Friday, March 11, 2022

Access Advance's duplicitous royalty policy pretends to be FRAND but fails to withstand scrutiny: patent pools must not borrow from Ponzi schemes

There's been some talk this week about Access Advance's March 8 announcement of an update to its duplicate royalty policy. And once more, it's a non-solution. It would be a running gag if there weren't actually companies suffering due to pending and threatened infringement actions.

For a recap, the Landgericht Düsseldorf (Dusseldorf Regional Court) entered a landmark set of rulings that resulted in the first finding of un-FRAND-liness of a patent pool's terms. The issue was not the rate (which the Dusseldorf court had blessed before) but double-dipping: defendants like Vestel previously took an MPEG LA license and are therefore already licensed to many patents in the Advance pool (potentially also Samsung's HEVC (H.265) patents), but are now being pressured to take an HEVC Advance license without a reasonable credit or refund mechanism to avoid double-dipping. There have been various cases in which courts relied on pool licenses as comparable agreements, but there had not been a previous case in which a pool's terms raised issues. To add insult to injury, the normally rather patentee-friendly court not only found Vestel's FRAND defense meritorious but also declared that the defendant and counterclaimant was entitled to antitrust damages (of an amount to be determined subsequently). Access Advance now claims to have gone beyond the Dusseldorf court's requirements, but as I'll show further below, they still refuse to do what really needs to be done.

Earlier this year Access Advance, the GE-Philips-Dolby-Mitsubishi entity that is all about generating supra-FRAND income (in the combination of patent royalties and pool management fees) for its four owners, already claimed to have learned its lesson. However, a first attempt to address the Dusseldorf court's concerns had some glaring deficiencies. In February, the Dusseldorf appeals court overturned an anti-antisuit injunction against Xiaomi and mentioned the patentee's FRAND defeat in the lower court. Meanwhile I've found out about a number of Access Advance v. TCL cases pending in Munich.

There we are, and now Access Advance is giving it a third try. While there is a web form to request a copy of the new terms ("Duplicate Royalty Adjustment Amendment"), the policy statement summarizing the terms of that amendment is publicly accessible (though it's not straightforward to navigate there). So let's talk about that one, but first I'm going to edit the headline (click on the image to enlarge):

What makes that policy duplicitous is that unsophisticated observers, or sophisticated ones with a two-second attention span, may look at it and actually be led to believe that Access Advance is heeding the Dusseldorf court's criticsm. In reality, it looks like the economics of the HEVC Advance patent pool simply don't make it a profitable option to treat implementers fairly if they are already licensed to some of the patents in question. That shouldn't be an issue: patent pools have to avoid double-dipping all the time, and if, say, 30% of the patents in a pool (assuming for the sake of the argument that they're all equally valuable) have already been licensed, it's common sense that someone taking a license from that pool shouldn't pay a lot more than 70% of the standard pool rate. But Access Advance apparently can't do that without losing money, in which case the approach in question is broken beyond repair.

Were this a motion-to-dismiss process in the U.S., and had the latest policy been presented as a Second Amended Complaint, a renewed motion to dismiss would tell the court that almost all of the issues previously raised still exist. Blog posts aren't formal proceedings, so I can show you very quickly and easily why they haven't really addressed all of the issues.

Only refers to other pools, not bilateral licenses or exhaustion

This link (a so-called anchor link) takes you directly to the part of my January post on the previous version of that policy. In that one, I wrote:

"The first glaring deficiency is that it talks only about patents that are "also included in the patent list of another patent licensing pool, or joint licensing program," without addressing the scenario in which someone took a direct (bilateral) license, which is not at all uncommon in this industry. Bilateral licenses can result from license agreements or even from patent exhaustion. The Access Advance folks know that, and it must be attributed to bad faith that their Duplicate Royalty Policy fails to address that problem."

Nothing changed in that regard: in order to be eligible for the (questionable) benefits of that policy, a company must be "a licensee of another HEVC patent pool or joint licensing program (e.g., the HEVC patent pool administered by MPEG LA)."

As far as I know, Vestel's FRAND defense and counterclaim were based entirely on a prior MPEG LA license, not on a bilateral license or patent exhaustion. But that doesn't make that the other scenarios wouldn't apply to other companies. In January I wrote "the root cause of the Dusseldorf disaster is deep and structural." If Access Advance was sincere about becoming FRAND-compliant, they wouldn't wait until some future defendant raises a related issue: they'd simply recognize that double-dipping, for whatever reason it would occur, isn't FRAND, period.

On a related note, it's debatable whether it's reasonable for Access Advance to limit the applicability of its policy to scenarios in which an implementer took the other pool license first. This is, again, a non-issue in the Vestel and Xiaomi cases: they signed with MPEG LA first. MPEG LA's HEVC pool is the older one, which means a company can have signed up to MPEG LA before the HEVC Advance pool even existed, but not the other way round. But let's look at it this way: Access Advance has told implementers to go seek a refund from MPEG LA, but that wouldn't even be an option if MPEG LA, too, had formulated a duplicate-royalty policy according to which no adjustments will be made if someone subsequently licenses some of those patents elsewhere. So even if one agreed that it's fair to say that the older pool license takes priority, just like an older patent application beats a younger one in an interference proceeding, Access Advance is now taking a position that is self-serving and self-contradictory at the same time.

Commitment, not merely referral

The previous version of the policy had one weakness that has meanwhile been addressed, but on the bottom line licensees still won't really get a FRAND deal. In January I wrote:

"In a duplicate-license sceario, Access Advance does nothing for licensees other than to 'refer the request to the applicable Licensor(s)' (footnote 3), which falls far short of what the Dusseldorf court wanted, which was legal certainty for licensees."

And I noted that no one needs a policy for a scenario in which a licensor and a licensee agree. If they're on the same page, there is no problem left to be addressed by the pool. All that Access Advance has done is to eliminate that "referral" part. Instead, "Advance commits to deduct the amount of duplicate royalties," but as you'll see, the way they calculate "the amount of duplicate royalties" raises the very same issues as before.

In a nutshell, they do commit to a deduction right away, which would be an improvement over the previous state of affairs if the deduction was FRAND, which it is not.

A commitment to unfairness is the same from the point of view of an implementer seeking a credit or refund as a non-commitment to fairness.

In the next two sections I'll raise the two reasons for which the deduction they commit to falls far short of meeting FRAND criteria.

Access Advance acknowledges that some licensors--presumably including Samsung--receive no royalty payments

In January I already wrote that "Samsung may contribute lots of patents but not actually get much (if any) money out of the pool." That wasn't only about Samsung: there may be others. But Samsung is by far the largest HEVC Advance licensor, and previously was an MPEG LA licensor. It makes sense to focus on that example for practical reasons.

More than one industry source has told me on background that Samsung may not get much--if any--money out of the pool because it may have contributed its patents just to get a cheap license (or the next best thing to a zero-zero cross-license) for its own mobile devices and TV sets. One my intuitively feel that it's simply Samsung's choice to use its patents as a bargaining chip. If they're interested only in their margins as a device maker, that's a legitimate parameter--but companies like Vestel are still entitled to a license to Samsung's patents on FRAND terms.

Samsung has an interesting history with respect to FRAND. The European Commission deemed it a FRAND abuser in its dispute with Apple, though there were extenuating circumstances as it was just retaliation for non-SEP assertions and Apple's royalty demands, whether over rounded-corner design patents or software patents covering only particular implementations of minor features, weren't perfectly reasonable either. In 2014, Samsung withdrew its SEP assertions against Apple, and since then has consistently been advocating the devaluation of SEPs, just like Apple. However, the net effect of Samsung's sweetheart deal with Access Advance raises FRAND question. Even if one gives Samsung the benefit of the doubt (it might not have anticipated what would happen in the Vestel cases), it needs to be concerned in its patent dealings with preserving its consistency because there's a lot more money at stake for Samsung with respect to cellular SEPs than video codec patents.

Why are Samsung's competitors--and even companies that may implement HEVC in market segments in which Samsung has no presence, though Samsung is that big elephant in the room that it's almost impossible not to compete with--potentially harmed?

Assuming that my industry sources are right, Access Advance's new policy still wouldn't entitle those who took an MPEG LA license early on to a deduction with respect to Samsung's patents. There's a conspicuous conditional clause there:

"If a Licensor of Advance's HEVC patent pool would receive a share of royalties from Company under the HEVC Advance Patent Portfolio License Agreement (the 'HEVC Advance PPL') on account of products of Company licensed under such Licensor's patents that are included in both Advance's HEVC Patent Portfolio and the HEVC patent portfolio of another HEVC patent pool or joint licensing program (a 'Dual-Pool Licensor'), Advance commits to deduct [...]" (emphases added)

As Yogi Berra famously said, when you come to a fork in the road, take it. Here, there is a fork in the road because an IF is a condition that can be met, but need not be met. So we have to ask ourselves the question of what scenarios exist in which the IF condition isn't fulfilled. The fact that Access Advance made the actual receipt of royalty payments a condition can only be interpreted as an admission that one or more licensors don't receive a share of the pool's royalty income. Otherwise the wording would have been much simpler and shorter.

Normally it's a given that someone who contributes patents to a pool receives a share of the royalties. Seriously, I've never heard anyone doubt whether, for instance, a company like LG or OPPO gets money from Avanci. The general public doesn't know exactly how much (though the DOJ's Business Review Letter discussed the Avanci model with a view to 5G), but no one would assume that it's zero.

If Samsung got value out of its participation in the pool other than royalty payments, that doesn't mean implementers who previously licensed its patents aren't entitled to a FRAND deduction. Samsung can't just say "sorry, we get no money out of the HEVC pool" (unlike from MPEG LA, a pretty transparent pool). The problem is that GE-Dolby-Philips-Mitsubishi use Samsung's patents to justify the pool rate. They still point everyone to the fact that the Dusseldorf court didn't take issue with the rate--but a rate that appears reasonable for the entire pool may be clearly supra-FRAND when you take out a huge chunk of patents (because of a prior license).

Patent pools must not become Ponzi schemes in the sense that the benefits promised to the first ones to join depend on revenue generated through subsequent transactions with others who will lose money. Here, Access Advance--which, again, is just about maximizing revenues from IPR for GE-Dolby-Philips-Mitsubishi, all of which but Dolby are a shadow of their former selves with respect to video codec innovation--attracted Samsung to the pool with what may be the closest thing to a zero-zero cross-license. But the deal only benefits Access Advance and Samsung if subsequently some other licensees have to pay the very high Advance rate, which they seek to justify, to a large degree, with the strength of Samsung's portfolio. Subsequent licensees must get a FRAND deduction if they've already licensed Samsung's patents through MPEG LA (or a bilateral deal).

A FRAND deduction must be reflective of the overall value derived by a licensor. With a conventional pool, that value is simply a royalty check. But value creation doesn't depend on money changing hands. If a patent pool gave Samsung raw materials in exchange for its contribution, it would still be good and valuable consideration, and a deduction to combat double-dipping would have to be reflective of that type of benefit or it would not be FRAND. Here, the value is that Samsung's own devices got licensed.

Access Advance should tell the world who the licensors that get no money out of the pool are, and what that is so.

Exorbitant pool fees are not reduced

With respect to pool management fees I wrote in January that "[i]n the aggregate of multiple types of fees and charges, Access Advance keeps roughly 40% of the royalty income, which is several times more than MPEG LA's cut according to what people in the industry say about it." Access Advance still declines to reduce those fees in a duplicative-license scenario. Let's resume where we left off further above. After the IF clause I discussed in the previous paragraph, this is what the policy says:

"Advance commits to deduct the amount of duplicate royalties for such Dual-Pool Licensor's patents under the HEVC Advance PPL from invoices sent to Company (i.e., pre-net),(3) thus preventing the Dual-Pool Licensor from receiving duplicate royalties for the Dual-Pool Licensor's patents."

The term "pre-net" is defined in footnote 3:

"Duplicate royalties are that portion of actual net royalty collections apportioned and distributed to Dual-Pool Licensors under Advance’s HEVC patent pool based on products of Company covered by one or more of the Dual-Pool Licensor patents licensed through both Advance’s HEVC patent pool and another HEVC patent pool or joint licensing program, and expressly exclude all fees and other allocations or deductions made by Advance prior to apportionment and distribution of net royalty collections to all Dual-Pool Licensors based on their respective licensed patents." (emphasis added)

So even in scenarios in which implementers already have a license to a substantial part of the portfolio, implementers would still have to pay the same sky-high pool management fees as if they hadn't previously obtained a license to many of the patents in question.

Courts should not be fooled. Competition enforcers might want to take a look should the problem persist, which it probably will because neither Samsung nor GE-Philips-Dolby-Mitsubishi appear prepared to ensure that all implementers are treated fairly.

Share with other professionals via LinkedIn: