Wednesday 18 December 2019

Biting the hands that feed you SEP technologies

Longstanding and economically efficient balance in Standard-Essential Patent licensing is being destabilized by misinformation and manipulation of commercial practices and of benchmarks in Fair, Reasonable and Non-Discriminatory licensing. This is epitomized in litigation between Apple and Qualcomm, commencing January 2017, until settlement in April 2019 and in the US Federal Trade Commission’s antitrust action against Qualcomm also commencing around the same time and now on appeal. Many in the cellular industry—who unsurprisingly like the idea of lower prices for something they assimilate and must pay for rather than that they create or sell—have lapped this up. But that does not justify the disingenuity or negate the resulting harmful effects that undermine incentives for ongoing and long-term technology development in the mobile communication sector including 5G and the emerging Internet of Things.

“Goal: Reduce Apple's net Royalty to Qualcomm” [1]

Since the 1990s, Apple among others have taken various and elaborate steps to "Devalue SEPs." As indicated in the next three endnotes, I have analysed the failings in the first three of Apple's proposed measures for "Reshaping FRAND — Licensing, Litigation & Competition Law" including:
  • “Base = derived from smallest priceable component (i.e. baseband)”[2]
  • “Rate = no higher than adjusted pro-rata share of SEPs”[3]
  • “Control for quality, over-declaration & royalty stacking”[4]
  • “Build favorable, arms-length ‘comp’ licensees”

And, regarding the fourth measure, for example, Apple’s actions over the years in the run-up to suing Qualcomm were in “Creating Evidence” for the lawsuit so “[it]can leverage [its]purchasing power” and “captur[e]IP value with purchase power.” To achieve this, Apple “selectively filter[ed][deal-flow] pipeline to identify the most desirable deals,” it “Us[ed]Liabilities as an Asset” and “Evaluated risk, cost and use[d]as evidence… as a comparable in disputes with others.” Apple sought to “create leverage by building pressure” that would “hurt Qualcomm financially,” put “Qualcomm’s licensing model at risk” and “drive Qualcomm to engage Apple” on a significantly weakened basis.

Apple’s multi-pronged attack included coercing its suppliers—contract manufacturers Foxconn, Pegatron, Wistron and Compal—into violating supply agreements, including those forged with Qualcomm two or more years before a couple of them even started making iPhones for the launch of the very first model in mid 2007. Apple’s stipulations demanded “[CM] does not settle any such claim or allegation, or make any admissions of liability or admissions relevant to the claim or allegation (related to the Goods), or take any other action that [CM] knows or should reasonably know will harm Apple’s position(s) with respect to the claim or allegation, without Apple’s prior written permission.” For example, according to one of them: “our customer has recently [as of April 2017]requested compal to stop the royalty payment to Qualcomm.” For CMs to continue paying would have left them incurring losses by not being able to recover licensing costs along with component and manufacturing costs, as was the norm. “Apple may notify Supplier that Apple will not pay Supplier any amount attributable to mobile technology license fees at any time, and after the date of any such notification, any mobile technology license fees that Supplier pays are the sole responsibility of Supplier.”
Timeline of Contract Manufacturers’ Subscriber Unit Licensing Agreements

Unlike Apple with gross profit margins of around 60 percent on its iPhones, corresponding figures for CMs are one tenth of that and CMs typically have operating margin percentages in the low single digits. Qualcomm had been receiving licensing royalties of $7.50 per iPhone until Apple forced its CMs to stop paying. That was, for example, one fifth the cost of the camera and versus total component, assembly and test costs of $325 [Endnote 5] for the iPhone 8 Plus. The unsubsidized retail price of this model was $799. Product costs and prices are from 2017/2018 for the cheapest model with 64GB of memory.
The restrictions Apple imposed on its CMs were in overt breach of its (2013-2016) Business Cooperation and Patent Agreement (BCPA) with Qualcomm for which “Apple [was]at risk for infringement, tortious interference and full royalties (plus and interest, penalties, etc).”  The BCPA required that “Apple shall not knowingly take (or continue taking) any action against or make any demand of any Qualcomm Licensee that prevents, restricts, or discourages such Qualcomm Licensee from which it purchases Apple Devices from complying fully with the terms of such Qualcomm Licensee’s QC License Agreement.”  Apple found it “Beneficial to wait to provoke a patent fight until after the end of 2016” when the flow of all the benefits it received from Qualcomm ceased with expiration of the BCPA.

IP value is unrelated to publication medium, hardware and manufacturing costs

We all understand and accept that most of the cost for books, recorded music, and movies is for the intellectual property added rather than in the publication medium. Similarly, with generic drugs typically being 80-85 percent lower in price than patented pharmaceuticals, we recognize that most of the value in the latter before patent expiration is in IP rather than in the cost of medicine manufacturing.
The parallels among book printing, audio or video disc manufacturing and chip manufacture are closer than one might imagine. In all these cases, production plant is quite generic, typically operated by independent manufacturers and can be rapidly reconfigured to serve the needs of different customers and products. Declining manufacturing process costs including in silicon foundries have little or no relationship with costs of content or technology development elsewhere. While Taylor Swift sounds and has very different IP ownership to Black Sabbath, CD production of their respective albums is as oblivious and independent of that as is TSMC’s chip foundry to the cellular or video codec SEP ownership and to the implementation of cellular modem designs by MediaTek versus Huawei’s HiSilicon. The production of CDs and chips is, therefore, rather like printing from a commercial perspective.
So why is there such resistance to paying total royalties of no more than around five percent of smartphone costs in patent licensing fees in patent-rich devices such as smartphones?
IP costs are typically buried and invisible in technology product bills of materials. Where a manufacturing company develops its own IP for the products it produces, the R&D cost is accounted for as a fixed and sunk indirect cost—not a variable cost in production, as with the addition of a camera or baseband chip. The same accounting treatment applies when that IP value is cross-licensed to other manufacturers for net-zero or much reduced net licensing charges. In some cases, concerted action among product suppliers insists that patented and other technologies are only licensed to those who contribute their IP gratis, as is the case with open source software and some royalty-free patent pooling arrangements, as is the case with the Bluetooth Special Interest Group and with the USB Implementers Forum. They do that because costs are recovered from their financial returns on selling products and services that incorporate those technologies. It is only when that IP is procured from elsewhere in licensing for cash payments and without the quid pro quo in cross-licensing the manufactures of others that true values are revealed anywhere, such in the management or financial accounts.

Qualcomm was targeted despite and because of its superior patent strength

Apple recognised internally that “Qualcomm is widely considered the owner of the strongest patent portfolio for essential and relevant patents for wireless standards.” However, with the measures identified above, in public statements, in its litigation complaints against Qualcomm and in its sponsorship of the FTC action against Qualcomm, Apple sought to undermine SEP valuations generally. Qualcomm leads among owners of wireless SEPs and SEPs reading on other standards applicable to Apple’s products implementations. Apple recognises Qualcomm superior position: “compared to others [Huawei, Nokia, Ericsson, InterDigital and Apple], Qualcomm holds a stronger position in each of the [ WiFi, audio/video and cellular SEP] categories, and particularly with respect to cellular and WiFi SEPs”.
Apple also recognizes that Qualcomm “has significant holdings in other areas, including many areas relevant to Apple… Compared to other licensors, Qualcomm has more significant holdings in key areas such as media processing, non-cellular communications and hardware. Likewise, using patent citation analysis as a measure of thorough prosecution within the US PTO, Qualcomm patents (SEPs and non-SEPs both) on average score higher compare to the other, largely non-US based licensors.”  And yet, Apple’s much-published arguments were also that Qualcomm’s non-SEPs are not worth much.
Qualcomm’s licensing-for-money business model was the most lucrative target for Apple in potential cost savings, because licensing by Ericsson and Nokia—that used to license defensively to protect handset business downstream and still have large downstream network equipment businesses—continue to be limited—particularly by their legacy comps including extensive cross-licensing—in the extent to which they can be fully and property rewarded in cash royalties for their patent value.

Good value for money

While some OEMs have managed to “hold out” from paying their fair share in patent licensing fees, the status quo in SEP licensing under FRAND terms has worked rather well overall. For example, billions of consumers use gigabytes of mobile data per month on smartphones that are their primary or only means of internet access. In the US, consumers spend more time on their mobile devices than they do watching television, in significant part because video streaming on these devices is substituting substantially for the former. At 3 hours 43 minutes per day in 2019, the average U.S. adult spends more time using all their mobile devices (including smartphones, tablets, etc.) than they do watching television.
Apple’ CEO Tim Cook even publicly recognized the value of improvements in cellular technology. For example, in an April 2016 call to investors he said “[T]he LTE rollout with India just really began this year, and so we’ll begin to see some really good networks coming on in India. That will unleash the power and capability of the iPhone in a way that an older network, a 2.5G or even some 3G networks, would not do.”
Licensors of the standard-essential technologies that have made much of the above possible are not simply free riding on previous accomplishments in their licensing demands.  Leading SEP developers, including Ericsson, Nokia and Qualcomm in particular, have no direct share of the large mobile phone product market, which is worth nearly $500 billion annually and provides stellar profits to Apple, or of the operator services market worth $1 trillion or of the revenues Internet platforms and applications including Android, Uber and Facebook generate from mobile devices.  Consequently, these licensors need and deserve adequate compensation, by other means, for their major R&D developments in SEP technologies.
New technology developments are enabling an accelerating pace of improvements including 5G which, for example, have unlocked access to the mmWave spectrum that is providing orders of magnitude more cellular network speed and capacity than was even thought conceivable less than ten years ago when 4G was first introduced. With significant further developments still required, 5G promises so much more in ultra-reliable and low-latency communications including connections to a multitude of things as well as people.
The pace of innovation is increasing, but it is largely the same relatively few companies that make most of the technical contributions to standard setting organizations, such as 3GPP for cellular, and that file SEP declarations to the ETSI IP rights database.  Some of those companies are increasingly dependent on licensing for fees rather than selling manufactured products to make a return on their large R&D investments.

Legal tussles and consequences of Decisions

There is much ongoing legal dispute with tensions among patent law, contract law and antitrust law: in Qualcomm’s appeal to Judge Koh’s Decision in the US FTC case and elsewhere among other litigants. Matters include those of patent exhaustion, where in the supply chain licensing may or should occur, tying and exclusive dealing, the meaning for FRAND, availability of injunctions, and jurisdictional issues among nations in contract interpretation and in global licensing.  Judgments on appeal in the US and EU and in the UK Supreme Court (i.e. Unwired Planet v Huawei; Conversant v Huawei and ZTE) could bring about major disruptive changes with significant unexpected as well as expected consequences. Litigants should be wary of possible and likely adverse long-term consequences in what they wish for, given the huge success of the mobile sector for manufacturers, operators and the enormous benefits it has provided to consumers.
I will finish with another idiom that comes to mind with the British Pantomime season upon us. While the idiom is from Mother Goose, it is more widely known from is origins in one of Aesop’s Fables. Don’t kill the goose that lays the golden egg! 


[1] This and all subsequent quotes, apart from a public announcement by Apple’s CEO, Tim Cook, are from internal documents at Apple, as found through discovery in the litigation between Apple and Qualcomm, and as revealed publicly for the first time in Qualcomm’s Opening Statement presentation at trial on April 16, 2019. In Re: Qualcomm litigation Case No. 3:17cv0108-GPC-MDD (S.D. Cal.) The trial was terminated very shortly thereafter that day with dispute settlement between Apple and Qualcomm. See also:
[5] TechInsights estimates for component, assembly and test costs. 

Keith Mallinson is a leading industry analyst, commercial consultant and testifying expert witness. Solving business problems in wireless and mobile communications, he founded consulting firm WiseHarbor in 2007.

No comments: