Sunday, April 15, 2012

Motorola calls $4 billion royalty figure 'misleading' but doesn't dispute the number per se

Late on Friday, Motorola Mobility filed its opposition brief to a Microsoft motion for partial summary judgment that asks a federal court in Seattle to find that Motorola breached contractual obligations by making excessive royalty demands that are nowhere near a FRAND royalty level. What lends the process concerning that motion additional relevance is the fact that last Wednesday's temporary restraining order, which prevents Motorola from the near-term enforcement of a patent injunction it might win in Germany on May 2, will be in effect (in its current form) until the Seattle-based court decides on this breach-of-contract issue after a May 7 hearing.

Here's Motorola's brief, which I'll discuss further below:

12-04-13 Motorola Response to Microsoft Motion for Summary Judgment

For every item of a summary judgment motion, three outcomes are possible: an item can be

  • granted,

  • denied for lack of merit, or

  • denied because it's not (or at least not yet) ripe for summary judgment since there is still room for disagreements on relevant facts, which may ultimately require the involvement of a jury.

Motorola's 30-page brief (naturally) disputes the merit of Microsoft's motion, but the emphasis is very clearly on claiming that the question of reasonable royalty demands must be put before a jury. Not only is this a common strategy (since it's easier to defeat a summary judgment motion on that basis) but it's particularly logical in this case, given that a previous Microsoft motion for summary judgment on this issue was denied in late February for two deficiencies that the court allowed Microsoft to fix in a new motion: on the one hand, the court found Microsoft's legal theory lacking as far as the obligation of a holder of FRAND-pledged patents to make FRAND offers is concerned (Motorola argues, as it does in Friday's brief, that only the outcome of a negotiation must be FRAND, but does not argue that "an opening offer can be 'blatantly unreasonable'", admitting immediately that "an opening offer must be made in good faith"); on the other hand, Judge James L. Robart determined at the time that "[o]n the record before it, the court cannot say that there is no genuine dispute as to any material fact with respect to whether Motorola's initial offers were on [F]RAND terms".

Microsoft's motion reflects that its lawyers heeded both aspects of the court's guidance. There is now a more elaborate legal argument in place as to why Motorola's contractual obligations (the fact that there are contractual obligations has already been decided in Microsoft's favor) were breached by "blatantly unreasonable" royalty demand s(that came with a 20-day ultimatum). And as far as the factual dispute over the appropriateness of certain demands (2.25% of the price of the end product is what Motorola demanded) is concerned, Microsoft's second-attempt motion essentially argues that no juror in his or her right mind could come to any other conclusion than the one that Motorola's demands were totally out of line -- a scenario in which a judge can decide without having to ask a jury.

If Motorola persuaded the judge of the alleged need for a jury verdict, the temporary restraining order would expire on or shortly after May 7 (the judge will try to decide on the day of the hearing, but delays are always possible), but that doesn't mean that Motorola would be free to enforce a German injunction (if one is ordered in the meantime) immediately: a follow-on order could effectively extend the existing one. The conduct on Motorola's part that was key to Judge Robart's temporary restraining order would still be the same: an attempted end-run around the FRAND enforcement case in Seattle through actions in other jurisdictions with different timelines and standards. But at least Motorola could argue that it should be allowed to enforce, while it's hard to see how Motorola would get (or even dare) to enforce a German injunction after losing yet another summary judgment decision in Seattle (the aforementioned late-February decision was a partial victory for Microsoft, with two of its requests having been granted).

In connection with the H.264 video codec standard, Microsoft compared Motorola's royalty demand (applying the percentage Motorola wanted against the royalty base Motorola stated) to what it pays, or would without an annual royalty cap pay, to the MPEG LA patent pool (click to enlarge):

Motorola's opposition brief counters that graphic with the following assertions:

"As discussed below, Microsoft's $4 billion estimation is misleading. And as demonstrated in Motorola's opposition to Microsoft's original summary judgment motion, comparing Motorola's proposed royalty terms to that of the MPEG LA or other patent pools is simply unsound. Dkt. No. 90 at 19-21. Pools have rates that are not negotiated and are purposefully low, having no relevance to a negotiated one-on-one license."

The first sentence addresses the right side of the chart (Motorola's demand), and the second sentence the comparability of the left side (patent pool rates) to the right side. Interestingly, Motorola doesn't say that the $4 billion figure is inaccurate or exaggerated: it just says that it's "misleading", and denies that the different rates can be compared.

With respect to comparability, let's just assume, for the sake of the argument, that some court or regulatory agency decided to give Motorola the benefit of the doubt and base any decision on the hypothesis of patent pools being cheap -- even though the number, nature and stature of the companies involved in the MPEG LA pool makes it hard to believe that this was a conspiracy to drive down prices. It's true that the consortium members are interested in adoption of the standard, but every other business also has to consider the impact of pricing on volume and will nevertheless just look for the best bottom-line result. A salesman in the computer software (and previously consumer electronics) business with whom I worked closely in the first half of the 1990s frequently said in internal discussions on pricing that he learned this rule of thumb: "Half the price, ten times the volume." But even if one agreed with Motorola's proposition that pools are cheap just for the sake of being cheap, the discrepancy between the rates it asks for and the pool rates is so unbelievably huge that this gap can't be bridged just with this argument. At the most (and even that is doubtful), that argument could explain away a small part of that discrepancy, with its remainder still being the opposite of FRAND.

Even though the word "misleading" is more of an admission than a rebuttal of the $4 billion figure (which Microsoft considers a conservative estimate because it's based on the assumption of an average PC price ($500) that appears rather low), I really would have liked to understand the reason for calling the number "misleading". Unfortunately, Motorola's brief is heavily redacted. I understand that Motorola can't disclose, for example, the exact terms of its license agreements with other companies, such as Nokia. But with entire long paragraphs or even sequences of paragraphs (such as two pages in a row, from the middle page 17 to the middle of page 19) having been redacted, I couldn't identify any basis for the "misleading" claim. On the contrary, the non-redacted parts of Motorola's brief support Microsoft's estimate. For example, Motorola's brief defends the idea of charging Microsoft, for each copy of Windows it ships, based on the price of the end product:

"Finally, Microsoft criticizes Motorola for proposing that the royalty be based on the selling price of, among others products, computers (i.e., the end product) rather than the selling price of the Windows operating system, even though Microsoft does not sell computers. [...] But this offer was consistent with Motorola's practice of using the price of the end product as a royalty base and was made to show Microsoft how Motorola valued its patents (i.e., on the end product). Moreover, this was a reasonable opening proposal, in light of the Supreme Court's decision in Quanta Computer, Inc. v. LG Elecs., Inc., [...] (2008). In that case, the Supreme Court held that '[t]he authorized sale of an article that substantially embodies a patent exhausts the patent holder's rights and prevents the patent holder from invoking patent law to control post-sale use of the article.' [...] Accordingly, in order to ensure that its downstream rights were protected, Motorola's opening offer made clear that it wanted to hold Microsoft responsible for collecting Motorola's normal percentage of the end-product price."

Patent exhaustion is a concept I've discussed on this blog on several occasions (if you look for examples, just click on that term in the list of tags at the end of the post). Here, Motorola's argument is that if it sells a license to Microsoft to incorporate H.264 in Windows, it sort of grants a license to an entire PC that ships with Windows -- therefore, Motorola says, it needs to protect itself and ask for a royalty based on the full price right away. That argument is an insult to human intelligence for anyone who knows what patent exhaustion does and does not mean, but even if it meant what Motorola suggests by stressing a need to protect its "downstream rights", it would not justify Motorola's demands.

First, patent exhaustion doesn't mean that each component, or combination of components, of a Windows PC is licensed to practice Motorola's H.264 patents in whatever way only because Windows itself is. Motorola doesn't say so, but it implies this by claiming it needs to get paid by Microsoft for the full price of the PC. From a patent exhaustion point of view, a PC as a whole would be licensed only to the extent that its use of H.264 comes down to using Windows itself. If, for example, the same patented inventions are practiced by an alternative operating system (say, Ubuntu, which theoretically could be installed later by the user or be shipped as a dual boot option) or by a graphics chip built into the same device (an issue that the ITC also had to evaluate in its investigation of S3 Graphics' first, and ultimately-dismissed, complaint against Apple), then exhaustion doesn't apply to those use cases.

Second, whether one interprets patent exhaustion broadly or narrowly, Motorola's demand comes down to double-dipping (or multiple-dipping). It would charge Microsoft for the full price of the PC, Canonical (the Ubuntu maker) for that same full price, and a graphics chip maker like NVIDIA. What it should do, however, is charge each of them for what their respective components do, and base its royalty on the contribution its patented inventions make to each component.

And in any event, it's clear that not every component of a PC even has the basic capability of practicing those patented inventions. If someone buys a PC with a gold-plated case, just like there are some $100,000+ jewelry smartphones out there, the golden case won't ever play a video. Nor would an enterprise-class database for $10,000 or more.

In light of that, I don't expect the patent exhaustion argument to benefit Motorola's case.

I'd like to highlight a few more points that Motorola makes and that appear to make sense at first sight but don't withstand further analysis.

In a footnote, Motorola says the following:

"The value of patent transfers and grant back licenses cannot be underestimated. Indeed, Microsoft was part of a consortium that last year spent $4.5 billion to buy Nortel's patent portfolio and days ago spent $1.1 billion to purchase 800 non-essential patents from AOL."

The part that the footnote relates to is, unfortunately, redacted, but the footnote can still be analyzed on its own. What Motorola does here is to toss out billion-dollar amounts that were paid for the acquisition of certain portfolios in order to claim that "patent transfer and grant back licenses" must be very valuable. However, Motorola admits that AOL's patents were "non-essential", making them a different subject. Concerning Nortel, this was the sale of a bankruptcy estate, and before the final auction, Microsoft and Motorola were among the various industry leaders who expressed concern (in filings with a bankruptcy court) over the impact of bankruptcy on the related licensing commitments. But at any rate, comparing licenses to acquisitions is about the same as comparing the rent paid for one house to the price of another house.

The next item relates to the following graphic from Microsoft's motion, which relates to the WiFi (or WLAN) standard named IEEE 802.11:

Motorola says that suggesting "that the royalty base should, at most, be the cost of the WiFi chip inside the Xbox, rather than the Xbox itself" is "akin to valuing a music album, television show or movie based on the value of the blank CD or DVD recording medium". The mistake here is that the value of a blank medium is a function of a relatively low cost of goods and a low cost of recording, without any intellectual property rights going beyond the material involved, while computer chips usually do come with significant intellectual property rights. This is also an issue between Motorola and Apple in connection with wireless standards. But a Qualcomm chip comes with a huge amount of licensed intellectual property (including some of Motorola's own patents, in fact), and its value is far above that of what would otherwise be tantamount to scrap metal.

Then Motorola says that Microsoft "has valued the WiFi functionality at 40% of the price of an Xbox system", or "far more than the value of the WiFi chip inside the adapter". Motorola points to this webpage promoting a "Xbox 360® Wireless N Networking Adapter". Its currently-advertised price is $59.99, not the $79.99 that Motorola says was last year's price, but that's merely a gradual difference. Motorola doesn't say (at least not in the part that it shows to the public) that the gaming console business is known to be a razor-and-blades business: companies give the razor away at subsidized prices in order to make money on the blades (software and add-ons). Here, Motorola compares the price of a blade (an add-on product to bring WiFi to older devices) to the subsidized price of a razor. It would be a different thing if Motorola claimed a higher royalty on a blade that is WiFi-focused than the primary issue, which is that Motorola wants to receive royalties on the end price of an entire Xbox or Windows PC.

Also, it certainly is more costly to manufacture and sell an add-on hardware component than to integrate a small chip into the original product.

Finally, Motorola says that the Entire Market Value rule, a way to ensure an appropriate royalty base in connection with patent infringement damages, is not applicable here because it is "directed to determining a reasonable royalty in the context of patent infringement damages -- not negotiating a RAND rate during real-world negotiations". However, there are good reasons to assume that infringement damages relating to non-standard-essential patents are in no event lower than a RAND rate, but in many cases even higher. Anyway, the Entire Market Value rule solves a problem that both the determination of infringement damages and the determination of FRAND royalties face, and I can't see a compelling reason for which it shouldn't be applicable to FRAND royalties. At least I don't see a better proposal on the table.

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: