Interdigital Technology Corp v Lenovo Group Ltd [2021] EWHC 3192 (Pat)

[2021] EWHC 3192 (Pat)Case No HP-2019-000032
IN THE HIGH COURT OF JUSTICE
CHANCERY DIVISION
PATENTS COURT
Venue Royal Courts of Justice, Rolls Building, 7 Rolls Buildings, Fetter Lane
London EC4A 1NL
Date Date: 26/11/2021
Before
THE HON MR JUSTICE MELLOR
Between
INTERDIGITAL TECHNOLOGY CORPORATION ClaimantsINTERDIGITAL PATENT HOLDINGS, INC.INTERDIGITAL, INC.INTERDIGITAL HOLDINGS, INC.LENOVO GROUP LIMITED DefendantsLENOVO (UNITED STATES) INC.LENOVO TECHNOLOGY (UNITED KINGDOM) LIMITEDMOTOROLA MOBILITY LLCMOTOROLA MOBILITY UK LIMITED
Mark Chacksfield QC, Thomas Jones and Edmund Eustace (instructed by Gowling WLG) for ClaimantsDaniel Alexander QC, James Segan QC, Ravi Mehta and William Duncan (instructed by Kirkland & Ellis International LLP) for DefendantsHearing Hearing date: 25 th November 2021
Approved JudgmentI direct that pursuant to CPR PD 39A para 6.1 no official shorthand note shall be taken of this Judgment and that copies of this version as handed down may be treated as authentic.COVID-19: This judgment was handed down remotely by circulation to the parties’ representatives by email. It will also be released for publication on BAILII and other websites. The date and time for hand-down is deemed to be Friday 26 th November 2021 at 4pm..............................THE HON MR JUSTICE MELLOR

Introduction

[1]The Claimants (“IDC”) are members of a group which owns a substantial portfolio of patents covering 3G, 4G and 5G technologies in the mobile device sphere. By this litigation it seeks to make the Defendant companies (who I will call Lenovo, but which cover the Lenovo and Motorola brands) enter into a global FRAND licence.[2]Technical trials A and B have taken place. In Trial A the IDC patent in question was found valid, essential and infringed by HHJ Hacon. Judgment in Trial B is still awaited (from me). The Non-Technical (FRAND) trial is set to start at the beginning of term in January 2022. Very broadly, the trial will examine whether a licence offer (or offers) made by IDC is FRAND and if not, what terms are FRAND. In the usual way, the parties will seek to establish what they say are the FRAND terms through two routes: one is by examining what are alleged to be comparable licences and the second (often used as a cross-check) is via what is known as a top-down analysis.[3]This judgment arises from argument at the PTR ahead of the FRAND trial and it concerns the confidentiality of certain patent licence agreements and information derived from them which will be used in the comparables part of the case. When it comes to the detail, there are three distinct parts to the application.[4]These confidentiality issues matter to the parties but also to the counterparties to the various agreements. For this reason I did not give judgment immediately. As I said during argument, I wished to give further consideration to the contentions put forward by certain counterparties, and to reflect on the arguments I received from the parties.[5]As a well-established licensor, IDC disclosed some 70 Patent Licence Agreements (‘PLAs’) as part of its disclosure for the FRAND trial. Of those, 26 are relied upon by the parties as comparables and 8 others are referred to in the evidence. It is not in dispute that the terms of the PLAs (and in particular the financial terms) are highly confidential, not just to IDC but also to each of the counterparties to those agreements.[6]In July 2020, Birss J made an Order by consent (“the July CRO”) which established a threetier confidentiality regime comprising(1) a general confidentiality tier to protect ‘General Confidential Information’ or ‘GCI’(2) an Approved Persons only tier (‘AP’) and(3) what is called Lawyers Eyes Only (‘LEO’) but which includes external non-legal experts. The LEO tier excludes any in-house employees of the parties but such persons can be included in the Approved Persons tier. As matters currently stand, IDC have 2 AP representatives and Lenovo have 4, but these include personnel involved in licensing negotiations.[7]All the PLAs disclosed by IDC include information currently designated LEO, although there is also a specific fourth tier for a recently signed PLA. I am told there is a very limited amount of Lenovo LEO material in the case.[8]Thus far, the parties and their advisers have managed to progress the litigation under the terms of the July 2020 CRO and, indeed, the evidence for trial is very nearly complete. Under the current LEO regime, not only can no-one at Lenovo see an unredacted copy of any of the PLAs relied on as comparables, substantial redaction of the expert reports has been required to provide copies for Lenovo personnel to read. The redactions are such that Lenovo personnel do not currently have access to the evidence on comparables which will form an important, if not critical part of the case at trial. Furthermore, Lenovo has recently served a statement of case in draft as to the lump sum which it wishes to say should be paid under a FRAND licence. The document remains in draft because Lenovo is not at present able to give instructions as to what the figure for the lump sum should be, because certain inputs which will be used to calculate that lump sum figure cannot presently be disclosed to any Lenovo personnel.[9]At a short CMC on 25 October 2021 I was invited to make an Order by consent which inter alia required IDC to serve redesignated (in terms of confidentiality) versions of its evidence and pleadings. It did so on 19 November 2021. Lenovo estimate that IDC served over 1300 pages of redesignated material, significant parts of which contain information alleged to be confidential at the various tiers I have outlined.[10]In the meantime, on 4 November 2021, and anticipating that the redesignation exercise would not achieve what Lenovo required, its solicitors wrote suggesting that the LEO tier should simply be removed. After further correspondence, the proposal has been honed somewhat and the parties have made significant progress in seeking to agree a new confidentiality tier, intermediate between LEO and Approved Persons. This is embodied in a proposed new Confidentiality Protocol in which the parties have agreed (in broad outline): 10.1. There should be a limited number of disclosees, so called Specified Persons, who should be named with the counterparties having the right to object to them; 10.2. There should be a licensing bar in place i.e. the disclosees should not be involved in any licensing negotiations for a specified period.[11]There remain a series of points in dispute as to the details of the Protocol, and it is fair to say that these points largely derive from the regime which I established recently in Godo IP Bridge 1 v Huawei Technologies[2021] EWHC 2826 Pat (“IP Bridge”). There are a number of significant differences between the circumstances I had to consider in IP Bridge and those in this case, not least that the litigation in IP Bridge was still at a relatively early stage and this case is very close to the FRAND trial. Nonetheless, it almost goes without saying that each case turns on its own facts.[12]One of the consequences of Lenovo’s suggestion on 4th November 2021 to dispense with the LEO tier altogether was that IDC had to give notice of the proposed change to all the counterparties to the PLAs which had been disclosed. Shortly before the PTR hearing commenced I received representations in varying levels of detail from a total of 14 such counterparties. Some made their representations via a solicitors’ letter. Other representations were summarised in a letter to me from IDC’s solicitors. Of the 14, 2 raised no objection, 1 expressed no positive view but the remainder made strong representations that the confidentiality of their agreements should be preserved. I will have to return to discuss some of these representations below.

Applicable principles

[13]The applicable principles are not in dispute. Lenovo reminded me in their skeleton of a number of the seminal cases in this area, Al-Rawi (Al-Rawi v Security Service [2012] 1 AC 531) in particular. Mr Segan QC (who argued these issues for Lenovo) rightly emphasises the two features of a common law trial which the Supreme Court identified as fundamental to our system of justice – the open justice principle and the principle of natural justice. All the cases cited by Lenovo were reviewed by the Court of Appeal relatively recently in Oneplus v Mitsubishi [2020] EWCA 1562, [2021] FSR 13. Floyd LJ summarised the principles at [39] as follows:
‘39 Drawing all this together, I would identify the following non-exhaustive list of points of importance from the authorities: (i). In managing the disclosure of highly confidential information in intellectual property litigation, the court must balance the interests of the receiving party in having the fullest possible access to relevant documents against the interests of the disclosing party, or third parties, in the preservation of their confidential commercial and technical information: Warner-Lambert[1975] R.P.C. 354 at p.356; Roussel[1990] R.P.C. 45 at p.49. (ii). An arrangement under which an officer or employee of the receiving party gains no access at all to documents of importance at trial will be exceptionally rare, if indeed it can happen at all: Warner-Lambert[1975] R.P.C. 354 at p.360; Al-Rawi[2011] UKSC 34 at [64]. (iii). There is no universal form of order suitable for use in every case, or even at every stage of the same case: Warner-Lambert[1975] R.P.C. 354 at p.358; Al-Rawi[2011] UKSC 34 at [64]; IPCom 1 at [31(ii)]. (iv). The court must be alert to the fact that restricting disclosure to external eyes only at any stage is exceptional: Roussel[1990] R.P.C. 45, p.49; Infederation at [42]. (v). If an external eyes only tier is created for initial disclosure, the court should remember that the onus remains on the disclosing party throughout to justify that designation for the documents so designated: TQ Delta[2018] EWHC 1515 (Ch) at [21] and [23]; (vi). Different types of information may require different degrees of protection, according to their value and potential for misuse. The protection to be afforded to a secret process may be greater than the protection to be afforded to commercial licences where the potential for misuse is less obvious: compare Warner-Lambert[1975] R.P.C. 354 and IPCom 1; see IPCom 2 at [47]. (vii). Difficulties of policing misuse are also relevant: Warner-Lambert[1975] R.P.C. 354 at p.360; Roussel[1990] R.P.C. 45 at pp.51–52. (viii). The extent to which a party may be expected to contribute to the case based on a document is relevant: Warner-Lambert[1975] R.P.C. 354 at p.360. (ix). The role which the documents will play in the action is also a material consideration: Roussel[1990] R.P.C. 45 at p.49; IPCom 1 at [31(ii)]; (x). The structure and organisation of the receiving party is a factor which feeds into the way the confidential information has to be handled: IPCom 1 at [33]. 40 To this I would add that the court must be alert to the misuse of the opportunity to designate documents as confidential. It remains the case that parties should not designate such material as AEO, even initially, unless they have satisfied themselves that there are solid grounds for establishing that restricting them in that way is necessary to protect their confidential content.’
[14]As his summary indicates, Floyd LJ analysed a variety of previous cases in the paragraphs leading up to [39] and [40], and a number of those judgments were given in FRAND cases. I will not set out the paragraphs in this judgment, but I have had particular regard to paragraphs [25]-[27] regarding Al-Rawi, [29]-[31] regarding IP Com 1 and [33]-[36] regarding TQ Delta. Naturally, in the summary, paragraphs (i) and (ii) have particular force in the present circumstances.

Application to the facts

[15]As I mentioned, there are three distinct parts to this application: 15.1. The first and perhaps the most important concerns whether I should establish yet another tier of confidentiality, but one where Specified Persons at Lenovo are able to see all the financial information which is currently LEO only. 15.2. The second concerns the status of certain anonymised data, in which the identities of IDC’s licensees are concealed, but a full range of unpacked royalty rates are presented. 15.3. The third concerns three particular extracts from conclusions set out in the Expert Report of one of Lenovo’s experts, Mr Meyer. Lenovo suggest that the information in these extracts is not confidential at all.[16]Before coming to the points debated between IDC and Lenovo, I need to mention certain overriding points and points taken by various of the counterparties, which apply to all three parts of the application. It is fair to say that in general terms, the vast majority of the counterparties who made representations object strongly to any relaxation of the protection afforded to their licence agreements and, in particular to the financial details in them. It is clear that they regard the financial details as highly confidential and it is difficult to disagree. I proceed on that basis.[17]Nonetheless, many of the counterparties take a pragmatic view (recognising that some information has to be made available to Lenovo so it can participate in the FRAND trial) but have put forward a variety of suggestions including the following: 17.1. Whilst accepting that Lenovo must be able to see the end product of the analysis of the comparables, one counterparty suggested that Lenovo did not need to see the underlying licence agreements or their specific financial details. Lenovo’s representatives could be provided with abstracted or anonymised data representing the rates that can be derived from the various comparables. It was suggested that in several cases brought by Philips, the in-house representatives for the Defendants received only anonymised, abstracted details of the comparable licences in the case, without needing to see the underlying documents themselves. 17.2. One counterparty had agreed, back in July 2020 and at a time when a small number of licences were said to be involved, to the disclosure of its royalty rate at the AP level. Now that a large number of licences are involved, this counterparty objects to the Approved Persons having unrestricted access to all the rates of each counterparty. It says that Lenovo will gain an unfair competitive advantage by having access to the full range of IDC’s confidential SEP rates. It points out that the basis of its concerns were recognised by Sir Alastair Norris in Sisvel v Xiaomi [2020] EWHC 2641 at [49]: “Having failed to agree a licence in arm’s length negotiations with the Claimants or certain other counterparties they are now to be provided with a wide range of highly sensitive confidential information about the Claimants’ and those other parties’ pricing structures. Such a dynamic creates a perverse disincentive to reach an arm’s length settlement and a perverse encouragement of FRAND litigation.” 17.3. The same counterparty also warns of the risk of third parties being able to put “2 and 2 together” from pieces of information made public. It concludes by suggesting that if any re-designation is made, it should be done in accordance with [14]-[16] of IP Bridge. 17.4. Finally, I need to keep in mind that at least some of IDC’s counterparties may have licences with Lenovo or may seek to negotiate such licences in the future or, indeed, they may have such negotiations ongoing at the moment. I am reminded that I need to be astute to avoid Lenovo obtaining an unfair advantage in any such negotiations via disclosure in this action, albeit those concerns need to be balanced against the requirements for Lenovo to be able to have meaningful participation in this important FRAND trial.[18]With those principles in mind, I turn to consider the points in dispute.

The proposed Specified Persons tier

[19]IDC accept the proposed Specified Persons regime in principle. I have no doubt IDC were right to do that in the light of the applicable principles and the authorities. This new tier is essential to allow Lenovo (via their Specified Persons) to be able to participate in a meaningful way in the lead up to and in the FRAND trial. None of the counterparty representations persuade me otherwise.[20]The points in dispute concern details as to the implementation of this new tier. I can deal with them relatively shortly.[21]First, the number of disclosees – IDC suggest one is sufficient, Lenovo say two. Lenovo have identified their two nominees: one is a General Counsel in the Intelligent Devices Division of Lenovo and the other is in a finance role. I have no doubt that Lenovo should have these two individuals as their Specified Persons.[22]Second, the length of the licensing bar, IDC suggest 5 years, Lenovo say two is sufficient. At least two of the counterparty representations contended for a 5-year licensing bar. The dispute is a slightly odd one, in view of the fact that the two Lenovo Specified Persons do not, as I understand it, have involvement in licensing negotiations and evidently do not intend to for at least 2 years. In these circumstances it is difficult to see therefore what prejudice Lenovo would suffer if the licensing bar on these two individuals is 5 years, particularly in view of the size of its organisation. As presently advised, I consider there is good reason for the licensing bar to be 5 years in view of the range of the detailed and commercially valuable information which will be available to them in this tier. As in IP Bridge, this will be the position pending any further order.[23]Third, IDC suggested a ‘read-only restriction’ of the type I imposed in IP Bridge. Lenovo resist this. Lenovo’s 2 Specified Persons are based in the US but the current intention is that they will come to London and attend the trial, other things being equal.[24]I have no doubt that a ‘read-only restriction’ is not warranted in the present circumstances. Both Specified Persons will have given extensive undertakings. To require them to attend a Kirkland & Ellis office or to have some supervised remote access would be unduly burdensome in the lead up to and during the trial.[25]The fourth point concerns the 35 or so PLAs which are not relied upon in the expert evidence. IDC say there is no reason why they should not remain LEO and I am inclined to agree. For its part, Lenovo suggested a different and broader solution, whereby any materials currently designated LEO which are not included in the agreed trial bundles can retain that LEO designation, whereas any LEO materials included in the agreed trial bundles will be subject to the Specified Persons tier of protection. Although I do not know what other LEO material there is in the case other than the PLAs, it seems the trial bundle proposal is a practical division, provided that LEO materials are not being included in the trial bundles unnecessarily. If one party suspects that to be happening, I may be asked to rule on the point, but I sincerely hope that will not be necessary. I understand that there will be a considerable number of files in the trial bundle in this case and the parties should work to reduce the volume of material not increase it.[26]The fifth and final point concerns when this new regime should come into force. On the one hand, Lenovo require the Specified Persons regime to come into force as soon as possible. On the other hand, the counterparties should be given time to register any objections to the regime I have decided upon plus the other rulings on confidentiality set out later in this judgment. Since the counterparties are already on notice of the proposed changes (even if not apprised of the details), I consider it is sufficient to give them until 4.30pm on Thursday 2nd December 2021 to make any further representations in response to this judgment. If no further representations are received, then the Specified Persons regime will come into effect on 3rd December 2021. If any representations are made, they should be forwarded to me without delay and I will rule on them as quickly as I can.

The Anonymised Schedule 2

[27]In the anonymised Schedule 2 document, the identities of the licensees to the 26 comparables have been stripped out, being replaced by numbers 1-26. The table includes Lenovo’s analysis of the effective per unit rate which Lenovo says is calculated by Mr Bezant compared with Mr Meyer for each of the anonymised licensees, along with a ‘total IDC licences weighted average’ figure and sub-total weighted averages for the 7 comparables pleaded by Lenovo and for the 19 or so comparables relied upon by IDC. These unit rates are all unpacked rates, not the actual rates in the licences in question.[28]Lenovo suggest that this anonymised Schedule 2 should be designated non-confidential. It contends that exactly this sort of information was published by Birss J in the public version of his judgment from the FRAND trial in Unwired Planet and by Judge Selna in TCL v Ericsson. As regards Unwired Planet, I note that at [464] Birss J published an overall % figure as representative of the value of Ericsson’s 4G SEP portfolio but no range of figures which led to that. Furthermore, Mr Segan QC contends that if this type of information retains any confidentiality protection, the conduct of the FRAND trial will be extremely difficult.[29]For their part, IDC suggest this anonymised Schedule 2 document still contains valuable commercially sensitive information but contend that the General Confidential Information category is the appropriate level of protection.[30]At this stage, I am inclined to agree that this table should remain as General Confidential Information. I remain of the view that even though the rates are unpacked, this array of information could provide advantage to Lenovo and to third parties if it is made public without restriction. I do not rule out the possibility that this type of information may well be included in a public judgment in due course, but the position will be reviewed as the FRAND trial progresses and the relevance of information becomes clearer. Finally, I am not convinced that GCI protection will render conduct of the trial overly difficult. Counsel are well able to make submissions and to cross-examine by indicating the figure they are talking about without reading it out in public. As a general point, I take a cautious approach at this stage. When I have read in more fully and have a better grip on all the issues, I may well revise my views on confidentiality.[31]However, the weighted average figures in Schedule 2 are somewhat different. I agree that these three rows can be designated non-confidential, not least because (see below) I am agreeing to Mr Meyer’s weighted average per-unit rates being re-designated nonconfidential and I see no reason why Lenovo’s analysis of Mr Bezant’s competing figures should be treated any differently.

The extracts from Mr Meyer’s Report

[32]As indicated above, the three particular extracts are all from the conclusions at the end of Mr Meyer’s Report. Paragraph 252 gives the results in terms of a range of fixed fee royalty amounts and his conclusion as to a comparison with his chosen comparables. It is proposed that a redacted version of his Table 49 should be prepared which will show just two weighted average per-unit rates, with the identities of the licensees in that table and their effective per-unit rates redacted. Finally, Figure 8 compares Lenovo’s analysis of the current offers with the outcomes of two slightly different analyses he has conducted on his comparables.[33]If Lenovo are content that no confidentiality should attach to these extracts, then I consider it is appropriate to re-designate them as non-confidential. These extracts really define part of the battle lines which will be in issue at trial and it is difficult to see how any public judgment from the FRAND trial could be meaningful without this type of information being included. Furthermore, I find it difficult to see how any reverse engineering of such information could provide any information which would be valuable or useful.

Two final points

[34]First, IDC asked me to make clear that each of the figures discussed in [27]-[33] above represent Lenovo’s analysis of the underlying materials, analyses that remain disputed by IDC. At this point I am not expressing any view as to their correctness – that is a matter for the FRAND trial. I am content to include this clarification in case any of the figures discussed in [31] and [32] are made public before the trial. If any of those figures are made public in advance of the trial, they must be accurately described. I would hope that the parties would, in the usual way, refrain from partial or selective publication in advance of this trial.[35]Second, Lenovo suggested that the parties seek to agree ways in which other protected information in the case can be anonymised to preserve the confidentiality of the information, whilst enabling discussion to take place in open court. IDC agree this is a sensible suggestion and so do I. Notwithstanding the caution apparent at certain points in this judgment, my aim is to conduct as much of this FRAND trial in public as possible. The Court will sit in private provided I am persuaded that it is essential to do so. If I detect that unreasonable claims to confidence are being maintained, I will rule accordingly.[36]I ask Counsel to agree the draft Order to give effect to this Judgment.