Saturday, September 8, 2012

Microsoft and Google's Motorola Mobility attack each other's FRAND royalty expert reports

Two months prior to a FRAND contract trial in Seattle, Microsoft and Google subsidiary Motorola Mobility are having a Daubert fight, i.e., motions to exclude each other's expert testimony. The court's plan is to set the terms for a FRAND license that would cover Microsoft's worldwide use of Motorola Mobility's patents declared essential to the H.264 (video codec) and IEEE 802.11 (WiFi, or WLAN) standards. Microsoft's objective is to get a license without having to overpay, while Motorola Mobility primarily wants to retain the option of pursuing and enforcing injunctive relief, and at a minimum wants to receive a very high royalty for its standard-essential patents in order to offset as much as possible of the ultimately inevitable cost of licensing Microsoft's numerous non-standard-essential patents that read on Android.

This FRAND dispute in the Western District of Washington has also resulted in a preliminary injunction barring Motorola from the enforcement of a couple of German H.264 patent injunctions. Motorola Mobility appealed this order, and the United States Court of Appeals for the Ninth Circuit will hold a hearing on Tuesday, September 11.

The Daubert motions are all about what the FRAND royalty rate should be, not about injunctive relief per se. Indirectly, however, there is a connection with injunctive relief. Motorola claims that the court does not have the authority to create a contract "ab initio" (from scratch), but even Motorola has previously agreed that if it original 2.25% royalty rate demand was "blatantly unreasonable", it has breached its FRAND-related contractual obligations, which in turn would bar Motorola from seeking and enforcing injunctions against Microsoft. In other words, if the court's plan, which Microsoft supports, works out, it's going to be "game over" for Motorola's quest for injunctions against Microsoft, but even if Motorola was right and this litigation could never result in a license agreement after the November trial, it would also be "game over" for Motorola's injunctions if the court determined a FRAND royalty rate compared to which Motorola's 2.25% demand was "blatantly unreasonable".

The expert reports themselves are sealed, but the parties' filed public redacted versions of their Daubert motions on August 27 and their responses on Wednesday (September 5), and those pleadings provide useful information as to where the parties stand. Motorola still stands by its 2.25% demand, which is ambitious and aggressive but little surprise in light of the tactical needs I outlined in the previous paragraph. Microsoft consistently argues that pool rates for large numbers of patents (including those held by a variety of major companies) declared essential to the two standards in question, such as the MPEG LA AVC/H.264 and the Via Licensing IEEE 802.11 pool rates, are within the FRAND spectrum, and the outcome of this rate-setting should not be far above those pool rates.

The parties are so far apart that there can't be an outcome that both parties accept at the same time. At least one of them will appeal, and I wouldn't even be surprised if both appealed in the end, with one arguing the court-determined rate is too low and the other one saying it's too high.

But before we get to a decision, there must be a trial at which the parties present evidence (such as expert reports) that supports their positions, and prior to that, the court will have to throw out FRAND rate theories that are contrary to law, unreliable, or irrelevant. FRAND is a range, not a point, and it's obvious that Motorola will ask for more than Microsoft will offer. The court will be inclined to resolve as much as possible at trial, but theories that are fatally deficient should be thrown out now, at the Daubert stage.

I believe the Daubert process is much more of a threat to Google's (Motorola's) 2.25% demand than to Microsoft's pool rate counterproposal, and one of the reasons I think so is that the tactical necessities I outlined above leave Motorola with no real choice: no matter how much of a long shot this may be, Motorola needs to convince the court that 2.25% is the way and the light. And if it can't convince this court, it at least wants to be able to appeal without having softened that stance. Motorola is strategically stuck. It has dug itself a hole, not only because it asked for 2.25% in the beginning but especially because it kept digging and digging as the process evolved, such as by admitting that a "blatantly unreasonable" initial demand would constitute a breach of contract.

By contrast, if Microsoft had concluded that a fair rate is not just the pool rate but, say, ten times the pool rate, Microsoft could offer that rate or something close to it. For Microsoft it would be pointless to take an extreme position because even compared to 100 times the pool rate, Motorola's demand would still have to be considered "blatantly unreasonable". Microsoft wouldn't want to pay 100 times more than it considers reasonable, but since there's a whole range of outcomes that would work for its purposes, it would be foolish to overplay its hand and risk that, in a worst-case scenario, its position is thrown out ahead of trial. But Motorola may even have to risk a Daubert order against its position: in that case, it can at least appeal, but if it voluntarily lowers its demand to the point that its initial 2.25% demand is undoubtedly unreasonable, it loses without a fight. Quite a dilemma.

In the following I'll look at some of the issues here that the parties raise in their Daubert motions and responsive pleadings.

Motorola's criticism of Microsoft's FRAND rate theory based on multilateral negotiations

Motorola's Daubert motions calls Microsoft's various expert reports "wishful thinking". The largest part of Motorola's argument is that FRAND terms result from bilateral negotiations (one patent holder, one implementer), while Microsoft's experts base their theories on "multilateral" negotiations that are held when a patent pool is formed. The second part of Motorola's motion levels further criticism and claims that Microsoft's expert opinions are not based on a methodology that provides reliable results, but it's also centered around the bilateral/multilateral argument.

In its opposition brief, Microsoft says that considering the terms of ex ante, multilateral licenses (pool rates) is a way to arrive at a FRAND rate, and all that matters is whether result is FRAND. Microsoft counters Motorola's claim that its experts' consideration of bilateral license agreements (apparently 50 of them) is the only reliable approach by noting that "the MPEG-LA pool considered by Microsoft's experts reflects the agreement of 29 licensors and more than 1,100 licensees for RAND royalties for 2,339 standard-essential patents, a combination which dwarfs the 50 agreements Motorola claims to have considered". And Motorola's argument that an ex ante (before standardization) perspective on FRAND rates is not acceptable runs counter to Judge Posner's position on FRAND royalties in Apple v. Motorola. Here's a key passage from Judge Posner's decision:

"There is another decisive objection to Motorola's damages claim. The proper method of computing a FRAND royalty starts with what the cost to the licensee would have been of obtaining, just before the patented invention was declared essential to compliance with the industry standard, a license for the function performed by the patent. That cost would be a measure of the value of the patent qua patent. But once a patent becomes essential to a standard, the patentee's bargaining power surges because a prospective licensee has no alternative to licensing the patent; he is at the patentee's mercy. The purpose of the FRAND requirements, the validity of which Motorola doesn't question, is to confine the patentee's royalty demand to the value conferred by the patent itself as distinct from the additional value--the hold-up value--conferred by the patent's being designated as standard-essential."

Judge Posner said great things about injunctive relief in connection with standard-essential patents, but the above position that FRAND rates must reflect "the value of the patent qua patent", stripped of any post-standardization hold-up value, is no less important. It should be the guiding principle of all FRAND rate-setting discussions in the world.

Theoretically, Motorola could have chosen to accept the Posner FRAND rate doctrine and to try to argue that even on that basis, 2.25% is a FRAND rate. But in that case, it wouldn't be able to criticize Microsoft's expert reports for looking at the ex ante value of such patents. And it would have to support its own 2.25% rate based on the value "qua patent" of each of its patents. Motorola apparently concluded that it can't win a battle over the value "qua patent".

Motorola's Daubert motion faces a high hurdle. It will succeed only if the court concludes that Microsoft's experts came up with numbers that are clearly outside of the FRAND ballpark. If Motorola "only" convinces the court that Miorosoft's theories are lowball offers within the FRAND range, that's not enough to get them tossed ahead of trial. And in order to deem the results of ex ante negotiations to be unreliable or even contrary to law, a court would have to determine that, contrary to Posner's "value qua patent" take, SEP holders are entitled not only to compensation based on the pre-standardization value of their patents but to payments that additionally -- and unjustly -- reward them for their participation in standard-setting. I doubt very strongly that Motorola will be able to persuade any U.S. district court, nor any U.S. appeals court, that a patent should become more expensive to license once it's included in a standard.

Microsoft argues that Motorola's expert report captures hold-up value

Motorola's opposition to Microsoft's theories based on the results of ex ante negotiations is mirrored by Microsoft's criticism that Motorola's own expert, Charles R. Donohoe, captures hold-up value (in other words, the value of being able to threaten with injunctions affecting a company's ability to implement an entire standard) in his theory supporting Motorola's 2.25% demand.

In its opposition brief, Motorola vehemently denies this. But Microsoft's Daubert motion provides some quotes from Mr. Donohoe's report that make it plausible that Motorola really wants to be compensated based on the economic value and commercial relevance of the H.264 and IEEE 802.11 standards as opposed to the pre-standardization value of its contributed patents:

"Wi-Fi functionality has become pervasive" and "devices can[]not be competitive [without it]"

"H.264 functionality is an important feature of both Windows and Xbox"

"[a] wireless connection has certain advantages over wired connections"

"H.264 [] functionality has certain advantages over other video compression formatting standards"

"[IEEE] 802.11 technology had clear advantages and utility over wired networks and cellular technology", and "this factor supports Motorola's offer of 2.25% for its [IEEE] 802.11 [patents]"

If Mr. Donohoe's report focused on why Motorola's declared-essential patents cover technological breakthroughs that justify a 2.25% rate, I doubt that passages like the ones quoted above would be found in it. Again, I haven't seen those expert reports, so I can form my opinion only on the basis of the Daubert pleadings, but I trust both parties when they provide literal quotes from thsoe reports. And the quotes listed above appear to indicate that the Donohoe report is the antithesis of Judge Posner's value-of-patent-qua-patent take.

The 15 Georgia-Pacific factors and the Entire Market Value Rule

A significant part of the parties' Daubert dispute is all about the applicability of certain patent damages theories to the FRAND rate-setting exercise. Microsoft describes the 15 Georgia-Pacific factors as "open-ended" and "unweighted", while Motorola apparently believes that a theory based on the outcome of a negotiation that takes place under this framework (and, Motorola being Motorola, after standardization) is most likely to work for its purposes. In other words, Motorola thinks that its best shot at defending its 2.25% demand is to address the Georgia-Pacific factors.

At first sight, Georgia-Pacific is a reasonable framework. If you read the 15 factors I just linked to, there's something in it for everyone. But that framework is actually quite treacherous. If a court doesn't insist that at least some weighting of the different factors occurs, it's easy to downplay some highly important factors and to overstate the relevance of less important or even inapplicable ones.

A particularly important example here is factor #13. That one is based on she same line of thought as the Entire Market Value Rule (EMVR), which Motorola says should not apply here. Even the EMVR has been criticized as making it too easy for patent holders to overstate what they're entitled to (for example, Professor Brian J. Love wrote a paper entitled "Patentee Overcompensation and the Entire Market Value Rule"). But at least the EMVR provides courts and parties with a tool for potentially separating those parts of a complex piece of technology that a particular patent has nothing to do with from the smallest saleable unit that the patent relates to. And in today's high-tech industries, patent damages (or court-determined royalties) would be unsustainably high if the EMVR was not applied properly.

Under Georgia-Pacific, apportionment is factor number 13 out of 15. That's almost an afterthought. And it's because this is a four-decades-old framework and was developed in connection with a plywood-related product that is absolutely not comparable to today's smartphones, operating systems or gaming consoles. For today's high-tech products, identifying the relevant component of a product is the very first thing to do. But Motorola doesn't want this, and it can't: if it admitted that it used the wrong royalty base initially by demanding 2.25% based on the sale of the relevant end product, it would have to admit that its original demand was "blatantly unreasonable".

Google's (Motorola's) preference for a royalty theory based on a bilateral negotiation under the Georgia-Pacific framework would be much, much less of a problem if Motorola accepted Posner's value-of-patent-qua-patent guideline and, as a result, a focus on the pre-standard-setting value of its patents, a value that can only be assessed properly after applying the Entire Market Value Rule. On that basis, the difference in methodologies between Microsoft and Google (Motorola) would come down to the difference between bilateral and multilateral negotiations, and probably some argument over how to weigh the different Georgia-Pacific factors. They would always arrive at different numbers, and there would be plenty of room for disagreement on the value of Motorola's patented inventions, but they could have a constructive process, and possibly even reach an agreement ahead of trial. But Google (Motorola) would have to give up its dream of winning and enforcing injunctions, or at least receiving gigantic amounts of royalties, based on its SEPs.

If you'd like to be updated on the smartphone patent disputes and other intellectual property matters I cover, please subscribe to my RSS feed (in the right-hand column) and/or follow me on Twitter @FOSSpatents and Google+.

Share with other professionals via LinkedIn: