Unwired Planet International Ltd v Huawei Technologies Co Ltd & Anor

This judgment deals with the remedies to follow from my main judgment dated 5^th April 2017 ([2017] EWHC 705 (Pat)). There is no need to rehearse the background. The first issue I have to decide is what sort of relief should be granted in this case. UnwiredPlanet press for a final injunction to restrain infringement of the two patents found to be valid and infringed, albeit they also accept that the injunction should be stayed on terms pending any appeal by Huawei to the Court of Appeal. Huawei contend that I should accept certain undertakings from Huawei and not grant a final injunction. This first issue is the major question I have to decide. The other issues are: second what to do about damages, third a point on certain declaratory relief, fourth costs, and fifth permission to appeal.

The main judgment decided that a particular form of licence represents the FRAND terms between the two relevant parties, UnwiredPlanet and Huawei, in the relevant circumstances. That licence will be referred to as the Settled Licence. It is annexed to this judgment. The form of the Settled Licence was finalised a few weeks after the main judgment. There is no dispute that the Settled Licence in that form reflects the decisions made in the main judgment.

The Settled Licence is a global licence. Throughout the proceedings Huawei had maintained that they were not prepared to enter into a global licence with UnwiredPlanet. Huawei had made it clear that they did not accept that a global licence was FRAND as a matter of competition law. This was reflected in the main judgment, for example paragraphs 524-572. I held that such a licence was FRAND and that since UnwiredPlanet had established infringement by Huawei of valid patents and since Huawei have not been prepared to take the FRAND licence, a final injunction should be granted (paragraphs 793-795 and the summary paragraph 806(18)).

I had made it as clear as I could to the parties during trial that the trial itself was the hearing at which the question of whether or not to grant an injunction was to be decided. However because Huawei had refused to engage with the terms of a worldwide licence, the fully worked out Settled Licence did not exist at the date the judgment was handed down. Reluctantly I decided that the injunction could not be dealt with at the handing down but would be dealt with at a later hearing (see main judgment paragraph 794). Although Huawei were to that extent benefitting from their earlier non-engagement, nevertheless that seemed to me to be the proper course.

A preliminary question is whether there is anything to decide at all. Huawei submit that I have not actually made the decision whether to grant a final injunction or not. However in case they are wrong Huawei have also issued an application to reconsider the grant or refusal of a final injunction under the jurisdiction confirmed by the Supreme Court in Re L-B [2013] 1 WLR 634. UnwiredPlanet pointed out that in the main judgment I said an injunction “should be granted” (see summary paragraph 807(18)) but UnwiredPlanet did not press a formal point that the issue could not be considered on its merits at this hearing.

Since no injunction was granted when the main judgment was handed down and since I stated in terms that I would deal with an injunction at a later hearing, I have not yet exercised the court’s power one way or another on this issue. It is open to me to decide the matter and there is no need to consider exercising the Re L-B jurisdiction.

UnwiredPlanet contend that a final injunction should be granted. They agree that it should be stayed on terms pending appeal, but that is all.

Huawei submit that the present circumstances are now materially different from the circumstances during trial and at the time the judgment was handed down. Now Huawei offer two undertakings to the court. They are set out in the witness statement of Huawei’s UK solicitor Simon Ayrton of Powell Gilbert. He explains that pending appeal Huawei are prepared to provide the following undertakings to the Court:

“First, that within seven days of the expiry of the time to appeal against the Final Order or, if there is an appeal against the Final Order, its withdrawal or final determination, Huawei will enter into the form of licence settled by Mr Justice Birss (i.e. the Licence) or such other licence as may be finally settled by the courts.

Second, that until the expiry of the time to appeal against the Final Order or, if there is an appeal against the Final Order, pending its withdrawal or final determination, Huawei will abide by the terms of the Licence as if the same were in full force and effect.”

The “Final Order” is the order whose terms I am deciding now. The “Licence” is what I have called the Settled Licence.

Huawei submit that I should accept this undertaking and refuse to grant a final injunction. They submit that UnwiredPlanet are protected pending appeal because Huawei will pay royalties in the meantime (subject to a point about a letter of credit to deal with what happens if Huawei win their appeal). They also submit that the effect of this undertaking is that one way or another there will be a licence in place with Huawei as a party. Mr Ayrton says that these undertakings mean that Huawei are willing to enter into whatever licence the UK courts settle and in the meantime will abide by the Settled Licence as if it is in force. If no other licence is finally settled by the courts then Huawei will enter into the Settled Licence. So the final injunction will serve no useful purpose for the duration of the Settled Licence or such other licence as is finally settled by the courts. Mr Ayrton then points out that the Settled Licence will expire in 2020 (and a 2020 expiry was an agreed term of the UK licence at trial). Since patent EP (UK) 2 229 744 which was considered in trial A and found valid and infringed does not expire until 2028, the injunction will still be in being in 2020 when the Settled Licence expires. Having the injunction in place at that time would unfairly distort the positions of the parties in any renegotiation which would need to take place.

UnwiredPlanet take a number of points in response to these submissions. Their first major argument is that Huawei’s submissions mix together the question of what the final relief should be absent any appeal, with the question of what the terms should be pending appeal. They contend that, subject to an issue about letters of credit, there is no dispute about the appropriate terms pending any appeal. Huawei in effect should behave as if the Settled Licence is in force and pay royalties in the meantime, subject to repayment by UnwiredPlanet. But the fact that the final injunction should be stayed pending appeal does not mean it should not be granted. They argue that today the position remains that Huawei refuse to enter into the licence found to be FRAND by the court and so an injunction ought to be granted. They argue that what Huawei are trying to do is decline to enter into the licence settled by the court at least while the appeal is pending but still avoid an injunction. They say this builds an interim measure into what should be final relief.

UnwiredPlanet’s second point is to dispute Huawei’s assertion that the effect of the undertakings is that it is inevitable that Huawei will enter into a licence with UnwiredPlanet at the conclusion of the appeal process. UnwiredPlanet do not agree that the undertakings offered make it watertight, in every conceivable scenario on appeal, that a licence will be in force between UnwiredPlanet and Huawei at the end. That depends on examining the points on which Huawei wish to appeal, which I will return to below.

Like many aspects of this case, this dispute about relief raises some new problems. One suggestion was that the real problem was that the Settled Licence had no terms dealing with appeals. Huawei submitted that if it did then they might be able to enter into it now but still argue on appeal that a UK licence was the right outcome. I agree that it might be possible to craft terms in a licence which could allow for something like that and perhaps the next time a court is called upon to settle FRAND terms, that could be considered. But neither side suggested such terms at trial and I made it clear the time to do that was then, not now (e.g. paragraph 581). Moreover I am not convinced it is as simple as Huawei suggest.

Another point is the issue of what happens in 2020. This arises from the fact that the eight year term of the Settled Licence runs from 2013 until 2020. That term arose because the parties at trial were agreed that the UK licence should work that way. When I came to consider the worldwide licence I started from the UK terms, finding (paragraph 593) that the points the parties were able to agree upon as FRAND in the UK licence were just as applicable to a worldwide licence. That included recognising that it would be an eight year licence running from expiry of the 2009 Ericsson-Huawei licence until the end of 2020.

When, during the trial, I saw the agreed expiry date of 31^st December 2020 in the UK licence my first reaction was one of surprise. It was only three years away and at least some of the patents will last much longer than that. But on reflection I could see that such a thing made sense and I was not surprised the parties had agreed on it. One of the real problems with FRAND and SEPs is that all sorts of relevant circumstances change. Old patents expire and new ones are granted. Standards themselves change. One might think new patents cannot be essential to old standards but even that is not as simple as it might appear. Fixing a rate for a period of years involves a risk (paragraph 563) and so the idea that today one can agree what the rate should be for many years into the future is difficult. If patents courts are to support orderly FRAND licensing of standards essential patents, as I believe they should, then a workable approach needs to be developed which allows for these changing circumstances.

Huawei are right to raise this problem. It is inherent in settling licences in this field. It would have arisen just the same if Huawei had won on the UK/worldwide issue. It has to be sorted out whoever is correct about the final form of relief.

In order to decide what to do I believe it is useful to look at what the final remedy should be absent any appeal and then consider the position pending appeal separately. I note that in Kirin-Amgen v Transkaryotic Therapies Inc (No 3) [2005] FSR 41 (decided in 2001 but reported in 2005) Neuberger J (as he then was) took a similar approach when he was considering the terms of a final patent injunction and the terms of a stay pending appeal. In paragraphs 20 and 21 the judge considered what the correct order would be “absent TKT’s appeal”.

I will start by looking at what the right final relief should be in a patent case with a FRAND licence absent any appeal. Thinking about the problem of what happens in 2020 highlights an aspect of UnwiredPlanet’s case which goes too far. UnwiredPlanet contend that, absent any appeal, the patentee is entitled to a final injunction which should run from now until the expiry of the relevant patents and should continue even if the defendant enters into a licence. They argue that such an injunction will not prejudice Huawei, even if Huawei become licensed, because the prohibition is only to not infringe. Therefore, they say, if Huawei have a licence, then their relevant acts are not infringements and not prohibited by the injunction.

In my judgment this approach is unreal. By that logic if I granted the injunction today and Huawei signed the Settled Licence tomorrow the injunction would remain in existence. What happens if a year later a dispute arises about an alleged repudiatory breach of the licence? Huawei could then be in an impossible position because they would risk being in contempt of court if they continued to sell equipment during a period when there was an argument that the licence had come to an end.

With this problem in mind and assuming an injunction should be granted at all I will consider, absent appeal, what the correct form of a final injunction in respect of patents the subject of a FRAND undertaking should be when the court has settled a FRAND licence but the defendant has not entered into it. I will call this a FRAND injunction. The answer is simple. A FRAND injunction should be in normal form to restrain infringement of the relevant patent(s) but ought to include a proviso that it will cease to have effect if the defendant enters into that FRAND licence. If as in this case, the FRAND licence is for a limited time, shorter than the lifetime of the relevant patents then the injunction should also be subject to an express liberty to either party to return to court in future to address the position at the end of the term of the FRAND licence. In any case the FRAND injunction should also be subject to an express liberty to apply in the event the FRAND licence ceases to have effect for any other reason.

Normally in English law when the court grants final relief a party is not entitled to come back to court in future even if circumstances change but these unusual terms arise from dealing with patents the subject of a FRAND undertaking. A FRAND injunction in this form reflects the finding that what the patentee is entitled to today, bearing in mind its FRAND undertaking, is a licence on FRAND terms but if the defendant has the ability to take the licence and does not do so, then an injunction is appropriate for as long as the defendant does not enter into that licence. If the defendant enters into the FRAND licence there should be no injunction at all. The fact the FRAND licence is limited in time does not justify an injunction continuing into the future. The court should not pre-judge at this stage what should happen if or when the FRAND licence ceases to have effect.

In this case the FRAND licence is the Settled Licence. It will expire on 31^st December 2020. Thinking about the position on 1^st January 2021 at this stage now in 2017, it cannot be said that UnwiredPlanet must have an injunction on that future date but neither can it be said today that Huawei must be free to sell products or commit any other act on that future date which would infringe if unlicensed. The answer to those questions depends on the circumstances at that time. At times in argument before me it seemed to be contemplated that the process of resolving what is to happen after the end of the Settled Licence only has to start on 1^st January 2021. That is wrong. If the patentee has failed to start a process of FRAND negotiation well in advance of the expiry of the current FRAND licence then no doubt the court will be unsympathetic to the patentee even if no licence has been finalised to start the day after expiry. So too if the patentee has dragged its feet in the negotiation. Conversely if the patentee engages reasonably but the putative licensee does not, then the court’s sympathies may well lie the other way round.

To deal with an appeal the FRAND injunction in the form described can simply be stayed pending appeal on terms to secure appropriate payments in the meantime.

I turn to consider whether a FRAND injunction should be granted or whether I should accept the undertakings offered by Huawei instead.

The grant or refusal of an injunction is an exercise of the court’s discretion. That discretion is circumscribed in various ways. Nevertheless when the holder of an intellectual property right has established infringement and a threat and intention to infringe in future, an injunction will normally be granted. In the words of Art 3 of the Directive on the Enforcement of Intellectual Property Rights (2004/48/EC) an injunction in those circumstances is normally effective, proportionate and dissuasive.

It is not unusual in intellectual property cases for the defendant, having defended the claim vigorously throughout, to turn round after the judgment has been given against it and seek to offer undertakings not to infringe in order to avoid being subject to an injunction. Rights holders generally resist this course and insist on an injunction. In part that is because it involves a public vindication of their rights. Indeed that is exactly what the defendant is usually seeking to avoid by its belated offer of undertakings not to infringe. The practical effect of the two things (injunction and undertaking) is the same, and actually undertakings can be rather easier to enforce under English civil procedure, but nevertheless rights holders usually insist on an injunction for the reason I have explained. The claimant has been forced to come to court and vindicate its claim. An offer of undertakings after judgment is usually regarded as made too late. In general the courts will order the injunction and refuse to accept such belated undertakings.

Turning to this case, in my judgment it is not irrelevant that the undertaking offered by Huawei is only offered now, after many years of litigation and after judgment has been handed down. If, from the outset, Huawei had given an unqualified undertaking to enter into whatever licence the court (including an appellate court) had decided was FRAND, then an injunction at this stage would not have been appropriate. Indeed Huawei argued that the terms of the qualification on the claim for an injunction sought by UnwiredPlanet from the beginning (main judgment paragraph 679-683) run counter to such an injunction. That would have been a better point if Huawei had offered this undertaking at an earlier stage. It is also notable that Huawei argued, and I accepted, that the qualification was ineffective.

On the other hand the patentee’s FRAND undertaking does make this situation very different from a normal IP case. The question is whether, given the FRAND undertaking, even a very belated undertaking to take whatever FRAND licence the court settles should mean that the court should refuse a FRAND injunction in the appropriate form. The possibility of an appeal has nothing to do with it because Huawei’s ability to appeal is and can be protected whichever approach I take.

Huawei contend that absent an appeal they would sign the Settled Licence now. I do not doubt that but it is no different from the position of any defendant who changes their tune after they have lost a case. Throughout these proceedings Huawei have maintained a determined stance that they would not accept a worldwide licence in this way. It has been a fundamental issue. In my judgment the undertaking offered by Huawei now is too late. By refusing to offer an unqualified undertaking before trial and before judgment Huawei forced UnwiredPlanet to come to court and vindicate its rights. The right thing to do now is grant a FRAND injunction albeit one which will be stayed on terms pending appeal.

I have reached this conclusion without considering UnwiredPlanet’s submission that the undertakings in the form offered by Huawei are not watertight. My initial view – from looking at the words of the undertakings themselves and in particular the first undertaking - was that they were watertight. In other words that it was legally inevitable that whatever happened on appeal Huawei would end up being a party to a licence covering at least the United Kingdom. However what emerged during the course of argument is that one cannot assess the effect of these undertakings without also taking into account the points Huawei wish to argue on appeal. They are addressed below. At this stage all I will say is that it seems to me that a conceivable result of the points Huawei wish to argue and which I am satisfied have a real prospect of success is that an appellate court could decide that the court cannot settle the terms of a licence. If that happened then on the face of the wording the undertakings do not oblige Huawei to enter into any licence. Huawei say that that outcome on appeal is not their case. Rather their case is that the licence should be a UK only licence. I accept that the outcome I have described is not the outcome of the primary case Huawei contends for, but it is a possible consequence of the arguments Huawei advance. It could arise if Huawei are correct that there is more than one set of terms which are FRAND and possibly in other circumstances too. For this reason also I prefer to grant a FRAND injunction stayed pending appeal rather than accept the undertakings offered.

A point arose on the letter of credit which UnwiredPlanet have offered to deal with repayment of any sums paid in the interim pending appeal. UnwiredPlanet proposed a letter of credit drawn on a bank called Pegasus Bank in Dallas. Huawei contend that the bank should be one in the UK. UnwiredPlanet say that to achieve that will cause very substantial and unjustified extra cost. Huawei’s representatives explained that they are not aware of Pegasus Bank and should not be compelled to accept a letter of credit from it. The fair approach is to require UnwiredPlanet to give a bank guarantee to Huawei’s satisfaction. If that comes at a cost then those costs should be recoverable from Huawei if no repayment is made to Huawei, but borne by UnwiredPlanet if any sum is repaid.

Counsel for Huawei raised a concern that the draft order provided by UnwiredPlanet includes a penal notice. In the circumstances that is inflammatory and unnecessary. It is only necessary to serve an order with a penal notice if it is to be enforced under CPR r81.4. Since on any view the injunction will only bite after an appeal, UnwiredPlanet can always serve a further copy of the order, endorsed with a penal notice under CPR r81.9, in future if they feel it is necessary. There is no reason why the order should contain one now.

This is closely related to the first issue. The point is that if Huawei enter into the Settled Licence then no problem about damages for past acts arises because they are covered in that agreement. But as of now Huawei have not entered into the Settled Licence. Absent appeal if Huawei never did enter into the Settled Licence then damages will be due and an order for damages should be made. The order should be in the same form as the FRAND injunction, i.e. that it will cease to have effect if the defendant enters into the Settled Licence. The way to deal with appeals is to stay the matter pending appeal.

It is common ground that some declarations should be made. Two are agreed, one that UnwiredPlanet’s various offers are not FRAND and the other that Huawei’s offers were not FRAND. It is also common ground that some sort of declaration should be made about the Settled Licence but there is a dispute about the terms. The drafts assume the licence is annexed to the order but I will annex it to this judgment instead. Making allowance for that trivial change (in italics), the parties rival declarations are

Huawei’s point is that UnwiredPlanet’s formulation does not capture the finding that only one set of terms are FRAND in a given set of circumstances. Huawei’s version aims to do that. UnwiredPlanet resist Huawei’s drafts as tautologous and unduly complicated. They argue that the first negative declaration is redundant in the light of the second positive declaration and submit that the real purpose of this wording is to give Huawei something to appeal against when the effect of the appeal would otherwise be academic.

I agree that a declaration about the status of the Settled Licence ought to be made. However the declarations proposed by Huawei are too complicated and I am not convinced there is any good reason to take that approach. The declaration will be:

“the licence annexed to the judgment represents the FRAND terms applicable between the parties in the relevant circumstances” .

That seems to me to adequately capture the conclusion reached in the main judgment.

Finally Huawei also seek a declaration that UnwiredPlanet had not abused its dominant position. The purpose of this was, I think, to create a reason for appealing the Huawei v ZTE point and arguing that no injunction was appropriate even if Huawei’s undertakings had been accepted and no injunction granted. Whether that was a kind of bootstrap argument is not something I have to decide since I have granted an injunction. Since the judgment makes a clear finding that UnwiredPlanet has not abused its dominant position and states as much in summary paragraph 807(17), there is no need for that declaration and I will refuse it.

UnwiredPlanet argue that they should be regarded as the successful party (following Roach v Newsgroup Newspapers [1998] EMLR 161 and other cases) and therefore the order relating to costs should be that Huawei pays UnwiredPlanet’s costs of these proceedings subject to some issues which UnwiredPlanet accept they cannot recover. The issues to be excluded are the claims against Google and Samsung and the allegation that the MSA was void under Art 101 TFEU. Although Huawei adopted part of Samsung’s Art 101 case, it is convenient to refer to all these excluded issues as the Samsung issues.

UnwiredPlanet’s actual costs of the non-technical trial are about £12.4 million. Huawei’s costs were about £7 million. UnwiredPlanet seek a payment on account from Huawei of about £6.4 million which is arrived at in the following way. The starting point is a costs’ estimate confirmed in July 2016. At that stage UnwiredPlanet estimated their total costs to the end of the proceedings would be about £8.5 million. Huawei queried how a total of £12.4 million could have been incurred by the end of the non-technical trial (the hearing ended in December 2016) given that estimate. It emerged that at least part of the explanation was that the estimate was seriously in error. Nevertheless realistically UnwiredPlanet accepted that as a starting point to calculate a payment on account the July estimate should be taken. Starting from £8.5 million UnwiredPlanet deduct 24% for the Samsung issues. That comes to £6.4 million. If, contrary to their case a deduction should be made for the costs of the FRAND rate, those costs represent 43% of UnwiredPlanet’s costs. So the payment on account should be 57% of £6.4 million which is £3.6 million.

Huawei contend that the correct order for costs is no order. In other words each side bears its own costs. They argue that this was an unusual trial and it is not realistic to identify a single overall winner. There were three major issues: (i) worldwide vs UK licence; (ii) the FRAND rate and (iii) Article 102 TFEU. Although UnwiredPlanet won on the first and third issue, Huawei submit that Huawei won the rate issue or at the very least it cannot be said that UnwiredPlanet won that issue. If Huawei can be regarded as the winner of the FRAND rate issue then the fair order is no order as to costs. Analytically that result could be arrived at either on the basis that there is no overall winner or on the basis that even if UnwiredPlanet were the overall winner, the correct way to reflect Huawei’s success on the FRAND rate, given that it represented broadly 40-50% of the costs, would be to make no order. The latter approach would notionally set off Huawei’s costs of the FRAND rate against UnwiredPlanet’s costs on issues (i) and (iii) and taking into account the irrecoverable Samsung issues. If “no order” is not accepted, the best for UnwiredPlanet ought to be that they recover the costs for issues (i) and (iii) but not issue (ii).

As regards any payment on account, if a payment were to be ordered Huawei contend that the right starting point is the July 2016 estimate (£8.5 million) however the deduction of irrecoverable Samsung issues should be larger than 24%, after all when Samsung came out of the proceedings the trial estimate shrank from 13 weeks to 7 (i.e. 53%). From there the result depends on the earlier findings.

With that explanation of the parties’ arguments I can decide what to do. The principles applicable to costs are not in dispute and I will not rehearse them.

I agree with Huawei that it is fair to consider this case as consisting of the three major issues identified. UnwiredPlanet clearly won issues (i) and (iii). The debate is about issue (ii) – the FRAND rate. The simple point is that in terms of royalty rates the results on issue (ii) were % values lower than UnwiredPlanet’s case and higher than Huawei’s. In terms of amounts of money to be paid there is no question that on the judgment UnwiredPlanet will receive much more than was being offered by Huawei. Nevertheless it is fair to observe that the vast majority of the difference is due to the worldwide nature of the licence and issue (i) rather than the differences in the rates (issue (ii)).

I do not believe it is accurate to say that overall there was no winner. The appropriate way to characterise the result is that UnwiredPlanet are the overall winner and entitled to their costs, but that the FRAND rate issue (ii) should be considered to see if deductions are appropriate. Issue (ii) included dealing with the MNPA and HPA techniques, the analysis of comparables and the aggregate royalty calculations. It is a suitably circumscribed issue which lends itself to being considered in this way and it is pitched at an appropriate and fair level of generality.

I turn to consider who won the FRAND rate issue.

Considering just the benchmark rates for 4G multimode handsets, Huawei contended for 0.040% whereas UnwiredPlanet contended for 0.13%. The judgment set this benchmark at 0.062%. UnwiredPlanet point out that they had to come to court to get a rate higher than 0.040%, which is true but the same can be said the other way round, Huawei had to come to court to get a rate lower than 0.13%. In terms of numbers the outcome is closer to Huawei than to UnwiredPlanet. A rate of 0.062% is about half the rate UnwiredPlanet contended for (0.13%) but only about one and a half times Huawei’s contention (0.040%).

UnwiredPlanet pointed to an offer made by Huawei on a “without prejudice save as to costs” basis in October 2016. Offers made on that basis are not shown to the court at trial but can be relied on when dealing with costs. They can play a very important part in the assessment of who won on an issue involving a sum of money. The offer was put on a patent by patent basis but it includes a lump sum offered as damages for the UK for the period 2013-2016. Mr Moss of EIP for UnwiredPlanet calculated the lump sum offered for that period as being about $459,000. It can be contrasted with an estimate of the amount due under a UK portfolio licence on Huawei’s case as to rates of about $453,000. UnwiredPlanet submitted that this represented Huawei’s “line in the sand” so that to receive any amount above that represented success for UnwiredPlanet. The “line in the sand” metaphor is not helpful in this case. It is just another way of repeating the point I have already noted that UnwiredPlanet say they had to come to court to get a benchmark rate higher than 0.040%. To repeat the answer, that is true but it is just as true that Huawei had to come to court to get a rate lower than 0.13%. UnwiredPlanet cannot avoid that problem just because Huawei made a without prejudice save as to costs offer. I do not find the without prejudice save as to costs offer makes any material difference.

UnwiredPlanet emphasised the amounts of money which Huawei will have to pay as a result of the judgment. The figures are all based on estimates for the sum due for 2013-2016 and therefore do not represent the total sum due, but since they are all on the same basis they can be used for comparison purposes. In terms of a lump sum, the most Huawei were offering was the sum above – i.e. about $453,000. By comparison the amount Huawei would have to pay for the period 2013-2016 under the Settled Licence is about $23 million. On the other hand, as Huawei point out, an estimate on the same basis of how much Huawei would have had to pay for 2013-2016 on UnwiredPlanet’s original royalty case which remained their case from 2014 to 2016 is about $150 million. That original case involved a 0.2% rate for 4G. The estimate on the same basis using UnwiredPlanet’s case as at August 2016, when the 4G rate dropped to 0.13%, is about $100 million. Finally, I should mention the same sum based on the UK portfolio licence rates in the judgment. That would be about $0.9 million.

UnwiredPlanet emphasise these numbers as showing how they will receive much more than Huawei offered and Huawei make the familiar reply that the amounts are much less than UnwiredPlanet claimed.

An important factor Huawei emphasise is the effect of the decision about the OM and China rates. Taking 4G multimode handsets the judgment was that for Major Markets (MM) the rate should be 0.052% whereas for China and Other Markets (OM) the rate should be 0.026%. Of course, 0.026% is lower than the 0.040% benchmark contended for by Huawei and that has an important effect which Huawei were entitled to emphasise. In simplistic terms if the licensee’s total sales were $1 million in Major Markets and $1 million in Other Markets including China, then the royalty due using the two rates set by the court would be $520 (MM) plus $260 (OM) which is $780 in total. You could express the $780 total as a single percentage of global sales of $2 million. It would be 0.039%. In other words, just below the 0.040% benchmark rate. If the sales mix was shifted to favour the OM and China markets, this notional global blended equivalent rate would fall even further towards 0.026%. Mr Lasinski performed a more sophisticated version of this calculation based on estimates of Huawei’s actual sales mix in the relevant periods. The notional global blended rate he comes to for 4G handsets is 0.030%. Huawei pointed out that 0.030% is lower than certain figures used in their FRAND Statement of Case as benchmark numbers.

At one stage it did seem to be suggested by Huawei that this approach showed that Huawei had won the FRAND rate issue because these calculations showed that it had achieved a lower effective global benchmark rate than rates it was actually offering UnwiredPlanet in these proceedings. That submission goes much too far, for three reasons. First, Huawei were not offering such global rates at all. They never offered a global licence. These rates were only being used to arrive at a UK rate. Second, Huawei made it very clear that their case was that a rate for China and elsewhere should be much lower than the benchmark numbers it was advancing. It was never suggested that these benchmarks put forward by Huawei should be applied to China. Third, UnwiredPlanet question whether the way Mr Lasinski approached the sales mix for 2013-2016 and the future was fair. If the detail of these calculations was important then I would need to examine this third issue in more detail but it is not necessary to do that. The basic point, that a notional global blended rate will lie between the MM and OM rate, is obvious. Where exactly it comes out does not matter.

The answer to the question of who, if anyone, won on the FRAND rate depends on the proper way to look at it. UnwiredPlanet argue it is like the assessment of the amount of damages to be paid. They had to come to court to get more than the defendant was offering. They have achieved that and so they have won. A defendant in that case can protect itself in costs with a without prejudice save as to costs or an offer under CPR Part 36 which works in a similar way. Huawei had tried to take that approach but not succeeded. Looked at this way, UnwiredPlanet won.

Huawei argue that this issue is more like a tariff setting exercise in the Copyright Tribunal and draw my attention to AEI Rediffusion v PPL [1992] 1 WLR 1507 and in particular to Mummery LJ at 1518 in which he held that when the tribunal finds that both side’s proposed terms were unreasonable and the tribunal has arrived at a result somewhere in the middle, it is not correct to proceed on the basis that the outcome must produce a winner and a loser.

I remind myself that the approach to costs in the Copyright Tribunal, at least at that time of AEI, was not the same as in court, and that under the CPR unreasonableness is not required in order to deprive a party of his costs (as noted by Lord Woolf MR in the same case at p1523).

I also bear in mind that the reason I am asking this question is about the incidence of costs relating to an issue when I have decided that UnwiredPlanet are the overall winner. Taking all these factors into account, in my judgment it is appropriate to see this issue as more like a tariff setting exercise than a simple claim for damages. Given the FRAND undertaking a rate had to be settled somehow. Huawei offered rates which were too low to be FRAND and they will have to pay at a higher rate than they were offering. They did not win the rate issue. But it would not be just for UnwiredPlanet to recoup the very substantial costs of the FRAND rate issue, along with the rest of the costs they will recover. Their offered rates were not FRAND either and were quite some way further from the end result than Huawei’s. The correct thing to do with the costs associated with the FRAND rate is to deprive UnwiredPlanet of them but go no further than that.

So the costs order will be that Huawei should pay UnwiredPlanet’s costs of the non-technical trial save for the Samsung issues and save for the costs of the FRAND rate issue.

Turning to the payment on account, the first point is UnwiredPlanet’s figure of about 24% to represent the costs of the Samsung issues. That figure is supported by Mr Moss’ evidence and comes from a detailed analysis of the costs which was done by a costs drafting firm DeNovo. It is challenged by Mr Ayrton in his evidence. He believes that a much larger deduction should be made and makes a number of detailed points about the way UnwiredPlanet’s costs have been handled. It is not going to be productive to delve into the detail on this issue since it is only concerned with a payment on account. Unlike other patent cases I am not being asked to make an overall percentage costs order. Mr Ayrton suggests using the reduction in the length of the trial when Samsung settled as a reasonable benchmark. I recognise that costs incurred pre-trial are not always reflected in the time at trial but I accept Mr Ayrton’s approach at least as indicative. The fact that trial length was reduced to nearly half its original estimate when Samsung settled is firm evidence that a deduction of only 24% does not fairly reflect the amount of costs which Huawei should not be liable for. But a deduction of 50% is too much. I find that a fair discount to use would be 40%.

Huawei also raised a good number of concerns about UnwiredPlanet’s costs but for the purpose of assessing a payment on account, by starting from the estimate as at July 2016, these matters do not need to be considered. Therefore the amount of a payment on account will be calculated as follows.

The starting point is the July 2016 figure of £8.5 million. I deduct 40% for Samsung issues, which comes to £5.1 million. I will then deduct 43% to reflect the non-recovery of the FRAND rate costs (recognising that 43% of £5.1 million is different from 43% of £6.4 million) which comes to £2.9 million. That is an appropriate sum to be paid as a payment on account of costs.

If a payment is to be made by Huawei they ask that the order should provide for the due date to be calculated by reference to the date the order is sealed rather than the date the order is made, because the sealed version is needed for the internal processes necessary to set up a payment. I accept that.

Huawei seek permission to appeal on essentially three grounds. Ground 1 is what they call the global licensing point. This includes submitting to the appellate court that more than one set of terms can be FRAND in any given circumstances, that the court ought to have held that the UK licence was FRAND and that the court ought not to determine FRAND terms including rates for territories other than the UK. It also involves submitting that it was not appropriate to grant an injunction excluding Huawei from the UK market unless it took a global licence. These points are properly arguable issues of principle and I will give permission to appeal but with one brief observation.

The arguments include a reference at paragraph 1(iv) to the fact that the licence includes SEPs held by the tenth party but excludes certain SEPs acquired from Samsung and Huawei’s dealing in telematics. Albeit I recognise that Huawei were reluctant to do so and that the agreement was in the context of the UK licence only since they refused to engage with worldwide terms at all, nevertheless Huawei did agree all those terms in the UK licence with UnwiredPlanet. If they had wanted to challenge them or contend for other terms, Huawei always had the means to do so. I have considered whether to refuse permission to appeal on that narrow point but decided that it would be better to allow Huawei to make the points they wish to make.

Ground 2 is the hard-edged non-discrimination point. The key argument is whether the judgment was right to require a distortion of competition. That is a properly arguable issue of principle and I will give permission to appeal. Paragraph 4(iii) under this heading submits that the court was wrong to hold that the discrimination between Samsung and Huawei would not tend, having regard to the whole circumstances of the case, to lead to a distortion of competition. This is just rearguing the court’s assessment of the facts and evidence (see the reference to Dr Niels’ cross-examination at paragraph 517). I refuse permission to appeal on that point.

Ground 3 is the Huawei v ZTE issue, injunctive relief and abuse of dominance. These are properly arguable matters of principle and I will give permission.

A small point is in paragraph 5(iv) in which Huawei submit that the judgment errs in identifying a “ratio” of Huawei v ZTE because no distinction between ratio decidendi and obiter dictum exists in European jurisprudence, citing National Navigation v Endesa [2009] 1 CLC 393 (Gloster J) and 2 CLC 1004 (Moore-Bick LJ). Huawei did not raise this when the draft judgment was provided to them. Looking again at the paragraphs of the judgment which are referred to (paragraphs 738-739 and 744(iii)), had I been aware of National Navigation when the judgment was written, I would have used different words, but the substance would not have changed. The point in National Navigation is that there is no distinction between a binding ratio and merely persuasive obiter dictum in decisions of the CJEU. One cannot disregard clear propositions of law in a European judgment just because they were not part of the ratiodecidendi. I do not believe the paragraphs in the main judgment disregard any aspect of Huawei v ZTE as obiter or merely persuasive. The point being made was that the CJEU decided what was not an abuse rather than deciding what was. Nevertheless since the main judgment uses the word “ratio” (probably inaccurately) I will give permission to appeal on that point too.

UnwiredPlanet’s position on permission to appeal was that they did not wish to appeal if Huawei did not, but if Huawei have permission then they seek to appeal on two points. The first is that the judgment failed to take into account the blended global nature of the benchmark rate. The second is that to exclude infrastructure servicing revenues from the royalty base was perverse.

The significance of the first point is that, as UnwiredPlanet submit, once I had derived a benchmark rate it was in fact used as a maximum and then two downward adjustments were made. They argue that the judgment overlooked the fact that the benchmark was a global blended rate and included contributions from territories with lower rates as well as higher rates. I do not believe that is an accurate characterisation of the judgment for these reasons. A key component in the benchmark rate was the value E for Ericsson’s portfolio. That was derived taking into account rates which were global blended rates but also rates which were not. In order to allow this judgment to be given in public I will not say any more. Nevertheless in case UnwiredPlanet’s characterisation of the judgment is correct, I will give permission to appeal on this point.

The second point is about one of the other terms. UnwiredPlanet argue that the judgment ought to have preferred UnwiredPlanet’s wording, even though I found that UnwiredPlanet’s wording was too extreme, on the basis that the decision that wording was too extreme was reached without a proper evidential basis. Since it was up to UnwiredPlanet to file such evidence if they wanted to and given the history of this litigation that submission is unmeritorious. Furthermore the submission that as an alternative I ought to have invited Huawei to propose alternative wording or invited further argument is unmeritorious too. UnwiredPlanet well understood and were content with the approach being taken that I was seeking to reach a conclusion on a full set of licence terms and avoid a further hearing (cf paragraph 581 of the main judgment). I will refuse permission to appeal on this point.

In summary:

Definitions

Grant of License

Applicable Law and Dispute Resolution

Notices

Non-waiver

Force Majeure

Miscellaneous

For [Insert applicable Huawei reporting entity] in Major Markets:

For [Insert applicable Huawei reporting entity] in Other Markets and China:

For [Insert applicable Huawei reporting entity] in Major Markets:

For [Insert applicable Huawei reporting entity] in Other Markets and China: