Thursday, September 6, 2018

Apple, Foxconn et al. want to end Qualcomm's double-dipping practice (chips and patents) once and for all

The previous post was about a magistrate judge's annoyance at Qualcomm's gamesmanship. But what's sure to cause Qualcomm and its lawyers a lot more headache is the pressure from two important summary judgment motions: the FTC's motion in the Northern District of California seeking to clarify Qualcomm's self-imposed obligation to license rival chipset makers and--which is what this present post is about--a motion by Apple and four contract manufacturers (Foxconn, Compal, Wistron, and Pegatron) in the Southern District of California, leveraging the Supreme Court's 2017 Lexmark ruling against Qualcomm's double-dipping practice (this post continues below the document):

18-08-31 Apple Et Al. Motion for Partial Summary Judgment by Florian Mueller on Scribd

For a recap of something discussed in greater detail last year, Qualcomm's business model is basically like this:

  • As Judge Koh recalled in a recent order, "an analysis conducted by Qualcomm in 2015 showed that revenues from Qualcomm's licensing program were 'equivalent in size to the sum of ~12 companies with a form of technology licensing,' including leading cellular SEP licensors such as Ericsson, Nokia, and Interdigital."

  • Customers have to accept an out-of-this-world royalty rate in order to be able to buy Qualcomm's chips.

  • Those who don't want to buy Qualcomm's chips must still pay patent license fees and will sooner or later be "persuaded" to enter into exclusivity arrangements under which their total cost (the sum of chip supplies and patent royalties) is reduced (but apparently still very high).

The above sounds like a "great" (for Qualcomm, not for the economy at large or for society) deal. However, Apple and its four contract manufacturers say that "Lexmark forbids this double-dipping."

The doctrine of patent exhaustion, which the Supreme Court clarified, upheld and reinforced in Lexmark, is fundamentally irreconcilable with the notion of a company selling chipsets and collecting patent license fees from the customers buying its chips. But Qualcomm appears to have believed for a long time that it would get away with it. One, because no one would dare challenge it. Two, by virtue of three structural elements that the summary judgment motion addresses:

  • "Qualcomm divides the chipset sales and software license across two agreements"--one for the purchase of the chipsets and another one, named Master Software Agreement, for the software license.

  • Foreign sales (such as to Asia-based contract manufactures) don't help as Lexmark clear stated that "[a]n authorized sale outside the United States, just as one within the United States, exhausts all rights under the Patent Act."

  • The motion labels as "corporate shell games" a structure under which different Qualcomm entities sell certain products and technically own certain patents. More than six years ago I already reported and commented on a related restructuring.

The motion invites Judge Gonzalo Curiel to take a Lexmark attitude and focus on a simple commercial question: are we talking about authorized chipset sales or not? If authorized, there's exhaustion; if Qualcomm wants to argue that exhaustion doesn't kick in, it has to show that something was unauthorized about the sale.

The most technical question here ("technical" in a literal sense, not legally technical) is whether, as the motion says, "Qualcomm’s chipsets and software indisputably are a matched pair by design, each intended to be used with, and only with, the other." The moving parties describe this as an undisputed fact. It will be interesting to see whether (and if so, how) Qualcomm will seek to raise doubts about it, but absent a single reasonable use case for that software outside of Qualcomm's chipsets and for Qualcomm's chipsets without the related software license, I guess it just can't deny what matters here.

Exhaustion can only be determined on a patent-by-patent basis (even if certain key questions, particularly anything related to Lexmark, will overlap). It's not a portfolio question.

That's because it must be shown for each patent that it is practiced by Qualcomm's chipsets. It wouldn't be practical to have one or more courts make this determination for thousands of patents. And a motion for partial summary judgment is subject to rather strict page limits--the whole idea of summary judgment is that you raise an issue that the court can easily decide. Apple and the likes of Foxconn therefore picked three exemplary patents (from the ones Apple and the contract manufacturers tackled before). Footnote 1 says that "[a] favorable ruling on this motion would inform the parties as to the exhaustion issues for the other patents." In other words, the parties need some guidance, but then they should be able to figure out exhaustion for numerous other patents.

Footnote 6 explains, in other words, that the movants are simply in their right to present alternative claims: should they prevail on exhaustion, infringement won't be reached. While the motion doesn't say so, this would even apply in a case where the theories aren't just alternatives but where they would be contradictory. Here, however, it's technically still possible that Qualcomm's chips practice a patent (thereby triggering exhaustion, provided that a sale is authorized and a court is underwhelmed by an attempted end-run based on a separate software license and a nested corporate structure) while, for an example, Intel's chips don't. However, to the extent that a patent is truly standard-essential, it may be practiced by Qualcomm and infringed by everyone else--but who cares if it's exhausted?

The FTC's motion for partial summary judgment in San Jose has the potential to force a settlement of that entire antitrust matter. If the San Diego motion on exhaustion succeeds, there will still be a host of financially and strategically important issues to resolve, but the case would be streamlined in a highly important way and, in a best-case scenario, it's possible that no (alternatively, fewer) patent infringement questions would have to be taken to trial. And above all, either motion would have major impact on the market.

Share with other professionals via LinkedIn: