December 29, 2014

On the draft IEEE patent policy

IPR policies of telecoms Standardization Setting Organizations (SSOs) mainly define how patents that become part of the standard should be treated, typically through the principle of a (F)RAND ((Fair), Reasonable And Non-Discriminatory) licensing commitment. These policies have been in place for many years, decades even, but are quite naturally subject to reviews from time to time. In spite of such reviews and occasional calls for more regulatory content though, the policies are more often than not left largely unchanged.

This likely demonstrates that the original policies continue to serve the market rather well. Quite some wide-ranging thought and foresightedness have gone into them in order to carefully accommodate the diverse and complex interests of users and innovators of advanced communication technologies in the marketplace. And ultimately, to make the standards commercially successful in the long-term.

However, a radical IPR policy change is now underway within the Institute of Electrical and Electronics Engineers (IEEE), an SSO responsible for numerous standards including "WiFi", or IEEE 802.11 as it's formally known. Since last year, a proposal for an update of the patent policy of the IEEE Standards Board Bylaws is being worked on by the IEEE-SA (IEEE Standards Association), a body within the IEEE responsible for standards. Compared to previous IPR policies of telecoms SSOs, this proposal is very detailed and far-reaching. At least to me it also gives the impression of being aimed at significantly changing how wireless SEP licensing will work in the future. Let's take a detailed look.

To begin with, the policy states that the determination of a "reasonable rate" for a SEP "should" consider (emphasis added):

a) "the value that the functionality of the claimed invention or inventive feature within the Essential Patent Claim contributes to the value of the relevant functionality of the smallest saleable Compliant Implementation that practices the Essential Patent Claim" (emphasis added),

b) "the value that the Essential Patent Claim contributes to the smallest saleable Compliant Implementation that practices that claim, in light of the value contributed by all Essential Patent Claims for the same IEEE Standard practiced in that Compliant Implementation" (emphasis added)


c) "existing licenses covering use of the Essential Patent Claim, where such licenses were not obtained under the explicit or implicit threat of a Prohibitive Order, and where the circumstances and resulting licenses are otherwise sufficiently comparable to the circumstances of the contemplated license" (emphasis added).

Requirements a) and b) appear to mandate how the apportionment of the invention's value should be done. The phrase "smallest saleable compliant implementation" must reasonably mean a component or sub-assembly within the end-product, and in my mind the so-called "chip-set" might well be a likely candidate for this. The invention's value to such a component only is all that shall be considered. As I touched upon in my previous post on cellular FRAND royalty levels, this is actually not in line with how damages for patent infringement are, or should be, judged by courts. Indeed, as the US Federal Circuit recently clarified in the D-Link case, it's the value of the patented invention to the end-product that is the yardstick. In clarifying the "Entire Market Value Rule", it stated that according to the “substantive legal rule”, the “ultimate reasonable royalty award” – that is, the result of e.g. multiplying a royalty rate with a royalty base – ”must be based on the incremental value that the patented invention adds to the end product” (emphasis added). Thus, mandating SEP value apportionment to consider only a component within the end-product is not the law.

Requirement c) is even more remarkable. This is because ever since the patent system was created, all licenses - including (F)RAND SEP licenses - have by definition been obtained "under the explicit or implicit threat of a prohibitive order". I say "by definition" because this is quite simply the basic legal right of a patent holder, a right to exclude. And as I elaborated in an earlier post, (F)RAND doesn't change this per se. So in other words, c) simply says that none of the WiFi licenses in existence at the time of adoption of this new IPR policy may be used as a reference for any new WiFi licenses. This "purging" of the historical record of WiFi licenses seems intended to "cement" the component value apportionment mandated by a) and b). Presumably because the royalty considerations of those preexisting licenses may have been based on the value that the SEP portfolio brings to the end-product rather than to a component within. 

Interestingly, in the above-mentioned D-Link case, the defendant argued for disqualification of existing license evidence on those very grounds, but the Federal Circuit rejected this: "In short, where expert testimony explains to the jury the need to discount reliance on a given license to account only for the value attributed to the licensed technology, as it did here, the mere fact that licenses predicated on the value of a multi-component product are referenced in that analysis—and the district court exercises its discretion not to exclude such evidence—is not reversible error."  In fact, licenses "predicated on the value of a multi-component product" are perfectly normal in practical wireless SEP licensing. And incidentally, they're also in harmony with the law.

To me, a), b) and c) seem to be specifically designed to usher in a brand new era of very low royalty rates for WiFi. Clause c) makes all old licenses irrelevant as references. And clauses a) and b), while not necessitating it, certainly encourage a significant lowering of the absolute royalty rate by mandating consideration of the SEP's value to (at most) the chip-set only, thus creating a low-product-value bias with regards to the absolute royalty. I'm not sure how much support there will be for such a change, but it's not in line with what the US Federal Circuit has said about the law.  

However, there's more of interest in the proposed new IEEE patent policy, relating to prohibitive orders:

"An accepted ...[RAND licensing commitment] ... precludes seeking, or seeking to enforce, a prohibitive order except as provided in this policy." (emphasis added),

where "as provided in this policy" means:

"The submitter of ... [a RAND licensing commitment]... agrees that it shall neither seek nor seek to enforce a prohibitive order ...unless the implementer fails to participate in, or to comply with the outcome of an adjudication, including an affirming first-level appellate review, if sought by any party within applicable deadlines..." (emphasis added).

At first glance, this actually looks a bit similar to the process used e.g. in the US Federal Trade Commission's Consent Order in the Google/Motorola case. As a potential "softer" means to combat the very destructive"patent-holdout" behaviour, I myself saw that as a quite interesting path. However, when looking more in detail at the wording, it doesn't look like that's the purpose here. The phrase "an adjudication, including an affirming first-level appellate review" gives the distinct impression that an affirming first-level appellate review is actually a requirement. This makes it quite a different story from the quoted US FTC process. Unless I misunderstood the phrase, the whole thing looks like an attempt to in practice ban prohibitive orders altogether in SEP portfolio cases. Of course, also this is something that the Federal Circuit has clearly declared an error, as I mentioned in my earlier post on portfolio licensing and patent hold-out.

So overall I'd say that the patent policy changes proposed by IEEE-SA represent large steps in wrong, unmotivated and even unsupportable directions. What appears to be a ban on prohibitive orders is not in line with US Federal Circuit precedent and would only worsen the patent hold-out problem and put artificial downwards pressure on royalty rates. The mandating of value apportionment based on only a component and the explicit disqualification of old license agreements as references also don't align themselves with Federal Circuit precedent. When stripped down to its bare essentials, the proposal looks rather like a path-clearing for a drastic cut of the absolute SEP royalty level for WiFi.

But isn't that good? Well if it had been shown that there was a clear problem with royalties being too high, affecting the market in negative ways, then the IPR policy could definitely be in need of a review and as a consequence possibly even an update. But evidence of concrete problems to that effect yet remain to be demonstrated. We can again look at the D-Link case, a case that was specifically about SEP licensing for the IEEE standard 802.11 ("WiFi") and which involved several prominent implementers of that standard. The Federal Circuit confirmed that no evidence of royalty stacking or patent hold-up - two phenomena sometimes claimed to create 'too high' SEP royalties - was presented and that they should never be presumed to exist just because they are theoretical possibilities: "The district court need not instruct the jury on hold-up or stacking unless the accused infringer presents actual evidence of hold-up or stacking. Certainly something more than a general argument that these phenomena are possibilities is necessary. Indeed,“a court should not instruct on a proposition of law about which there is no competent evidence.""

There are others who have questioned the purpose of this new proposed policy. For example, the IEEE USA board (!) recently sent a letter to the IEEE board, straightforwardly asking for an explanation of exactly what's wrong with the existing patent policy and what concrete problems the new policy proposed by IEEE-SA is supposed to solve.

The SSO IPR policies set the delicate basis for achieving a long-term balance between incentive for innovation and incentive for implementation. This balance is for the common good of technological innovation, a healthy marketplace for implementers, and ultimately consumer choice. However, it seems to me that in the new draft patent policy for IEEE, shorter-term business interests of implementers may have been allowed to take on rather more weight than they should have. To my knowledge, no obvious market need has been demonstrated for any of the changes proposed. Nor is there any legal mandate for them and in fact, some changes even tend to run contrary with US Federal Circuit opinion.

To the extent that there's an actual need for an update of its patent policy, hopefully the IEEE will eventually come up with a more balanced proposal than the present one.