TQ Delta, LLC v Zyxel Communications UK Ltd & Anor

This is a primarily a dispute about a trial date for a RAND trial in a case where I have recently heard, but not yet given judgment, on a technical trial concerning two patents in the patent portfolio owned by the claimant (“TQ Delta”). Trial dates are normally fixed by listing. However, it has become increasingly common for trial dates for FRAND/RAND trials in telecoms patent litigation to be bitterly contested by the parties, at considerable cost.

There has been a very significant amount of evidence served on this application. I have read a total of four witness statements from the parties’ solicitors which collectively amount to about 60 pages, accompanied by long exhibits. This hearing has occupied the best part of a day. At the conclusion of the hearing I gave a detailed judgment. I hope that this will avoid a repeat of this type of application in the future.

The Patents Court is very fortunate that experienced solicitors, such as those who represent the parties in the present case, generally agree procedural aspects and therefore disputes such as the present never reach the courts. Unfortunately, this is not the attitude that has been taken in the present case. There have been numerous procedural disputes in this litigation. The inter-solicitor correspondence, like the witness statements, is voluminous and heated. This is unhelpful to the courts and to the parties as it increases the time and costs of the case. If the parties continue to fail to cooperate, I shall take a much tougher view on costs than I have been asked to do so far. Both sides need to behave sensibly to progress this matter to trial.

The patents in suit in the technical trial are two of TQ Delta’s patents from within a substantial portfolio of patents which are digital subscriber line (“DSL”) related, which it acquired from a company known as Aware Inc. in 2012. DSL technologies are those commonly used to provide fixed line broadband internet to residential and commercial premises. DSL technology is prescribed by internationally recognised technical standards which allow for interoperability between DSL products.

The Defendants (“ZyXEL”) are part of the world wide ZyXEL group of companies which is responsible for manufacturing and selling various types of DSL compliant equipment. TQ Delta contends that the patents in suit are essential to the implementation of certain standards, and that they have been infringed by ZyXEL. ZyXEL denies infringement and claims that the patents are invalid.

TQ Delta complains that it has been trying to license the ZyXEL group on a RAND basis for over four years but without success. It says that it made offers for a global portfolio licence on RAND terms, which have been rejected. The Defendants say that they have made a RAND offer for a global portfolio licence which has also been rejected.

TQ Delta complains that the over-arching strategy of the ZyXEL group is to avoid progressing negotiations and to delay litigation for as long as possible, thereby avoiding any payment to TQ Delta, whilst continuing to infringe TQ Delta's patents by dealing

It is not the purpose of this judgment to express a view on whether the defendants are pursuing a "hold out" strategy, where a potential licensee postpones for as long as possible any payment because it wants to hang on to its money and exhaust the resources and will of the opposite party. That issue may have to be decided on another occasion.

TQ Delta first contacted ZyXEL seeking to license its patents in 2013. It then issued US proceedings, having failed to reach agreement, on 9th December 2013. The UK case was not started July 2017. ZyXEL suggested that that indicated delay. I do not agree. TQ Delta was entitled to progress its case in the US and then to choose other jurisdictions to sue if it wished.

TQ Delta hoped that a trial of the technical issues would be available within 12 months of issue, or shortly thereafter. A timetable was set at the CMC on 21 November 2017. At that time ZyXEL submitted that the RAND issues should be stayed pending the conclusion of the technical trials. The result was that the two patent actions concerning the technical trials were listed to be heard in October 2018 with a RAND trial to follow in January 2019.

On 20 November 2017, ZyXEL issued an application for an anti-suit injunction in the United States. TQ Delta has sought to draw adverse inferences from that application. I draw no such inferences. ZyXEL were entitled to issue an anti-suit injunction. The fact that it was ultimately dismissed is why the technical trial took place. In this connection, the parties’ solicitors have disputed in three witness statements whether ZyXEL's US counsel submitted to the US court that the English courts were troll-friendly. That dispute is irrelevant to the issue that I have to decide. It does not matter what US counsel said at that hearing and none of that evidence should ever have been put in.

Shortly before exchange of evidence in the technical trial, ZyXEL issued an application on 1 June 2018 seeking to adjourn the technical trial to the window for the RAND trial in January 2019 and to adjourn the RAND trial from that window to an unspecified date following judgment on the technical trial. This was because of ZyXEL’s concern that their chosen expert for the technical trial would not be a fit state to give evidence. Had that application succeeded it would have entailed a considerable delay to the RAND trial. However, at that hearing Mr. Purvis QC, representing ZyXEL, made an alternative proposal, which was that both trials could be heard together in January 2019 and I acceded to that proposal.

Statements of case on RAND were then exchanged on 25 July 2018. As a result of various changes to the pleadings, the deadline for expert evidence on the RAND trial was pushed back to 7 December 2018. Then on 14th December ZyXEL applied to strike-out aspects of TQ Delta's expert evidence and made a further application in respect of disclosure, seeking a schedule of privileged documents. Those applications were heard by Arnold J, whose judgment is at [2018] EWHC 3651. The result of the hearing before Arnold J was that TQ Delta was given an election as to whether the

TQ Delta decided to run a full RAND case relying upon various matters which it could not have done had it wished to maintain the January 2019 date for the RAND trial. Mr. Speck QC has drawn my attention to paragraph 51 of the judgment of Arnold J:

Mr. Speck submits, and I accept, that the indication that the RAND trial would be adjourned for the minimum period necessary, enabled TQ Delta to make an informed election. It relied upon that indication. On the other hand, Mr. Nicholson QC for ZyXEL has quite rightly pointed out that I am not bound by what Arnold J said; indeed, I am required to consider, when re-fixing the date, potential prejudice to both parties and to ensure that each party has sufficient time to prepare for the trial. I need to take into account the evidence of Ms. Bould (on behalf of ZyXEL) in her tenth witness statement as to alleged prejudice to ZyXEL if the trial is fixed before April 2020, as well as the evidence of Ms. Brodie (on behalf of TQ Delta) in her tenth witness statement as to alleged prejudice to TQ Delta if the trial is delayed beyond September 2019.

Since TQ Delta made its election there has been little cooperation in listing the RAND trial. TQ Delta drew attention to the following positions taken by ZyXEL in intersolicitor correspondence.

ZyXEL’s immediate response to the election was to suggest that all directions in the RAND case should be stayed. On 4 January 2019, TQ Delta wrote to ZyXEL inviting it to provide details of which counsel and experts would be available for trial following a request via counsels' clerks that had not elicited the information. The response was that neither Mr. Purvis nor Mr. Nicholson had availability for a ten-day trial in 2019 and they would check their experts' availability for 2020. On 10 January, ZyXEL indicated in correspondence that it had “downed tools” on RAND. Further correspondence ensued, wherein ZyXEL refused to consider listing of the RAND trial.

On this application, TQ Delta has proposed various dates between April and September 2019 for the RAND trial, whereas ZyXEL assert that the RAND trial must not be heard until April 2020 at the earliest.

In order to arrive at a realistic estimate for the length of the RAND trial, I shall outline each party's case. I should add that I am considering this application before having delivered judgment the technical trials and therefore there is a degree of speculation about the result. TQ Delta does not wish me to indicate what the result will be, with reasons to follow, although I offered to do so in case it would be helpful to the parties. I will assume for the purposes of this judgment that at least one of the two patents in

TQ Delta's pleaded case on RAND is to adopt a comparable licence approach. It relies upon two licences which it alleges to be comparable: first, a licence known as the Zhone licence, which was pleaded in July 2018; and secondly, a licence which is described as “the confidential December licence”, a licence which was entered into very shortly before the hearing at the pre-trial review before Arnold J. For present purposes, I am prepared to treat that licence as confidential.

ZyXEL's case is based upon what is commonly known as a “profits available” or

“economic benefits” approach. It contends that the terms of the licence for one or more SEPs should reflect the economic value of the patented technology. It argues that the licence fee should be determined by assessing a share of the profits that it alleges are currently made by ZyXEL from the manufacture of the DSL equipment in issue, then apportioning a fraction of that to DSL functionality, then apportioning that fraction as representing TQ Delta's share of the overall patent pool essential to the DSL standards in issue.

So, each side's approach to the case is very different, not just in detail but in principle. On the other hand, to those familiar with litigation concern patent licences of right, or references to the Copyright Tribunal, or Wrotham Park damages enquiries, both of these approaches are entirely familiar; they are not that complicated, and the courts are used to dealing with them.

It is worth considering the procedure in the Copyright Tribunal and how it has been transformed over the years. The jurisdiction of the Copyright Tribunal was summarised by Arnold J in Phonographic Performance Limited v The British Hospitality Association and others [2010] EWHC 209. As the learned judge noted, the Copyright

In brief outline, Section 118 provides that a proposed licensing scheme may be referred to the Tribunal by an organisation claiming to be representative of persons claiming that they require licences in cases to which the scheme would apply. Section 119 provides that, if while a licensing scheme is in operation a dispute arises between the operator and either a person claiming he requires a licence or a representative body, that person or body may refer the scheme to the Tribunal. Both section 118 and section

Therefore, the Copyright Tribunal is required to assess RAND considerations in respect of collective licensing schemes, in disputes which may be of enormous commercial importance. Its procedure was criticised as being too slow an too costly and because hearings took far too long. As a result of those criticisms, in recent years the Copyright Tribunal has transformed its procedure by robust case management. Expert evidence is not admitted automatically and is very carefully looked at to see whether it is appropriate for an expert to be giving such evidence. By way of example, on the question of comparables in the Copyright Tribunal, expert evidence as to whether an agreement is a comparable (which is a question of fact) may not be admitted and even if it is admitted cross-examination may be strictly limited. The hearing of cases which might have lasted for six weeks now takes, typically, between 5 and 8 days.

ZyXEL's approach is criticised by TQ Delta on a number of grounds. ZyXEL assert that the benefit of adopting the technology is solely in respect of a mid-sized equipment manufacturer such as ZyXEL itself, which it chooses as the paradigm example, and that the costs of any royalties imposed on them cannot and will not be passed on by manufacturers of DSL equipment to customers. That proposition is challenged by TQ Delta.

Furthermore, ZyXEL assess TQ Delta's share of the relevant pool of essential patents by counting patents that have particular characteristics, such as the identity of the applicants or proprietors and key words appearing in the title or abstract. TQ Delta says that ZyXEL have done no more than identify a pool of patents within which it might be sensible to look for essential patents. They have not looked to see how many are essential at all. They then use this over-sized pool as the denominator in a fraction to further apportion the royalty due to TQ Delta as if all the pool represented essential patents.

Mr. Nicholson defended this approach on the basis that ZyXEL had applied the same assumption not just to the potential pool of patents, the denominator, but also to the numerator of the fraction, namely patents in TQ Delta's portfolio. They are all assumed to be potentially essential. Whether or not this approach is appropriate is a matter for the RAND hearing. However, it is necessary for me to bear in mind these issues when considering the length of trial.

This raises an issue considered by Arnold J on the strike-out application, namely that TQ Delta had served evidence, in particular evidence from Dr. Nedev, which was inadmissible.

As regards the assessment of the overall size of the patent pool, Dr. Nedev explained that any assessment needed to begin by compiling a pool within which it was sensible

Dr. Nedev then performed the exercise that he suggested was missing by a statistical sampling process that was similar to that undertaken in Unwired Planet v Huawei. He did not do it alone, but he supervised a team who participated in that exercise. Arnold J held that Dr. Nedev's criticism of ZyXEL's approach was permissible and admissible. However, he considered the explanation of what should have been done required a positive pleading, and that doing the exercise through a team with only the supervisor giving evidence was not satisfactory. He also considered that specific reasons as to why patents were considered to be essential or non-essential, needed to be apparent for the individual patents.

Therefore, at the PTR, Arnold J struck out the expert evidence of Dr. Nedev to the extent that it related to the review of the essentiality of the pool of 775 patents, which the defendants identified as potential SEPs. Those parts of Mr. Bezant's evidence that relied upon Dr. Nedev's review of essentiality were also struck out. However, Arnold J also ordered that TQ Delta had permission to amend its statement of case on RAND to introduce the confidential December licence, the economic benefits analysis contained in the expert evidence of Mr. Bezant and the issue of whether a RAND licence would cover all entities in the Unizyx group, including an entity known as MitraStar.

Since TQ Delta was already given permission by Arnold J to amend to introduce the confidential December licence as a comparable, no decision is required from me on that aspect of the amended pleading. However, I should just briefly touch on a submission made by Mr. Nicholson to the effect that it was TQ Delta's fault that the RAND hearing had to be adjourned in January because they had made a mess of their evidence. I do not accept that. Arnold J was careful to point out that the December licence could not have been introduced at an earlier stage because it had only just been entered into. In my view, given that TQ Delta wished, for understandable reasons, to rely upon the December licence, and given that ZyXEL were entitled to investigate the circumstances and terms of the December licence, an adjournment would have been necessary in any event. I do not consider that the question of fault is relevant.

Furthermore, since permission to amend has already been given in respect of the economic benefits analysis, I do not need to make any order in respect of that.

There is, however, a contested application to amend in respect of the essentiality analysis. The amendment is to allow TQ Delta to advance its positive argument as to how to conduct an economic benefits approach in case the court is attracted to such an approach. As I understand the pleading, the analysis is relied on as a cross-check. TQ Delta's primary case is based on comparables, but, as in Unwired Planet v Huawei, where Birss J used a similar calculation as a cross-check, it is put forward on that basis.

Having considered the pleading, I have decided to give TQ Delta permission to amend. It is not strikeable, and if that is case that TQ Delta wishes to advance, then I consider that it has pleaded it to a sufficient extent.

The question of the admissibility of Dr. Nedev's evidence requires separate consideration. TQ Delta takes the position that Arnold J's criticisms of Dr. Nedev's evidence, as it was served in December 2018, can and will be addressed. Its evidence is that Dr. Nedev will carry out the entire review so that the evidence will not in any way rely upon the work of others whose reasons will not be capable of being known or challenged. It explains that Dr. Nedev has already reviewed 59 of the patents himself and PA Consulting, his company, estimates it will take him 12 days to complete his review of the remaining patents in the sample. Arnold J also considered that Dr. Nedev's report must state reasons for the decisions that he has reached in relation to those patents. TQ Delta's position is that that will also be complied with. Until I see Dr. Nedev's revised report, which has not yet been served, I am not going to give permission to adduce it. Whether permission is ultimately given will, as I shall explain, depend on its contents, proportionality, whether the exercise that Dr. Nedev has undertaken is really likely substantially to lengthen the trial and the benefits that are to be gained from it.

I now turn to the reasons why it is suggested by ZyXEL that the trial cannot take place until April 2020. These reasons are summarised in the tenth witness statement of Ms. Bould. The first reason relates to a dispute, to which I have briefly referred, as to whether the benefit of using DSL standards accrues to ISPs rather than DSL equipment manufacturers. TQ Delta's case is that by using DSL, ISPs are able to make considerable cost savings in delaying the roll-out of alternative high-speed internet services, in particular optical fibre to consumer premises. Mr. Bezant's position is that this benefit should be shared between the ISP and the DSL equipment manufacturers, which results in an increased royalty for DSL SEPs.

ZyXEL do not accept that this is an appropriate approach but accept that TQ Delta is entitled to advance this case if it so wishes. However, Mrs. Bould suggests that to answer this case ZyXEL require a “sector expert” who is able to speak to matters of high-speed internet roll-out, alternative technologies and international requirements, including regulatory requirements for such services. She explains that her firm's early investigations demonstrate (which she correctly points out is self-evident) that the situation is very different from country to country. The approach of each country's national regulator differs in relation to the extent to which it requires high-speed internet to be rolled out and the extent to which the regulator has taken steps to facilitate roll-out, the costs of roll-out may differ from country to country, and so on.

Ms Bould suggests that ZyXEL may wish to call a sector expert who is able to comment knowledgeably as to the technology regulatory situation and infrastructure in at least the main markets throughout the world. I am not asked to give permission for such a sector expert to be called, not least because at present ZyXEL do not suggest that they have chosen one. Mr. Speck has powerfully submitted that the relevance of any such evidence, which does not relate to DSL technology, is at best peripheral, but I do not need to decide that at the moment.

However, I accept Mr. Speck's submissions that if such an expert were called, the additional time at trial required to accommodate this evidence would be very small. Ms. Bould has set out an estimate on the assumption that sector experts would be called by both sides and would require a total of two to three days' cross-examination. Mr. Speck's indication is that TQ Delta does not intend to call a sector expert and that cross-examination would be much shorter than that. If an expert is identified, then I think that no more than two hours will be required for cross-examination.

The second aspect that Ms. Bould has raised in relation to Dr. Nedev's essentiality analysis, on the assumption that his evidence which has previously been struck out is now admitted. ZyXEL suggest that Dr. Nedev's essentiality review, which looks at some 138 of the 775 patents in ZyXEL's pool as a statistically significant random sample and opines as to whether each of those 138 patents is essential to a DSL standard, is likely to be very substantially disputed by ZyXEL. She suggests that the work in doing this exercise, which could amount to considering essentiality of a huge number of patents will require a considerable amount of preparation and a considerable amount of time at trial. Her estimate for cross-examination of the experts in relation to essentiality is a total of six days.

In answer to this, Mr. Speck referred to the conclusions of Birss J in relation to a similar exercise conducted in the Unwired Planet case at paragraphs 333 to 335:

In summary, the experts reviewing the sample of patents which were deemed to be essential randomly selected a sample and reached a very substantial degree of agreement about which patents were and were not essential. In so far as they disagreed, it made very little difference to the numbers and was entirely unnecessary to resolve. Therefore, Mr. Speck's position is that the court at trial is very unlikely to need to resolve technical disagreements in respect of particular patents. It is primarily a statistical exercise in which differences between the experts may not matter or, if they do, are unlikely to make a substantial difference to the result, which itself is no more than a cross-check.

On the submissions that I have heard, I believe that Mr. Speck is correct. I do not consider that the court should allow anything like six days for cross-examination about essentiality. I consider that Ms. Bould's estimate is based on the premise that the experts will disagree about the essentiality of all of the patents in issue, which I think is extremely unlikely. The course that I believe should be adopted is that once Dr. Nedev's revised report is served there should be an application, if it is objected to, to admit it. The court can then consider whether it is appropriate and proportionate to admit it. If the evidence is then admitted, the experts will need to meet without prejudice and agree which patents are or are not essential. I do not consider that a significant amount of time at trial should be allowed for this kind of dispute in the context of a case where ZyXEL will have exercised its right to enforce a contractual undertaking to grant it a licence. One is postulating the position of a willing licensor and willing licensee who would not during the course of negotiations conduct contested infringement trials.

If the court, having seen the evidence of Dr. Nedev, considers that this proposed cross-check is disproportionate, or will take too long at the trial, or will lead to an adjournment of the trial, then no doubt that evidence will be excluded.

Ms. Bould’s estimate also includes 4 days for cross-examination of financial experts, 2-3 days for cross-examination of factual witnesses and 4 days for speeches. All of these estimates are, in my judgment, excessive.

In all the circumstances, I propose to allow ten days plus two days' pre-reading for the RAND trial. This may well turn out to be unnecessarily long and will no doubt be reviewed before the trial. Fortunately, Birss J has indicated that he is available and willing to hear this trial in September 2019.

I now need to consider ZyXEL's position that it will lose its counsel and will need to instruct new counsel for a September trial. It may well be that the same will be true of TQ Delta. It is very difficult to try to arrange the dates for hearings in telecoms patent cases based on counsel’s availability. There are a number of reasons for this. Counsel’s availability constantly changes. Busy counsel are frequently instructed in a number of cases many of which settle. Fraser J considered a similar issue in Alan Bates and Others v Post Office Limited [2017] EWHC 2844, where he observed that fitting hearings around the availability of busy counsel “has all the disadvantages of doing an intricate jigsaw puzzle, with none of the fun associated with that activity”. He considered that delaying hearings for such a reason was fundamentally the wrong approach. I agree.

There are a great many talented barristers. The truth is that many good advocates would be capable of presenting the parties’ cases at a RAND trial, as shown by the variety of barristers who have appeared in the Copyright Tribunal. Although I understand the importance to clients of their chosen counsel, I have no doubt that both sides will be well represented whatever date I fix for this trial.

I bear in mind that ZyXEL has not indicated any date that would be convenient for them before April 2020. The date that I have indicated, namely September 2019, will be convenient for other litigants and will not disrupt trials, because it is during the vacation. The parties will have the benefit of an expert judge. I also need to bear in mind that if I postpone this trial for a lengthy period that will, in my view, be prejudicial to TQ Delta who have provided confidential evidence as to the effect that such postponement would have on its ability to procure further licences under its patent portfolio.

Therefore, in all the circumstances, I am going to order that this trial should take place in September. I am not going to set a date in September. That is a matter for listing to work out and for Birss J to indicate what date will be preferable to him. Given the history of this matter, I would be very reluctant to accede to any further requests to adjourn the RAND trial, if, following judgment on the technical trial, such a RAND trial is required.