What’s in a Name? The Quintessential Misnomers That Are Untested SEPs: Thoughts on the Division Bench Judgment in Bansal v. Philips

[SpicyIP]

The Delhi High Court’s recent Division Bench ruling in Bansal v. Philips Division Bench Judgment may well become one of the most consequential SEP/FRAND decisions in India so far. In an incisive post, methodically breaking down the Division Bench decision, Aniruddh Bhatia explains in detail the decision as it deals with essentiality, infringement, FRAND dance, calculation of damages, confidentiality clubs and exhaustion. Aniruddh Bhatia is an advocate at an IP-focused law firm with eight years of litigation experience, six of which have been primarily in SEP/FRAND matters before the Delhi High Court. He has been involved in significant recent litigations across telecommunications, audio coding, and video coding SEPs, with notable work on procedural developments including confidentiality clubs, anti-anti-suit injunctions, and pro-tem (ad-interim) deposits. He would like to add the following disclaimer:-

“I am grateful to my colleagues for their spirited debates and opinions, as well as the occasional moral support. Truly grateful to the amazing Delhi High Court for the learnings and experiences. The court’s IP docket has provided a lot of material to write on, and too little time to do it in! The views expressed are strictly personal and academic, and are those of the author alone, and are not representative of any organization/individual I have represented or am affiliated with.“

What’s in a Name? The Quintessential Misnomers That Are Untested SEPs: Thoughts on the Division Bench Judgment in Bansal v. Philips

By Aniruddh Bhatia

For facts and a brief overview of the dispute, see Ambika’s post here.

And then there were two!

Among the four (4) post-trial SEP judgments (see, Ericsson v. Lava, Philips v. Sukesh Behl, Philips v. M. Bathla and the judgment considered here- Bansal v. Philips) in India, two (2) have, thus far, gone against the patentees/claimed SEP holders (Philips v. M. Bathla and Bansal v. Philips). This includes the latest Division Bench (DB) judgement (Bansal v. Philips) of the Delhi High Court which overturned the very first Indian post-trial SEP judgment in the SEP/FRAND space. Unless we are in for more over-rulings, a 50% post-trial success rate appears a bit problematic.

Perhaps, if one could say so, the recent post-trial outcomes and a 50% success rate may also invite further reflection on the concept of pro-tem deposits. If memory serves right, since Nokia v. Oppo, there has been a noticeable rise in requests for pro tem. While most of these requests are pending and each case turns on its own peculiar facts and circumstances, the evolving jurisprudence may nevertheless present certain interesting questions for future considerations. In light of the guidance provided by the current DB judgment, it would be interesting to observe some of the questions that the Court may face or may be called upon to answer in the near future in relation to pro-tem or interim monetary arrangements,

1. The extent to which ad-interim or pro-tem monetary arrangements in SEP disputes may need to account for the possibility that final adjudication, applying the standards articulated by the DB, may differ from interim assessments (one could argue that these things shouldn’t matter since the stages are quite different and give the example when Roche was denied an interim injunction but won post-trial against Cipla in that famous pharmaceutical battle- See, Roche v. Cipla (interim judgment by a Division Bench) and Roche v. Cipla– the post-trial judgment by another Division Bench),
2. The degree to which considerations or fundamental yardsticks governing grant of pro tem or interim reliefs/deposits be aligned or differ from the substantial obligations ultimately required to be discharged by SEP holders at trial,
3. Can pro-tem amounts be delinked from any kind of a FRAND analysis- so much so that pro tem reliefs can be granted for entire portfolios and based on offers (or counter-offers) that may, post a trial, be found to be non-FRAND?

But today is not a day for the concept of pro-tem. Given the nuances, these aspects can be debated till kingdom come, and to paraphrase Ecclesiastes, to many judgments (and to appeals), there will be no end.

Coming back to the DB judgment, one can’t help but note how it uses first principles to dissect several aspects of SEP/FRAND litigations. For a neutral researcher or student of law, this presents an opportunity in understanding how trials can reveal much about parties’ claims and how courts of law appreciate evidence and facts. Additionally, it is a mini-treatise on SEP-FRAND litigation in practice. The DB puts, and rightly so, the onus on the patentee to demonstrate the truth of its claims- right from the claim of essentiality to the claim of having engaged as a willing licensor.

Perhaps this nuance was always important to remember- that SEPs are just patents self-declared to be essential and it is the holder of such SEPs that voluntary undertook the FRAND commitment, with no metric or evaluation provided by any Standard Setting Organization (SSO) or any governmental body regarding what a FRAND rate is for a given portfolio of patents. Therefore, it must be for the plaintiff or the patentee to prove that its patent is indeed essential and that it has conducted itself in accordance with its FRAND commitment.

All about steps, issues, and onus.

Since the one who makes an affirmative claim should prove it, the judgment states that the Plaintiff, who alleges to be the holder of a SEP, has the onus to establish the following,

1. The patent is indeed a Standard Essential Patent (SEP): This requires the existence of a standard set by a Standard Setting Organization (SSO) (para 3) and “complete mapping” between the elements of the standard set by the SSO and the features of the suit patent as claimed, i.e. patent claim-to-standard mapping (para 6).
2. Defendants’ product infringes the patent– either “directly” by way of patent claim-to-product mapping (para 11) OR, as is done in SEP cases, by the “in-direct route” that is based on the law of transitivity (A=B, B=C, therefore, A=C. whereby it must be shown that both the patent claim and the defendants’ product maps on to the standard (para 12 and para 13).
3. Defendants’ illustrative defenses: As against this, the Defendant can (i) “assert and demonstrate” that its product can function without the aid of the patented technology in the suit patent, and/ or (ii) “question the manner” in which the product-to-standard mapping has been done, “and the legitimacy and the propriety thereof” (para 14).
4. The Plaintiff is a willing licensor who is offering the use of the patent on F/RAND terms and that the Defendant is not willing to take a license on F/RAND terms, thus being an “unwilling licensee”. Pertinently, the question of “whether the defendant is, or is not, a willing licensee would arise only if the plaintiff, in the first instance, demonstrates that it is a willing licensor” (para 16), this onus being on the plaintiff (para 17). To prove the same, the Plaintiff will have to provide information to the defendant basis which the defendant can evaluate the offer, and this information goes to the root of the question of Fair, Reasonable, and Non-Discriminatory (FRAND) as per the Division Bench (para 17).
5. And it is only then can the Court “compel a Defendant to take a license at the terms offered by the plaintiff, or at other FRAND terms as may be fixed by the Court, and injunct the defendant from using the SEP in default” (para 1).

That appears to be neat, but as the bench notes- it is easier said than done. In what is perhaps a sobering thought, the DB candidly says,

The entire exercise is so intimidating that it is doubtful whether any Court can arrive at a precise and accurate determination of FRAND rates, following all the steps in their proper sequence and complying with all considerations. At least in our country, it appears, frankly, impossible for such an exercise to reach its right, and legally correct, conclusion.

The DB identified the following issues (para 37)

1. Is the suit patent a SEP? The DB eventually held that it was not.
2. Is Philips barred from asserting rights in the suit patent owing to principle of international exhaustion of patent rights. The DB eventually held that Philips was so barred.
3. Has the suit patent been infringed by the Bansals? The DB held that Philips could not establish infringement- neither by the direct route, nor by the indirect route.
4. Assuming the answer to the earlier three issues to be all in favour of Philips, are the damages as quantified and awarded by the ld. Single Judge, sustainable in law? The DB eventually found that the manner of computing damages was not sustainable in law quite apart from the fact that since there was no infringement, there could be no damages either.

I only look at some aspects of this judgment and there will be other posts on detailed analysis of other aspects- given that the judgment runs into 150 pages.

Nature of the suit patent and the question of essentiality

Upon claim analysis, the Court concluded that the suit patent’s claims related to a decoding device and were a product claim. The complete specification, in the opinion of the Court, created confusion in as much as it started by saying that the patent claims a method and that the same was contradictory to the claim wordings and the title of the suit patent.

The Court also concluded that the plaint and the witness testimony demonstrated that the claim was for a component or a device, and not the decoding technology itself (para 76). However, the suit patent did not “identify the location of the device of which protection is claimed” (para 80). However, clarity on the aspect of location of the device, as per the Court, came from Philips’ expert witness who deposed that the process is done using a chip, which is an integrated circuit (para 81). Ultimately, the Court concluded that,

82. The position that results is that the device forming subject matter of the claim in the suit patent resides in the chip, or in the PCB, which is contained in the DVD Video Player. Any item or circuitry outside the PCB is, therefore, equally, outside the suit patent.

The question of essentiality: While the DB noted that there was a standard set by a SSO, namely, the DVD forum, yet Philips failed to establish essentiality since (i) there were no patent claim-to-standard mapping charts and (ii) the so-called Essentiality Certificates (ECs) filed by Philips in respect of the US and EU counter-parts of the suit patent were not tendered “by any governmental or statutory authority who is conferred with power in that regard”. They were merely issued by what the Plaintiffs called “independent expert patent evaluators” and the Court opined that the ECs would need to be proved in accordance with Sec. 45 of the Evidence Act. Pertinently, no one from the issuing evaluators- Proskauer Rose LLP or from Cohausz & Florack entered the witness box (para 92) nor was there any affidavit from any one from those firms “vouchsafing the correctness or validity of the ECs” (para 95). Therefore, “ECs have not proved in the manner known to law”.

Infringement: Since the patents were not mapped onto a standard, the DB opined that Philips failed the indirect test (law of transitivity) of establishing infringement.

As regards direct test for infringement, Philips did not file a product-to-claim mapping chart nor did it provide any details regarding how the alleged infringement analysis had been conducted by it. Philips had placed reliance on the affidavit of one Mr. Ravi Babu for the purpose of establishing infringement but the said Mr. Babu never entered the witness box. This being so, a third-party could not have vouchsafed the contents of affidavit of another.

These appear to be more of evidentiary mis-steps by the plaintiffs that ultimately proved to be fatal.

You shall not reap, what you did not sow: the return of the SSPPU argument?

The finding that the patented invention resided in the chipset perhaps led to the unescapable conclusion that any damages, or royalty, cannot be in excess of the decoding device/ component. Philips never claimed that every component of the DVD player was patented in its favour (see, para 165). Therefore, the Court concluded that,

167. Philips’ claim to royalty would have, therefore, to fail, as (i) the suit patent was not for the entire DVD player, or all components therein, but for the PCB, or chip, contained in the DVD player, and (ii) no royalty rate, charged by Philips for licensing use of the PCB or chip per se was forthcoming.

Does this mark the return, or rather a new consideration, of the argument that any compensation for the SEP holder has to be on the basis of the chipset or the value of the component wherein the patented technology resides, i.e. the Smallest Saleable Patent Practicing Unit (SSPPU)?

The said argument was previously rejected by a single judge in the post-trial Ericsson v Lava by relying on foreign judgements and industry practices. Interestingly, that judgement too, is under appeal.

What is the SSPPU v. end-user device debate?

A fundamental debate in the SEP/FRAND space is whether the royalty amount for the SEPs should be computed based on the SSPPU which houses the patented technology (for example, the chipset) or should it be based on the end-user device (the mobile phone or the smart car or a smart meter). Advocates of the latter approach usually say that the SSPPU approach undervalues the contribution of the patented technology to the overall value of the device. Advocates of the former usually say that this will award to the patentee a sum which is not warranted given the patent’s actual technical contribution. Another fundamental problem here could also be justifying the end-user device-based calculations when the products are as varied as an automobile- where cellular SEPs may be less relevant than safety features or mileage, or products such as smart meters which do not use cellular technology in the same way as a mobile phone. Moreover, with the advent of standardized technology in med-tech, the debate hits closer to home where public interest becomes an all-important consideration.  A more detailed discussion on this aspect requires a separate piece.

Of course, the Court’s conclusion was based on the specific claim construction. One cannot deny the fact that a claim construction, in a future case, may lead to a different conclusion- and then, may be, damages or royalty will have to be paid on the entire final product or the end user device, and not the SSPPU.

Interestingly, Philips did contend that it offered royalty on a per DVD player basis. But the court noted that that was merely a commercial decision and “cannot constitute the basis of determination of payable royalty in a litigation premised on a specific patent, claiming a specific device”. The DB also noted that the observation of the Ld. Single Judge that the decoding device was an integral part of the DVD player cannot justify computation basis the entire DVD player (para 166).

Philips had exhausted its patent rights

Once the DB concluded that the invention of the suit patent was contained in the chipset or the Printed Circuit Board (PCB), it also reached the conclusion that the defendants were protected by the doctrine of exhaustion, i.e. the first sale exhausts the rights of the patentee. This was because the defendants had submitted that they purchased MediaTek chipsets from two registered vendors of MediaTek (and letters from these vendors had been filed by the defendants). The defendants’ witness was never cross-examined on these aspects and the assertions, therefore, went unrebutted, as per the DB.

More importantly, the DB clarified that there is no requirement under the Indian law to establish that the dealer from whom the patented product was purchased by a defendant is duly authorized by the patentee to sell and distribute such product. It would be sufficient if such dealer was duly authorized under the law to sell the patented product. This, as per the DB, was due to the 2022 amendment to Sec. 107A(b), Patents Act wherein the words “duly authorized by the patentee” were amended to “duly authorized under the law”. And it was nobody’s case that the vendors from whom defendants had obtained the MediaTek chipsets were not authorized under the law to purchase or sell the MediaTek PCBs. Perhaps a piece specifically on the principle of exhaustion may follow from another author.

The FRAND rate and the role of Confidentiality clubs

Since FRAND determination is at the core of SEP disputes, it might be a good idea to start with what the Court says in this context. The Court stated that the onus to prove the FRAND rate is “squarely” on the Plaintiff (para 19) and “Ideally, this would require the plaintiff to disclose the entire agreements with the third parties in question, so that the defendant, as also the Court, is in a position to ascertain for itself that the rate is not influenced by extraneous or ulterior factors…” (para 20). It also noted that if there are confidentiality concerns, the same could be dealt with by constitution of confidentiality clubs.

This reasoning about “entire agreements”- which could mean (i) all or entire set of agreements as well as (ii) entirety of whatever agreements are produced- without any redactions- appears to tie neatly to another judgement (this time of a single judge) of the Delhi High Court in relation to constitution of confidentiality clubs (Nokia v. Acer, Asus, Hisense). The single judge had reiterated that in-house members of the defendants must be allowed access to the third-party license agreements that the patentee produces and that terms relevant for FRAND determination cannot be redacted. More importantly, since comparability is something that only the defendant can determine- and assist the Court in determining, the single judge went on to hold that,

“47. In the considered opinion of this Court, the Defendant’s prayer that Plaintiff be directed to provide inspection of ‘all’ comparable licenses is reasonable. The intent of offering inspection of the comparable licenses to a defendant in SEP suits is to enable the said defendant to satisfy itself that the rate offered is FRAND and this can only be done if ‘all’ comparable licenses are filed by the plaintiff. The Plaintiff plea that this is an oral prayer made by the Defendants and ought to be reserved for another round of adjudication fails to persuade this Court. This plea is relatable to the reliefs sought in the application and is therefore entertained. The Plaintiff is accordingly directed to disclose ‘all’ comparable licenses in the Confidentiality Club” And that 

“… The Plaintiff is not permitted to exercise discretion in selecting comparable licenses, and full disclosure of all comparable licenses shall be made as per the directions mentioned at paragraph 47 of this Order”.

Coming back to the DB decision, in context of the “Non-Discrimination” limb of FRAND, the DB noted that discrimination can arise if unequals are treated as equals and not just when the equals are treated unequally. It went on to say that the Courts must examine “the circumstances in which different rates are offered to different licensees, and whether that, by itself, would make the rates discriminatory” and that the words FAIR, REASONABLE, and NON-DISCRIMINATORY “form a kind of combined package” and the royalty rates must be “commercially fair” and “non-exploitative” (para 18).

The Court did give a note of caution about the informational asymmetries that exist in the SEP/FRAND space- “But then again, creation of a confidentiality club may only dispel, to an extent, misgivings harboured by either party regarding leakage of its confidential information, it still does not mean that all requisite information, so as to enable the defendant- or even the Court- to arrive at an informed decision that the offer of the plaintiff is FRAND is provided, even to the members of the confidentiality club” (para 21).

The Court also noted that “mere facial comparison of agreements may be insufficient to arrive at an informed conclusion that the rates” are FRAND and that the plaintiff ought to disclose the circumstances in which the agreements were entered into- these being “a relevant consideration” and the Court must be convinced that the agreements “are at a arms’ length, and negotiated on purely commercial considerations”.

For the uninitiated, agreements could be signed under a variety of circumstances. For example, patentees could go after smaller or inexperienced players to execute licenses on supra-FRAND terms and then use those as benchmarks. Equally, patentees may sometimes execute cross-licenses or give heavy discounts – again such agreements will represent a less-than-FRAND rate.

Perhaps, and stepping away from what may or may not be done in a litigation, what is really required in the SEP/FRAND space is transparency and honesty on part of both patentees and defendants. This may help us reach as close as possible to the ideal, even if the ideal may never be reached.

Interestingly, Philips did not place on record a single license agreement despite admittedly being in possession of the same. The DB, therefore, noted that there was no material on record before the single judge for arriving at the conclusion relating to damages or the finding that the royalty rates claimed by Philips from the Bansals were FRAND. The DB also opined that informal negotiations between parties can never constitute the basis for final adjudication of rates that may be payable by a defendant and such adjudication would require an objective assessment of agreements executed by the plaintiff-SEP holder with other licensees. The Court also found that the observation of the single judge that no evidence had been led by the defendants to rebut the evidence of Philips, was also unsustainable, as in fact, no evidence was led by Philips to support its assertion that the royalty rates extended to the defendants were FRAND.

A more detailed piece on how the Court dealt with FRAND will follow from another author.

Click here to view this post on SpicyIP and leave a comment.