Intellectual Property

What is 5G? Well, if you asked that question 30 years ago, people would think you were referring to US$5,000 (i.e., “five grand”), which at the time represented more than a 50 percent down payment on an average new vehicle in America. Today, 5G means fifth-generation cellular network technology that envisions high data rates, reduced latency, orders of magnitude increase in bandwidth, and massive device connectivity.

With 5G, the promise of autonomous vehicles safely and efficiently gliding down roads and highways everywhere can become a reality. Such a promise, however, can only be achieved through the thoughtful setting of technology standards so that every vehicle is on the same page of a very complicated playbook of vehicle-to-vehicle, vehicle-to-network, vehicle-to-infrastructure, and vehicle-to-pedestrian communications, much of which will be covered by thousands and thousands of patents. If, for example, one OEM's self-driving vehicle could not seamlessly and reliably communicate with another OEM's self-driving vehicle, the promise of safer and more efficient personal transportation quickly falls apart. The questions of which patents cover the technology necessary to run this complicated communications playbook and how to license them represent a major issue for the automotive industry.

Reminiscent of nineteenth-century settlers of Oklahoma, companies are already stockpiling patents on inventions that may be used to comply with 5G-related technical standards, positioning themselves for a modern-day land rush. In exchange for a standard setting body's adoption of a company's suite of patents, the company must contractually bind itself to refrain from seeking to enjoin unlicensed implementers, in favor of licensing them on terms deemed Fair, Reasonable and Non-Discriminatory (FRAND).

FRAND by its literal terms suggests a desirable even-handedness. In a technology ecosystem that must implement standards to enable the ultimate goal of a network where vehicles communicate with each other as well as with road surfaces, traffic controls and other connected endpoints, FRAND licensing of standard essential patents (SEP) is an unquestionable prerequisite. If the past decade's smartphone patent wars and the evolution of the mobile telephone market have taught us anything, it is that what is “fair” or “reasonable” to some, may be the antithesis of that to others. Non-discrimination sounds good until it is invoked to charge everyone—from an inexpensive 5G fitness bracelet to a more expensive autonomous drive vehicle—the same percentage of net sales.

In the 4G LTE world of today, dominated by smartphones, tablets and handhelds, such a disparity among devices is not the issue it will be when 5G standards will be applicable to a much broader range of connected products. And yet, fights already exist between SEP owners and SEP implementers over a range of issues, including how to determine the essentiality of an alleged SEP, the reasonableness of a “reasonable” royalty, and what it means to be “non-discriminatory.”

|

What is “Reasonable”?

One of the major issues in the current world of FRAND litigation is how to value a standard essential patented technology when it is merely a small, but necessary component in a larger, more expensive device. While paying 3 percent of the net sales of a smartphone to a SEP holder may be appropriate where the SEP covers a computer chip responsible for the entirety of the functionality of that smartphone, what percentage of an autonomous vehicle's value should be apportioned to a SEP covering a single chip found within a US$100 telematics control unit in a US$70,000 vehicle? And what happens when producing an autonomous vehicle requires the licensing and use of hundreds, or even thousands, of SEPs? Obviously, the manufacturer cannot pay each SEP holder 3 percent of net sales, lest the cost of licenses swallows the profit margin (if not the entire value) of the vehicle. Recent case law suggests that neither the courts nor industry players have yet settled on a single method for determining the FRAND value of a SEP.

In 2013's Microsoft v. Motorola, Microsoft sued Motorola over the latter's alleged failure to license its 802.11 and H.264 standards patents on FRAND terms. Motorola argued that royalties should be determined on the basis of a modified Georgia Pacific analysis based on competing experts' views of a hypothetical negotiation between a patent holder and a patent implementer at the time of the contemplated implementation. Microsoft, however, argued that royalties should be determined based on the incremental value of the patents in view of alternatives. Its suggested approach, argued Microsoft, was necessary to address the problem of “royalty stacking”; that is, the problem created when the sum of individual payments requested by different patent owners is higher than the payment that a single company would have requested if it owned all of the relevant patents. The court sided with Motorola, adopting the modified Georgia Pacific approach, but did not address the stacking issue. This analytical method and resulting decision were upheld on appeal in the Ninth Circuit.

In In re Innovatio IP Ventures, LLC Patent Litigation, the court applied a similar Georgia Pacific analysis, but addressed the royalty stacking issue by using a “top-down” analysis which took into account all SEPs for the relevant standard and adjusting the value of Innovatio's SEPs accordingly. In TCL v. Ericsson, the court there also adopted this “top-down” approach, and further adjusted for the regional strength of Ericsson's portfolio considering that it had filed fewer of its SEPs in Europe and elsewhere as compared to the United States.

In Ericsson v. D-Link Systems, Inc., the Federal Circuit likewise held that the jury must be instructed to apportion the value of a patented invention apart from the value it derives from being a part of the standard, as well as apart from unpatented features reflected in the standard. The court further held that there is no “one size fits all” version of the 15 Georgia Pacific factors, and that jury instructions in FRAND cases should be based on the facts of each case. The court noted that “[t]hese steps are necessary to ensure that the royalty award is based on the incremental value that the patented invention adds to the product, not any value added by the standardization of that technology.”

More recently, in Optis Wireless Tech. v. Huawei, the court instructed the jury that in addition to the modified Georgia Pacific factors, it “may consider any evidence of patent hold-up and royalty stacking.”

|

What is “Non-Discriminatory”?

Two recent rulings from the Northern District of California and the Eastern District of Texas provide insight into the different approaches that courts may take in deciding what FRAND commitments should be for potential licensees in a distribution chain (i.e., from the manufacturer of components to the manufacturer of complete devices or products). The different approaches are of critical importance for vehicle manufacturers who rely on lengthy and complex distribution chains. Depending on which of these two approaches becomes widely adopted, large ramifications on the value of an autonomous vehicle SEP, and in turn, the cost of licensing to the SEP implementer, await the automotive industry.

In Fed. Trade Comm'n v. Qualcomm Inc., the court held that Qualcomm was required to license its SEPs on FRAND terms to all potential licensees in a distribution chain, even where component manufacturers did not create a full product that practiced the standard on its own. The court rejected Qualcomm's argument that it was only obligated to license its SEPs on FRAND terms to manufacturers of devices that actually practiced the standard. The required licensing on the component level (e.g., a telematics unit) generally has the practical effect that the patent holder could not extract a further royalty from the manufacturer employing the component in a larger device (e.g., a vehicle) under principles of patent exhaustion.

By contrast, in HTC Corp. v. Telefonaktiebolaget LM Ericsson, the court held that FRAND commitments did not require that a SEP holder license its patents to the maker of components—the smallest salable patent-practicing unit. The court held that under French law controlling the interpretation of the standard-setting commitment entered into by the SEP holder, the patent owner was neither required nor precluded from offering a FRAND license based on the smallest salable components covered by the patent. The court further noted that there is a general industry practice of “offer[ing] licenses based on the price of the end-user device, and not on the [price of a] component.” Under this approach, a SEP holder could refuse to license the component(s) (e.g., a control unit in a vehicle) and instead seek licenses only from the manufacturers of the larger (and more expensive) products (e.g., the vehicle) employing the component.

Although the courts in HTC and Qualcomm relied on different governing laws—indeed, the laws of a different country in HTC—the two courts' differing approaches are important to note. While differences among the two approaches may eventually be resolved at the appellate level (or even via legislation), SEP holders and SEP implementers should continue to pay close attention to the differing approaches in the jurisdictions where their cases may develop. In the meantime, automotive OEMs are forced to formulate various supply chain, IP, financial and governmental relations strategies as this uncertainty plays out.

|

Conclusion

The competing interests of SEP holders and SEP implementers, to date, have made for contentious litigation despite everyone's agreement that rational implementation of FRAND concepts is necessary for widespread adoption of 4G LTE standards. As the market transitions to 5G, with its much greater product and economic disparity between small personal devices and large devices such as autonomous vehicles, players in the automotive space will need to be even more vigilant about evolving FRAND concepts, seek judicial change, actively participate in the relevant standard setting organizations, and even seek appropriate legislative or regulatory action.

Darryl M. Woo is an IP litigation partner at Goodwin Procter LLP in San Francisco, CA; and Raymond Millien is Vice President and Global Chief IP Counsel at Volvo Car Group in Gothenburg, Sweden. Both wish to thank Monte Cooper and Luc Dahlin of Goodwin Procter LLP for their assistance in preparing this article. This article reflects the authors' current personal views and should not be attributed to their current or former employers, or their respective clients or customers.