The battle to beat Google’s Android mobile phone OS is quickly turning into a legal bonanza. Apple is suing HTC, Samsung and Motorola, all makers of wireless phones with the Android platform. Oracle is seeking up to $6.1 billion in a patent lawsuit against Google, alleging Android infringes Oracle’s Java patents. And Microsoft is suing Motorola over its Android line.
That’s all perfectly fine from an antitrust and competition standpoint — leaving aside the harder policy question of whether using patent infringement litigation to block competition should be permissible. Enforcing property rights is a legitimate and rational business activity that, absent “sham” lawsuits, is not second-guessed by antitrust enforcement agencies or courts. There can be exclusionary consequences, but they are a result of the patent laws in the first instance, not of themselves anything anticompetitive by the patent holder.
A much more troubling aspect of the increasing IP (or “IPR” as they say across the pond) battles surrounding Android is the recent sale of Nortel’s 6,000 or so wireless patents at a bankruptcy auction in Canada to a collection of bidders including Apple, Microsoft, RIM, EMC, Ericsson and Sony. How Apple Led The High-Stakes Patent Poker Win Against Google, Sealing Ballmer’s Promise | TechCrunch. The winning consortium bid more than $4.5 billion — some five times Google’s opening bid and, according to some pundits, far more than the portfolio was worth — to gain control of the patents.
“Why is the portfolio worth five times more to this group collectively than it is to Google?” said Robert Skitol, an antitrust lawyer at the Drinker Biddle firm. “Why are three horizontal competitors being allowed to collaborate and cooperate and join hands together in this, rather than competing against each other?”
These are good questions. Patent “pools,” which are collections of horizontal competitors sharing patent licenses among themselves, are today generally considered procompetitive under the antitrust laws where they (a) are limited to technologically essential or “blocking” patents, and (b) do not contain ancillary restraints, such as resale price-setting or restrictions on participant use of alternative technologies. (MPEG, WiFi, LTE and other communications technologies are prime examples of patent pools.) The theory is that, with price effects eliminated, the cross-licensing of patents that might otherwise be used to block entry into a market reduces barriers to entry and increases efficiency.
Yet the consortium which won the Nortel wireless portfolio, revealing dubbed “Rockstar Bidco,” includes nearly everyone in the mobile phone and wireless OS businesses except Google. If these players agreed among themselves not to license their own patents to Google, that would be a per se illegal group boycott (also known as a concerted horizontal refusal to deal). Competitors cannot allocate markets or conspire to keep a rival out of the marketplace. It is unclear whether Google was invited to join Rockstar Bidco, but unless Larry, Sergey and Eric turned down such an offer, it seems a fair case can be made that the consortium bid was in effect an implicit horizontal agreement not to include Google. Post-auction, the reality of licenses will clearly tell us whether the joint ownership structure was a pretext to cover a refusal to deal. No one knows what the consortium intends to do with the Nortel patent portfolio; they won’t say. Microsoft, RIM And Partners Mum On Plans For Nortel Patents | Forbes.
This author happens not to be a fan of Android; I’m a very happy iPhone user since day one of the Apple wireless revolution. This does not mean, though, that I can agree with a business strategy in which all of the other players in the mobile phone industry gang up on Google. (It is unclear were Nokia fits into all of this, but given the steadily decreasing share for its Symbian OS, I suspect the inclusion or not of Nokia will not be dispositive.)
The antitrust issue this presents is a thorny one, which frequently comes up in connection with trade associations and technical standards. When competitors collaborate, is under-inclusiveness or over-inclusiveness worse? Which is the bigger threat to competition? That is, if a trade group opens a collective buying consortium, for instance, is it better from an antitrust perspective to require that it be open to all — so that some rivals are not deprived of the scale economies — or that the consortium includes less than all firms in the market — so that competition in purchasing will drive down input prices?
Another concern is that, by excluding Google, the Rockstar consortium allows the other competitors to utilize the patents without paying license fees (since they now own them), leaving Google alone to need licenses for its Android OS. Does Nortel Patent Sale Make Google An Antitrust Victim? | TechFlash. That is a variant of “raising rivals’ costs” (here one rival only), which has over the past three decades become a recognized basis for assessing the anticompetitive nature of unilateral, single-firm conduct. When a group includes horizontal competitors who collectively control a huge share of the market, raising rivals’ costs supplies the anticompetitive “purpose or effect” needed to make out a rule of reason antitrust claim, even if the group boycott concern is misplaced or ameliorated. Here the intent to slow down Android is clear; whether that is anticompetitive, exclusionary or not is more ambiguous. Apple, Microsoft Patent Consortium Trying to Kill Android | eWeek.com.
There are precious few judicial decisions in this area and the IP licensing guidelines from DOJ/FTC do not really speak to the question. For that reason alone, the Rockstar Bidco venture, in my view, merits a very close look by the U.S. competition agencies. Allowing Google’s mobile phone competitors to do indirectly, with joint patent ownership, what they could not do indirectly, by agreeing not to license to Google, would be an incongruous result. On the other hand, a remedy may be worse than the harm. In standards, for example, it is often the case that antitrust risks are mitigated by requiring the holder of an essential patent to agree to so-called FRAND licensing (fair, reasonable and non-discriminatory terms and conditions). That’s an appropriate remedy where under-inclusiveness is the problem, so long as there’s a market measure for a “fair” license (royalty) price. Where the licensor, as in this instance, is everyone except the licensee, I for one fear there would be no objective way to assess whether license rates were reasonable.
The lack of an effective remedy for a competition problem does not, of course, require that the transaction involved be blocked. At the same time, where a problem cannot be fixed, that is a good enforcement policy reason not to allow the structural market conditions giving rise to the issue in the first place. Put another way — a slight modification of an old aphorism — if there’s no remedy, maybe there should be no right. Whether the viability of the Rockstar consortium is decided by outgoing Assistant Attorney General Christine Varney or her September successor, the forthcoming answer should be interesting.
[Cross-posted at Lex Digerati]