Interdigital Technology Corporation & Ors v Lenovo Group Limited & Ors

This judgment contains my decision on the confidentiality issues left outstanding following my FRAND Judgment (in its Confidential version [2023] EWHC 538 (Pat) and in the public version [2023] EWHC 539 (Pat) but generally ‘the FRAND Judgment’ or ‘the Main Judgment’), i.e. whether I should remove some of the redactions in the current public version of the FRAND Judgment. I heard argument on the confidentiality issues at a hearing just before Easter on 5^th April.

As appears from the title page, I received both written and oral submissions from InterDigital and Lenovo as well as from the several interested third parties there listed. In addition, I received written submissions on behalf of Huawei, Innovius, PA Consulting, Acer, ZTE, Samsung, Xiaomi, Doro, Fairphone and a third-party licensee of InterDigital (‘AB’) who has legitimate reasons for remaining anonymous. Although InterDigital notified all those with an interest in information in the FRAND Judgment designated as confidential, some (about 10) did not respond, including LG and a number of the smaller licensees.

I have carefully considered all the submissions made to me but since many were similar if not identical, I can group the protagonists into the following categories:

To assist my deliberations, the redactions were divided into various categories. First, Mr Mehta produced a helpful table summarising the positions taken by each interested party (i.e. each licensee and data provider). Second, in their evidence and submissions, InterDigital used the following 8 categories, some of which I have supplemented by identifying some of the key passages in dispute:

Third, in their evidence and submissions, Lenovo addressed the following four categories:

I have not found it necessary to adopt any of these categorisations in their entirety. All were useful at highlighting particular issues. As I said during the hearing, I found the submissions made to me by the counterparty licensees the most illuminating and I am very grateful for their assistance.

In terms of the applicable principles, it is safe to say in oral submissions there was universal agreement that the applicable principles are those set out by Birss J. in his judgment on confidentiality in UnwiredPlanet v Huawei [2017] EWHC 3083 (Pat) at [23]-[24] (‘UnwiredPlanet Confidentiality’). Some reference was also made to earlier case law, including R (Mohamed) v Secretary of State for Foreign and Commonwealth Affairs (No.2) [2011] QB 218, but all earlier cases were considered by Birss J. and led him to derive the principles he said were applicable to ‘a case like this’ in [23]-[24] which I set out here:

‘23. Unless the public can see and understand a judge's reasons they cannot hold the courts to account. There is therefore a strong principle that all parts of a judgment should normally be publicly available. Nevertheless there are occasions on which judgments may be redacted. Redactions will require powerful reasons, supported by cogent evidence which addresses the details. Generalities will not do. Although redactions will be rare indeed when looking across the legal system in general, certain kinds of proceedings may regularly involve redactions due to the nature of the proceedings and the material involved. In any event however redactions must be kept to the bare minimum.

24.

Factors which will be relevant include:

i)

the nature of the information itself: for example cases in which some redaction may more readily be accepted could include technical trade secrets and private information about family life.

ii)

the effect of the publication of the information. This will be a critical factor. If publication would be truly against the public interest then no doubt the information should be redacted. If publication would destroy the subject matter of the proceedings - such as a technical trade secret - then redaction may be justified. The effect on competition and competitiveness could be a factor but will need to be examined critically.

iii)

the nature of the proceedings: for example privacy injunctions and competition law claims may require some redaction while an intellectual property damages claim may not. The point is not that different kinds of case demand a different approach, it is that the balance of factors will change in different cases (e.g. the need to encourage leniency applications in competition law).

iv)

the relationship between the information in issue and the judgment (as well as the proceedings as a whole). Obviously judges do not deliberately insert irrelevant information into judgments but not every word of a judgment is as important as every other word. It may be that some sensitive information can be redacted without seriously undermining the public's understanding of the reasons.

v)

the relationship between the person seeking to restrain publication of the information and the proceedings themselves (including the judgment). For example, a patentee seeking damages for patent infringement on a lost profit basis knows that they will have to disclose their profit margin in the proceedings and that those proceedings are public. A third party whose only relationship with the case is that they are a party to a contract disclosed by one of the parties to the litigation is in a different position.’

I agree entirely that those principles apply to this case and I propose to apply them to the facts. As Ms Love, for Apple, said in her oral submissions, the Court is engaged in a balancing exercise that is specific to the circumstances, specific to the case, specific to the information and specific to the position of the party whose information is under examination.

In several of the submissions from those in the first group, the danger of reverse engineering was emphasised. In particular, it was stressed that whatever was made public in my non-confidential FRAND Judgment would be analysed and studied in great detail, with links being made between different pieces of information to draw inferences as to royalty rates, terms and, more generally, InterDigital’s licensing practices. I keep this point in mind.

If, as was hinted in Lenovo’s skeleton argument, a different approach to confidentiality is required in cases of this type, that will have to be determined in a higher court. As I have previously commented, the SEP licensing market would operate much more efficiently with greater transparency. However, the reality must be recognised that virtually all the players in this market, whether licensors or licensees, seek to maintain confidentiality over their licence terms, and particularly the financial terms. It must also be recognised that, in order to decide this type of case, the Court benefits from the assistance and co-operation of third parties allowing their confidential information to be used, even when the Court exercises its powers to require/order disclosure from the parties.

This was indicated in my FRAND Judgment at [2], where I said the following:

‘2. There are two versions of this Judgment: the full version [2023] EWHC 538 (Pat), which comprises 225 pages, is available only to the parties and those in the appropriate level of the confidentiality regime because it contains a considerable amount of information confidential to one of the parties and/or third parties. I have received the parties’ contentions and submissions on what should be redacted from the full version to protect information said to be confidential either to InterDigital or one of their licensees. In line with the approach taken by Birss J. to redactions to his main judgment in UnwiredPlanet: see [2017] EWHC 3083 (Pat) (‘the Confidentiality Judgment’), I have taken a generous view of the claims to confidentiality at this initial stage, pending further evidence and representations. Having said that I have not accepted every claim to confidentiality because I must ensure that the public version fairly sets out the important parts of my reasoning. A version containing an initial set of redactions [2023] EWHC 539 (Pat) is available for general publication. It is highly likely that a further public version of this judgment containing fewer redactions will be published in the future and I aim to issue that version as soon as possible. In this regard, I am currently engaged in almost exactly the process described by Birss J. at [1]-[4] of his Confidentiality Judgment. I will appoint a further hearing at which confidentiality issues will be finally resolved, in which the parties and interested third parties can participate. This will take place shortly and before this term ends on 5th April 2023.’

I should point out that I received several rounds of submissions from the parties, InterDigital in particular, on what they contended needed to be redacted in the public version of the Judgment, before I issued the public version. With the benefit of hindsight, I can see that, at that first stage, I took a rather firmer view than perhaps Birss J. did of what information was necessary to include in the public judgment so that it fairly set out the important parts of my reasoning. The consequence of my initial firmer approach appears later.

Following the hand down of both versions of my FRAND Judgment, InterDigital’s solicitors took on the burden of notifying all the possible third parties who might wish to make representations on these confidentiality issues, and marshalling their responses. I am very grateful for all that work which was done under some time pressure.

Before turning to the principal issues, it is convenient to discuss the positions taken by Doro and Fairphone.

As I indicated, Doro and Fairphone were content for their effective per unit rates to be disclosed. However, their rates were included in general summaries - in Mr Bezant’s plots in [312], in the collection of 4G rates in [319 i)] and in the table in [577] which included effective per unit rates for 15 of the InterDigital 20. Aside from the point that the running royalty rates of the smaller licensees were generally much higher than the rates unpacked from the much larger lump sum licensees, the particular rates paid by Doro and Fairphone are not central to the reasoning in my FRAND Judgment. Doro’s effective per unit rate of $0.13 was, like that of Blu, an outlier (hence there was a small error in [587]). Fairphone’s effective per unit rate of $0.69 was more representative of the general point, although there was significant variation – the largest effective per unit rate amongst those 15 licensees was $2.24.

That leaves the parties in the first, third (i.e. Lenovo) and fourth groups. Since those parties in the fourth group made no representations but are not to be taken to have waived their rights to maintain confidence, I can proceed to determine the issues between those in the first group (who, with only some very minor exceptions, ask that the current redactions are maintained and therefore can be taken to represent those in the fourth group) and Lenovo.

Perhaps the most attention was focussed on the status of the bar chart X1 which features in the Confidential FRAND Judgment at [586]. Various combinations of redaction of parts of the bar chart were proposed. However, I reached the clear conclusion that even with the axes, all the figures and the identities of the Lenovo 7 redacted, the combination of the remaining diagram and other information in the public FRAND Judgment could and would be used to work out the unpacked royalty rates for all the Lenovo 7. The various proposals for redaction highlighted just how much information is portrayed in that striking diagram. I have therefore concluded it is not possible to reveal the diagram in the public judgment.

The issue does not end there. Even though I have decided I cannot make even a redacted version of the diagram public, I consider I can and should give a description of it, beyond what I said in [587] of the Public FRAND Judgment:

587.

Subject to an outlier (Blu 2020) representing the single low running royalty indicated at $[REDACTED], every other PLA in the InterDigital 20 implied royalty rates which were higher and, as indicated above, often significantly higher than the rates implied from the Lenovo 7.

X1 was aptly characterised by Mr Segan KC for Lenovo as the ‘fjord’ diagram. To explain: each license was represented by a bar, whose height was the unpacked per unit rate and width represented the volume of units under the license. The Lenovo 7 (all lump sum licenses) were depicted in various shades of blue, and the InterDigital 20 (all running royalty licences) were depicted in red. In the fjord analogy, the Lenovo 7 represented a depiction of the (more or less slightly undulating) sea and the InterDigital 20 represented a depiction of the side of a very steep fjord, the cliff going up in steps. These steps were very narrow because the volumes in the RR licenses were small. X1 remains a powerful illustration of the difference between the rates unpacked from the large lump sum licensees and the generally considerably higher rates paid by the much smaller licensees under their running royalty licences.

Each of these plots, tables or lists contain a lot of information, albeit in varying amounts. I must address each one individually.

The three plots in [312] indicate the 4G LER, CBR and PDR as calculated by Mr Bezant for each license in the InterDigital 20 or the Lenovo 7. Although Doro and Fairphone were content to have their points in these plots disclosed (see above), they were very much the exception. Even though these are mostly unpacked rates (particularly for the Lenovo 7), I am satisfied that InterDigital and all the other counterparties are entitled to have their rates redacted.

The issue over the tables in [372] and [575] concerns the effective per unit rates presented for each of the Lenovo 7. Lenovo proffered versions of these tables in which the various rates were anonymised and randomised, along with footnotes stating ‘The order of the licensees has been randomised using a random sequence generator so that the ordering no longer accords with the public version of the Judgment dated 16 March 2023’. Lenovo’s offers realistically accepted the confidentiality of even unpacked rates attributable to individual licensors. Lenovo’s principal argument in favour of disclosure of the anonymised rates was that it was necessary, in order to understand my FRAND analysis, to see the range of rates from which I selected LG 2017 as the best comparable.

The problem, even with these figures anonymised and randomised, is that it would be simple to identify the effective rate calculated by Mr Meyer attributed to Apple in view of my comments (already public) at [661] that Apple’s rate represented an ‘upper bound’ and at [797] an ‘outlier’. Even if it is not possible to attribute specific rates to the other licensees in the Lenovo 7, I am not persuaded that Apple should suffer this fate.

Furthermore, there are other ways to meet Lenovo’s principal argument which do not require their proffered versions. A sense of the range can be provided by reference to the upper and lower bounds of Mr Bezant’s (not Mr Meyer’s) effective per unit rates (presented by Mr Meyer), which (I emphasise) I did not accept. Mr Bezant had two rates at $0.52, with the lowest rate at $0.11.

The list at [319 i)] contains 4G rates from 14 of the InterDigital 20 set out in chronological order. The table at [577] contains the effective per unit rates calculated by Mr Meyer for 15 of the InterDigital 20. I agree that the list in [319 i)] could well provide a lot of information by comparison with the dates of various licence agreements. I also agree with the submission that there is already sufficient information about the trend which these figures illustrate in the previous paragraph – [318]. For those reasons I decline to remove the redaction of [319 i)].

The table in [577] is different. Lenovo presented me with an anonymised and randomised version of this table which I have concluded should be included in the public judgment. Even though it presents the unpacked per unit rates for 15 of the InterDigital 20, these figures illustrate two important foundations of my FRAND Judgment. First, the gulf between the rate I derived for Lenovo or the rate derived from LG 2017 and the vast majority of these unpacked rates from the smaller licensees. There are two notable exceptions – the rates for Blu and Doro (on which more below). Second, the very wide variation in these 15 rates - the degree of variation being indicative of a licensing policy which aims to extract as much money as is possible from each licensee i.e. without regard to what a FRAND rate is or should be. I consider that the public should understand these two foundations because they illustrate why I rejected InterDigital’s reliance on the InterDigital 20 and why I considered the Lenovo 7 to be a far better group of comparables.

The randomisation of these figures should prevent nearly all of them being attributed to particular licensees. The exceptions are Blu and Doro, for separate reasons. It is not difficult to discern from what I said in [587] that Blu had the lowest unpacked rate, and, as discussed above, Doro are content for their effective rate to be made public. As far as I can recall, the low rates charged to Blu and Doro were not the subject of any scrutiny or explanation at trial.

Although Blu did not respond to the invitation to make representations, I do not think I can proceed on the basis that they have no objection to disclosure of their effective per unit rate. Accordingly, I gave careful consideration to whether I should disclose the randomised figures in the table at [577], in the knowledge that it would identify Blu’s rate. Neither PLA is historic, both being relatively recent. However, I concluded disclosure was necessary to enable better public understanding of my FRAND Judgment. I draw some comfort from the thought that Blu (and Doro) might be rather proud of having negotiated rates which were very considerably lower than many others.

I can deal with the remaining issues adopting certain of InterDigital’s categorisations, dealing with them in reverse order.

Category 8 has two parts. The second part concerns the identity of licensee AB and I consider it should remain anonymous. The first part contains five redactions which I made for consistency despite the fact that none were requested by either InterDigital or Lenovo. I have reviewed these in the light of my considerations in this Judgment and will remove them.

InterDigital helpfully gathered certain redactions in their Category 7, where the information in question has already been made public elsewhere in the non-confidential judgment. I have also identified two other examples in the course of preparing this judgment. I will remove those redactions.

Category 6 concerns information/data provided by third parties. The redacted information falls into two broad categories: first, in the table at [161] information about the global top 20 mobile phone sellers over the period 2013-Q2 2021, comprising % market share, total units sold and average annual units sold and second, information from the PA Consulting reports in [829], [831]-[834]. The information in the first category must be readily available (provided it is paid for) and I consider it highly likely that any entity of any substance operating in this field will have this type of information already. Although I have concluded the degree of confidence in this information is relatively low, I do not consider I should unilaterally make it public. There is sufficient information in [162]-[164], extracted from the table, to understand the role that this market share information played in my analysis.

As for the PA information, the information in [829] has a very low degree of confidence and I propose to remove the redaction. The text quoted in [831] is from the PA 3G report and is on the historic side. As I have related in my FOO Judgment, although I did not say this expressly in my FRAND Judgment, this commentary (which I emphasise is largely historic) played a role in my rejection of the top down case. For that reason, on balance I consider it should be made public, to aid understanding of my reasoning and in particular because top-down cases based on patent counting studies seem to me to be reasonably popular, even if only deployed as a cross-check.

I retain the remaining redactions in [832]-[834] because they concern later PA Consulting reports (4G and 5G) and therefore have greater currency.

Category 5 contains a rather disparate set of redactions which I deal with in turn, giving brief reasons:

Category 4 I have already dealt with.

Category 3 (confidential information concerning non-financial terms or which cannot be used to directly ascertain the royalty rate) covers a rather disparate group of 37 redactions:

Category 2 (other confidential information factoring into the calculation of unpacked rates) starts with some redactions concerning RIM 2012. It is convenient to deal with those together with the Category 1 redactions which relate to RIM 2012 (see below).

The remaining redactions in Category 2 (and there are a considerable number of them) all relate to various data concerning the Lenovo 7. I have addressed some of these already. As for the remainder, having reviewed all of these, with two exceptions, I am satisfied that the redactions must remain in place. Most of them are points of detail and there is sufficient information in the public judgment to understand my reasoning without these further details being disclosed.

The first exception concerns some implied discounts which Mr Bezant calculated that Mr Meyer had applied to three of the Lenovo 7. As I said in [411], it is clear that Mr Meyer had not applied these discounts, but they were the product of taking the figures which InterDigital had ‘recognised’ in their annual reports as representing the part of a lump sum payment attributable to past sales. Revealing these figures cannot cause any harm to the 3 licensees. However they illustrate the effect of InterDigital’s approach to past sales, which is an important part of my Judgment, and I consider they should be in the public Judgment.

The second exception concerns RIM 2012 which Mr Meyer used to demonstrate a disparity between his figures for unit sales and total payments and those from Mr Bezant. In [582] I present the figures for RIM 2012 which illustrate the impact. Largely because those figures concern a historic PLA and because of the overall effect in overstating implied effective per unit rates, I consider they should be made public.

As for InterDigital’s Category 1, it includes redaction of actual and unpacked royalty rates from current or recently expired PLAs including in particular the Lenovo 7. These data are more or less current and perhaps the most valuable and commercially sensitive information. I propose to maintain the confidence in those figures.

Category 1 also covers actual royalty rates from some historic PLAs, RIM 2012 in particular.

As well as being historic, the rates in RIM 2012 (as I related in [129]) were the benchmark for InterDigital’s licensing program from 2012 until the reset in 2020. Those rates therefore form an important backdrop to all the licences agreed over that period. I therefore consider the terms of RIM 2012 should be unredacted. Even though two of the rates were 4G rates, they are largely irrelevant to any current 4G rate(s).

There is one point to add relating to the terms of RIM 2012. InterDigital pointed out that the 4GMM rate set out in 130 iv) was wrong. I will therefore issue a corrected version of the Confidential Judgment and the corrected figure will appear in the new version of the public Judgment.

All the licensees who made representations on confidentiality incurred costs in doing so. I should recognise also that whenever PLAs have to be disclosed in this type of FRAND proceeding, third party licensees incur legal costs (in varying amounts) to ensure information they regard as confidential is properly protected in the litigation. As far as I am aware, these third party licensees do not seek payment of those costs in the litigation in question.

Consistent with that, with one exception, none of the third parties who made representations for this confidentiality hearing sought costs, even though some were represented by both solicitors and Counsel at the hearing.

The exception was Apple. On their behalf, Ms Love did not seek all of Apple’s costs even of attending the hearing. Instead she submitted that Lenovo should pay a contribution to Apple’s costs of £30,000, being an estimate of the costs she said had been wasted by what Apple understood to be Lenovo’s rather belligerent initial attitude to confidentiality in contrast to the more reasonable stance taken by Mr Mehta during the hearing.

This request was made for the first time towards the end of the hearing, at which point schedules of Apple’s costs were produced. For understandable reasons, Mr Mehta was rather taken aback by the request, since no notice had been given. Although the actual request made was plainly linked to what Apple regarded as a significant change in Lenovo’s position made at the hearing, the schedules of costs had plainly been prepared in advance.

In these unusual circumstances, I decline to make any order for costs in Apple’s favour. Any request for payment of the costs of a third party at this sort of hearing ought to be flagged clearly in advance.

I recognise that significant costs have been incurred on confidentiality issues in these FRAND proceedings by InterDigital, its licensees and other third party data providers. Lenovo has also incurred significant costs in dealing with these issues, albeit to a lesser extent than InterDigital because none of Lenovo’s confidential information was in play in the FRAND Judgment. A lot of these costs cannot be avoided.

It will be understood that the principal balance I have attempted to strike in my approach to redaction is between the continued protection of commercially sensitive information (in particular the rates and conditions in current or recently expired licences) and ensuring that the public version of my FRAND Judgment sets out the important parts of my reasoning (which may require some details of licences key to the reasoning and/or of historic licences to be made public). The firmer approach I took initially has the consequence that I will remove relatively few redactions from the initial public FRAND Judgment and the new/revised public FRAND Judgment will be issued shortly. Assuming my approach to confidentiality is either not challenged or is upheld, it is to be hoped that the costs of this type of final confidentiality hearing can be considerably reduced in the future, on the basis that the Court can be expected to adopt a similar line to that set out in UnwiredPlanet and in my judgments in these proceedings.