moreover, revenue-baded market shares clearly tilt the balance in Apple’s favor (as explained here) [As to the future trend: Android is certainly doing spectacularly well lately, but we bet iPhone sales will increase once Apple abandons its (rather Steve Job’s) exclusive-good marketing strategy, which is very profitable (see previous hyperlink) but has costs in terms of market share. Android phones sell very well, among other reasons, because they are often subsidized by operators; iPhones on the other hand have traditionally been quite costly. The moment iPhones are cheaper Apple’s share should increase significantly] So, in reality, Android seems to face rather intense competition from Apple’s iOS, Windows, Blackberry; even its main customer (Samsung) has also developed its own OS Bada/Tizen (it also “multi-homes” by licensing Windows for some devices).
Against the background of what would appear to be a competitive smartphone market, the way to come up with a monopoly-like share would require 1) to distinguish separate markets for tablets (where Apple is the leader) and smartphones; and 2) to also take Apple and Blackberry out from the smartphone-only calculation by defining a relevant market for licensable mobile OSs, which intuitively seems a bit of a Procrustean move.
More importantly, forget about market shares for a second. The truly relevant question is: does Android enjoy significant market power? Can it profitably raise prices or decrease output or innovation? Because Android is OpenSource/FreeSoftware (obtainable for free, its source code is entirely disclosed, it can be freely modified/”forked” [see here for “what the fork is forking”?] and appropriated by third parties: just look at Replicant, CyanogenMod, MIUI or at Amazons’ Kindle) we don’t see how Google would be able to exert market power in any way. Even Microsoft and Nokia could take Android and do what they please with it (they could even try to fork/improve it and compete with Google).
Actually, could we even say for sure that there is a “market” for licenseable OSs when all licenses (except Microsoft’s) are FreeSoftware licenses?
Moreover, and as regards innovation, there are very few markets with innovation cycles as fast as the one for smartphones’ OSs having featured a number of leaders in recent years: Palm, Symbian, iPhone, Blackberry and now Android. And this is because given the prevalence of FreeSoftware barriers to entry are extremely low. The moment someone comes up with a more innovative (better) product (including an improved version of Android unrelated to Google), Google would also lose its current lead.
But, for the sake of discussion, let’s assume that Android is dominant and look at the theories of harm, which bring up some interesting issues In our second post we’ll discuss the predatory pricing claims, and in our third post we’ll deal with the bundling aspects of the case.
Mens Sana in Corpore Sano
Our motto (above) says it all.
At chillingcompetition, we like people who combine business and pleasure.
And we are particularly fascinated by those who mix serious lawyering and sport.
Alfonso is perhaps the physical incarnation of this.
Sport, especially basketball, was always big in his life as a kid.
When he moved to Brussels as a young lawyer, he started socializing running at the Aspria.
And in more recent years he has seemed willing to explore new, extreme sporting territories like gastronomic orgies of jamon serrano and trappist beer drinking.
With this background, we are proud to post above a picture of the fastest competition lawyer in town.
Alfie pays a drink to the reader who finds the man behind the helmet.
Competition Law Everywhere
In Montenegro, where I spent my hols, the price of a liter of gasoline is €1.38.
This price is the same everywhere in the country. And it is the same in all petrol station networks.
Me find this reasonably cheap as compared to most Western EU countries. But me also find this price curiously uniform.
Two obvious explanations spring to my obsessed antitrust mind: 1. cartel; 2. government regulation.
There’s worse though: a well-known car rental agency has confronted me with a real life example of abusive unfair pricing.
If you don’t return the car with the initial quantity of oil, the car rental company charges three times the price per missing liter of gasoline (!) – unless you pay black cash, in which case they charge you €1.5/liter to refuel the car…
I fully grasp that car rental firms want to recover vehicles with a sufficient quantity of gasoline, so as to avoid the costs of refueling them constantly.
But the alternative would be to charge customers for the missing quantity of gasoline at a rate equal to 1.38€ liter + an average increment covering the cost incurred for refuelling the vehicles.
I have not done the math short of data on this (+ I was on hols), but the 3x price formula looks really excessive.
The bottom line: yet another tactic that seeks to exploit the lazy mind bias of locked-in consumers.
Déjà vu? Microsoft announces Skype’s integration in Windows
On 15 August Microsoft announced on a blog post that Skype will come installed by default in Windows 8.1, and that it will be prominently displayed in its “Start” Menu (see Skype-right from (the) “Start”)
The news appears to have surprised many, who have publicly wondered whether Microsoft is actively looking for antitrust trouble (see notably here, here, or here).
And, of course, given my involvement in Skype-related competition matters, when I returned from my summer holidays I had a good number of emails from students, journalists, lawyers, friends and even family who were sending me the news and asking for an opinion. Since it would not be practical to reply to all those emails separately, I have decided to do it here.
[A disclaimer first: as frequent readers of this blog know I represent the two companies who chose to challenge the Commission’s decision authorizing the Microsoft/Skype deal. This means that I certainly am not an impartial observer, but it does not mean that the views set out here are to be attributed to my clients or my firm; they are exclusively mine. These views also refer to a conduct which is post-decision and therefore not the subject of the pending case].
My first comment is: Did anyone really not see this coming?
During the past few months Microsoft has pervasively integrated Skype with most of its products. Skype is now closely integrated with, for instance, Office, Office 365, Outlook, Outlook.com (formerly Hotmail), Windows Phone 8, Xbox, Lync (as announced only minutes after our Court hearing ended), and it was only a matter of time that it would come pre-installed in Windows. In the meanwhile, Skype’s only meaningful competitor in the consumer world (WindowsLiveMessenger) has disappeared and its users have been migrated to Skype. As a result, Skype’s user base has skyrocketed since the merger (going from approx. 150 to over 300 million unique monthly users), and rapidly growing.
[By the way, all this obviously voluntarily enhances the already powerful network effects at play in the only communication markets where interconnection is not mandatory, with obvious consequences]
Microsoft’s decision to bundle Skype pervasively with other Microsoft products, including – as just announced – Windows, may actually have come as a surprise to the European Commission. In its Microsoft/Skype decision, the Commission concluded that Microsoft would not have the incentive to tie Skype to other Microsoft “leading/dominant” products (e.g., para 155). No kidding.
Now let’s cut to the chase, can the integration of an application with a dominant operating system run afoul of the competition rules?
The European Commission itself has held various seemingly contradictory views over time. Microsoft, too, appears to have opposite views on this question. Let me explain this:
In the light of the spirit and the letter of the Microsoft’s 2004 infringement decision, the 2007 Microsoft Judgment, the 2009 Microsoft commitment decision, Skype’s integration with Windows would likely raise some antitrust flags (notably concerning the market for video calls, given that currently over 3 out of 4 video calls are made using PCs). As you know, in all of those precedents, the Commission and the General Court observed that pre-installation resulted in an unparalleled distributional advantage that could not be offset by the downloading of competing applications.
The Microsoft/Skype 2011 decision, however, arrived at exactly the opposite conclusion. The comments voiced out in the past few days in the media seem to have overlooked the fact that the Microsoft/Skype Decision – despite denying Microsoft’s incentives to tie Skype to its products – did actually address the possibility that Skype could be tied to Windows, and that it ruled out any competition concerns. The Decision acknowledged that pre-merger Skype was already present on approximately 60% of Windows PCs pursuant to agreements with OEMs, but alleged that there was data -not cited- showing that in practice pre-installation resulted only in a small share of Skype users (para 162). In other words, the Commission considered that pre-installation does not offer that much of a competitive advantage because users could easily and freely download Skype and other competing applications.
Query: does anyone see any inconsistencies between the Commission’s approaches to downloading? The Commission is certainly entitled to change approaches, but since the reasons for this change were not set out in the Decision, it’s difficult to identify with clarity what the Commission’s current approach to pre-installation vs. downloading is.
If you want to play more “find the differences”, try comparing the Commission’s prospective analyses and approaches to technical tying/bundling (and, for that matter, to interoperability degradations too) in Intel/McAfee (2011) and Microsoft/Skype (2011).
And whereas the Commission’s shifting viewpoints are remarkable, what is more striking is that Microsoft is, as of today, advocating two opposite legal standards, one for itself and another for Google:
As you may remember, back in April the FairSearch coalition (led in this case by Microsoft and Nokia) lodged a complaint against Google arguing that Google is abusing Android’s alleged dominance in the market for mobile operating systems by bundling certain “core Apps” with its operating system.
[The way I see it, in the case of Android the dominance and the bundlling are much more doubttful, but that is another story, and one interesting enough -I’ve just realized- to deserve some specific comments in the coming days].
So, in one case Microsoft is claiming that the pre-installation of Google apps on Android phones constitutes an abuse of a dominant position in the market for mobile OSs (no matter if users are free to download any competing application; btw, Skype for Android has no less than 100 million users!), but, at the same time, having Skype pre-installed in the dominant PC OS poses no problem (precisely because users are free to download other applications).
Anyone else sees any issue conflict?
A competition authority closed for business
A friend kindly pointed us to this “We’re-at-the-beach-so-dont-bother-looking-for-us-sign” with a comment: “Transparency and vacation should be the unrenounceable principles of any public authority“.
Chillin'Competition 