Warner Music UK Ltd & Anor v Tunein Inc

This judgment deals with two aspects of the consequential orders arising from my main judgment, which was at [2019] EWHC 2923 (Ch). I refer to that judgment and will not repeat the background and circumstances at length. In summary the main judgment decided that the defendant had infringed the claimants’ copyright in sound recordings of music when that music was played through the defendant’s systems (website and apps) from internet radio stations in categories 2, 3, and 4 but not category 1.

The two issues this judgment deals with are the terms of injunctive relief (and the scope of the inquiry as to damages) and costs. At the hearing I decided the first issue and gave the parties my decision but there was not time to give an ex tempore judgment on the point. This judgment contains the reasons for that decision. The second part of this judgment, which deals with costs, is the approved transcript of the reasons given at the hearing.

There are two questions to decide. The first is the scope and effect of the main judgment and its impact on the relief. The second is about the terms of any injunction.

Given the scale of the defendant’s activities, the case was managed so that the liability trial was conducted by reference to samples. There were sample sound recordings and there were also sample radio stations, divided into four categories. The total number of sample radio stations at trial was less than 40 whereas the defendant’s service carries about 70,000 to 100,000 internet radio stations.

The defendant submitted that any relief – injunction or inquiry as to damages – should be limited to the individual sample stations unless and until there had been a finding about other stations. The claimants did not agree. They submitted that the relief should be in general terms. I decided that the relief should be in general terms in that respect. These are my reasons.

The case management order relating to samples was part of the order of Master Price on 23^rd April 2018. That was the main directions order to bring the case to the trial which took place before Henry Carr J. Paragraph 4 provided for a split trial of liability and quantum. Paragraph 9 related to samples, as follows:

The four categories referred to in the main judgment as Categories 1, 2, 3 and 4 are set out in paragraphs 9(1)(i) to (iv). Paragraph 9(4) provides for a process after the trial to address the scope and effect of the main judgment. The defendant’s case was not based on a technical point about the way to invoke this mechanism but, in case there was such a point, the claimants issued an application under paragraph 9(4) for the relief they sought in any event.

In my judgment the answer to the defendant’s submission is that the conclusion reached in the main judgment justifies relief in general terms, not limited to the individual samples. The mechanism set out in the directions was an appropriate one because until one knew what the conclusion of the trial was, one could not say in advance what its effect and the consequences would be. For example, for all one knew at the time of making the directions, the result could have turned on a detailed examination of differences between different sample sound recordings or between different stations within the individual categories. However in the result it did not. Putting this in terms of the process envisaged by paragraph 9(4), I find that the scope and effect of the findings actually made means that relief in general terms is appropriate. The result does not turn on any details about individual sound recordings at all, nor does the result turn on distinctions between stations in the four categories, nor has there emerged a fifth category of station which the findings do not apply to.

Moreover the main judgment does not accept the defendant’s attempt to try to shift responsibility for rights clearance onto others apart from itself (judgment paragraph 148). After the main judgment there was a renewed dispute about whether PPL would be prepared to provide a list of UK licensed internet radio stations. At the hearing I directed that both sides write to PPL, and if a satisfactory resolution cannot be found then I will make further directions about that if need be. But I am far from persuaded that this difficulty justifies limiting the scope of the relief to the sample stations.

Therefore the effect of the judgment justifies injunctive relief in general terms, not limited to the samples.

In terms of the inquiry as to damages, it is a commonplace in intellectual property disputes that liability is decided by reference to particular instances of the activity alleged to infringe (i.e. samples) but that the scope of the damages inquiry is general. Now if, on the inquiry, instances emerge which raise issues which the main trial has not determined, then there are various ways forward. Experience shows that these problems are usually readily resolved by case management in the inquiry. It is not the law that detailed issues of liability not decided at the liability trial cannot be decided in the inquiry; the test is one of justice and convenience (see Norris J in Fabio Perini v LPC [2012] EWHC 911 (Ch) at paragraphs 44-51). If it is really necessary, then a distinct second liability trial could be conducted. In the present case I can see no justification at all for limiting the inquiry to the sample stations.

The second question relates to the terms of the injunction itself. An injunction is an equitable remedy and therefore a matter of discretion. When infringement of an intellectual property right has been established, an injunction in general terms usually follows, but there is no invariable rule. The fact it generally follows does not absolve the court from considering the facts and circumstances of a particular case. It is simply a reflection of the fact that the relevant circumstances usually warrant a general injunction.

Article 8 of the Info Soc Directive (2001/29/EC) provides that Member States must provide effective sanctions and remedies. The sanctions must be effective, proportionate and dissuasive. The IP Enforcement Directive (2004/48/EC) also contains relevant provisions. The measures, procedures and remedies provided must be fair and equitable (Art 3.1), they must be effective, proportionate and dissuasive (Art 3.2) and an injunction prohibiting continuation of the infringement is a remedy expressly provided for (Art 11). Two important aspects of the matter were addressed by Arnold J in EMI Records v British Sky Broadcasting [2013] EWHC 379 (Ch) at paragraph 90-95. They are that proportionality is something which must be taken into account; and that a fair balance has to be struck between the protection of intellectual property rights and the protection of the fundamental rights, such as the right to privacy and the right to freedom of expression protected by the ECHR and the Charter. Whether this amounts to a different approach from the one the court would have taken applying equitable principles anyway is not a matter I have to resolve.

There is a line of cases considering the difference between intellectual property injunctions drafted in general terms as opposed to specific terms, such as Coflexip v Stolt Comex [1999] FSR 473. These were addressed by Arnold J in Interflora v

Turning to the relief sought, the claimants seek an injunction drafted to restrain the defendant from infringing the claimants’ copyright. One immediate difficulty with that was that the claimants’ repertoire was not defined. The claimants resisted having to do that. I do not agree. There is a close analogy between this case and the cases brought by collecting societies such as PPL and PRS against defendants such as retailers and public house owners who do not have a licence to play the works in public. Once infringement has been established based on samples, a general injunction to restrain infringement of the collecting societies’ rights is granted but it is limited to a defined repertoire. Crucially, the collecting society also gives an undertaking to maintain a searchable database of its repertoire on the internet and to keep it reasonably up to date. The collecting society also undertakes to answer reasonable questions about particular sound recordings. This strikes a fair and proportionate balance between the rights of the copyright owners and the position of the infringer. Notably the rights in issue in this case are likely to be covered by these sorts of injunctions when PPL takes action because Warner and Sony are members of PPL.

So if a general injunction is to be granted, it must include a similar regime relating to the repertoire, for the same reasons.

The injunctive relief the defendant contended should be granted was very different from the one sought by the claimants. First the defendant contended that no decision should be made until after any appeal. (Both sides sought permission to appeal.) At the hearing I gave permission to appeal and also indicated that I was not prepared to put off a decision on the final form of the injunction because I believe it should be decided now, but I was minded to grant a stay of the final injunction pending the appeal. The stay was appropriate considering the balance of convenience because any injunction would be disruptive to the defendant, whereas the claimants license their rights. I also took into account that this stay gives the defendant plenty of lead time so that, in the event the appeal is unsuccessful, it cannot then say it had no warning what the consequences would be. The time allows the defendant to start now to prepare for the final injunction to come into effect.

The other features of the injunction the defendant contended for were that it was limited to sample stations, and that it would contain a term which provided that it was complied with if the defendant implemented a geo-restriction system which blocked access to the sample stations for a given user based on the defendant’s geolocation provider supplying data that the user was located in the UK.

Although I have addressed the sample stations already, that was under the rubric of assessing the scope and effect of the main judgment. That means a wide injunction is open to me, but I have not yet exercised the relevant discretion to decide what sort of injunction should be granted.

The defendant submitted that the facts of this case were similar to the parallel import case referred to in paragraph 24 of Interflora, with streams lawfully on the market in one state (in Category 3) being “imported” into the UK. I see that analogy to some extent but it is not sufficiently close to work in the defendant’s favour. In the Oracle case ([2012] UKSC 27) Arnold J was considering (at paragraph 23) that each individual parallel imported pack, with its serial number, might have to be checked by the importer. That is nothing like the task the defendant has to perform in order to respect the claimants’ rights.

The defendant relied on the evidence of Mr Jaquette to establish how difficult it would be for the defendant in practice to comply with an injunction drafted in general terms. Therefore, he argued, the injunction would operate to restrict the defendant’s users’ freedom of information and right of access to information and ideas; restrict the station broadcasters’ freedom of expression and freedom of business; and would restrict the defendant’s own freedom of expression and freedom of business. I was not convinced by this because it starts from a number of false premises. First, Mr Jaquette’s approach amounts essentially to the defendant continuing to operate as it does, trying to shift the onus onto others to clear its rights and relying on geo-blocking to prevent access to a limited set of internet radio stations which have been notified to it. That will not do. The defendant has been found to infringe on a large scale and its current approach to rights clearance has been inadequate. Having been found to infringe in this way, it is only right that the defendant should now take some responsibility for its actions and take positive steps to ensure it is not committing acts of infringement. Furthermore there was clear evidence that the defendant’s approach to geo-blocking does not work with sufficient reliability.

Second much of the defendant’s argument depends on drawing an analogy between the defendant’s business and the business of an internet service provider, referring to the UPC Telekabel case (C-314-12). However the defendant is not in that position. The defendant has carried out primary acts of copyright infringement on a large scale.

Part of the defendant’s approach is to try to set up a notice and takedown regime putting the onus on the rights holder to notify the defendant of unlicensed content. If the defendant was an ISP then that would be a very different matter but it is not. I agree that the claimants ought to provide proper notice of their repertoire and I have addressed that above, but that is all. If the defendant wants to target a category 3 internet radio station playing music in the claimants’ repertoire to users in the UK, then the defendant ought to take proper steps to ensure that there is a relevant licence of the UK rights in existence. I seriously doubt that would make its business unprofitable, but if it did then so be it. The right of freedom of business is not a right of freedom to conduct an infringing business.

Third, the fact that a given internet radio station is likely to broadcast some content which does not require a licence under UK copyright even if it includes other material which does, does not mean that the injunction sought by the claimants fails to strike a fair balance between the claimants’ rights and the rights of others (the defendant, the broadcaster and users). It is the defendant who chooses to make the internet radio station available to UK users. The UK is not a minor or incidental part of the defendant’s business. The claimants’ sound recordings are not a minor nor an incidental part of what the defendant makes available to UK users. The defendant can take proper steps to establish that the internet radio station is licensed under UK copyright. To repeat a finding I made at trial, I am not persuaded that the number of stations the defendant deals with is so large as to make this at all impractical.

Fourth, the rights the defendant has infringed are available to be licensed in the UK.

It may not be a case in which the defendant has a legally enforceable right to a licence, such as exists in the FRAND cases, but that does not matter. The claimants are willing licensors. The simplest way of avoiding any difficulty for the defendant is to take a licence. If the terms cannot be agreed then there are numerous ways in which willing parties can resolve that matter. I raised this at the hearing. The matter could be settled by arbitration or even by the court if the parties wished, possibly using the mechanism of s93 of the Arbitration Act 1996, which was amended in February this year to allow any High Court judge to sit as an arbitrator subject to approval of the LCJ.

Applying the principles, in my judgment an injunction limited to the sample radio stations would not be effective, proportionate or dissuasive. It would not equiparate with the scale of the defendant’s infringing activity. Nor would such an injunction strike a fair balance between the claimants’ rights and the various fundamental rights relied on by the defendant.

The just and proportionate injunction to grant following the finding in the main judgment is a general injunction to restrain copyright infringement with a provision providing reasonable clarity about the claimants’ repertoire.

There was no broad dispute in relation to the applicable principles. The applicable rule is CPR rule 44.2. The general rule is that the unsuccessful party will be ordered to pay the costs of the successful party, but the court may make a different order. In considering what to do, the court will consider all the circumstances, including the conduct of the parties, whether a party has succeeded on part of its case, even if that party has not been wholly successful, and any admissible offers to settle. That is r 44.2(4).

One point of dispute relating to principles is whether the economic value of a particular point, which in this case is one the unsuccessful party succeeded on, is something to be taken into account. Mr. Cullen submitted that while it was relevant when one came to the assessments that might take place under rule 44.3 or 44.4, it was of no relevance at the stage of applying rule 44.2. Mr. Howe submitted that that was wrong. He pointed out that rule 44.2 specifically provides that all the circumstances can be taken into account. It also provides expressly that the fact a party had succeeded on part of its case, even if it was not wholly successful, is something the court can properly take into account. I accept Mr. Howe's submission. The exercise is concerned with costs but that does not mean the court has to be blind to the importance of an issue on which a party, albeit successful overall, has lost. “All the circumstances” allows the court to take into account the relative importance of different points in the case when a party has won some and lost on others. Their economic significance is an aspect of that. The fact the rule expressly refers to partial success reinforces the point.

Turning to the case, stated very broadly, my conclusion was that the claimants, Warner and Sony, succeeded on the large majority of issues, if one counts issues.

Both sides accept that for that conclusion to be fairly reflected in costs means that some deduction should be made. Although Mr. Cullen did submit to me that he could have contended that no deduction should be made, he realistically made his case on the basis of some deduction. He contended it should be a modest deduction -- he put it at about 5% -- based on various analyses of counting pages and paragraphs of various documents in the case that the solicitors instructing him had undertaken.

On the other hand, Mr. Howe submitted that bearing in mind the value of the category 1 stations to the defendant, and therefore necessarily I suppose imagining if damages or a licence fee were paid, what the value might have been to the claimants had they succeeded, it was extremely relevant to take into account that a weighted average of approximately 67% of the listening hours in the UK via TuneIn are of category 1 stations. Therefore that tells you something very significant about the economic value of the issue about category 1 stations, which is one on which TuneIn succeeded. Therefore Mr. Howe submits that I should deprive the successful party, the claimants, of 75% of their costs.

So the parameters are 5% against 75% as a deduction from the claimants’ costs.

I should have said of course that the court always starts by examining who is the successful party. There was not really any argument about this, but I should identify that the claimants are the successful party in these proceedings. The costs should proceed from that basis.

Four other factors are advanced by Mr. Howe which he submits I should take into account. There is a point on the pre-action protocols because the letter before action did not mention category 1 stations, although Mr. Cullen submits that in fact they were in there, at least implicitly, in relation to the Pro App. That was a departure from good practice, submits Mr Howe.

Secondly, although the way the major issues have fallen out is the way I just described them, if one examines the case at a more detailed level of granularity there were other issues on which the defendant did succeed, and Mr Howe submits that is something that the court should have in mind when making its decision.

Third, in relation to the Pro App, in fact 80% of the usage of the Pro App is not music anyway, according to the defendant. Therefore the significance of the Pro App as a counterweight to the point made about category 1 is less than the claimants submit it is.

Fourthly, in relation to offers to settle, the defendant made without prejudice save as to costs offers to settle but it is not suggested that they were beaten. However the defendant submits, as a matter of conduct which the court should take into account, that in the course of the correspondence about settlement discussions, the claimants placed preconditions on negotiations, and that is unhelpful. Therefore it is said to be a

Mr. Cullen submits that the pre-action protocol point, which would be a minor one, is even less significant than it might otherwise be bearing in mind that it is about category 1 and he is accepting that some discount for category 1 needs to be made. I think Mr. Cullen is right about that. If things had been different, that might have led to a different answer, but as it has turned out I think that is right.

As far as who won or lost some of the detailed issues, I do not believe that examining that is going to lead to any significant change in the result and I will not embark on it. I will bear in mind the defendant’s point about the significance of the Pro App.

As far as the preconditions are concerned, Mr. Cullen made his submissions firmly on the basis that the defendant had not beaten the offers it made, and therefore these were irrelevant. I do not think that is quite the right analysis. In my judgment it is legitimate for a party to point out under the rubric of conduct, rather than necessarily under the rubric of beating or not beating offers, what exactly took place. I believe the court can, if it thinks appropriate, mark a concern about placing preconditions in the way of negotiations by taking it into account when making orders as to costs. It is not helpful in trying to settle cases for parties to refuse to even meet for negotiations by setting preconditions on those meetings, even if they ultimately win the case. I am going to take that into account.

Another matter which was advanced by the defendant related to an application in December last year which the defendant contended it should have the costs of. Whether or not the defendant is right about that, the total amount of costs that that application relates to is less than 1% of the overall costs which have been incurred by the claimants in these proceedings. The claimants’ total costs were about £1.4 million. Since I am trying to decide between one party who says I should apply a discount of 5% off the total and the other party who says it should be 75%, the margin of error in my decision is inevitably going to be much larger than the quantity of costs that that application related to. Accordingly, I will not take it into consideration as a separate matter.

Standing back, if I were simply going to analyse the matter on the basis of the amount of time attributable to the extra costs associated with category 1, then Mr. Cullen's submission is plausible and realistic, and I would assess a figure which reflected that at about 5%. The real issue is whether to take that approach or whether to increase the percentage for the reasons advanced by the defendant. The point is that this issue, although it may not have generated a significant marginal increase in the time, actually was of considerable commercial significance. Also, one needs to take a slightly different view, says Mr. Howe, of the way the issues go, because the underlying question of the correct approach in law under GS Media (Case C-160/15), was just as relevant to category 1 as it was to categories 2 and 3. Therefore if one asks the question a different way and seeks to identify the costs relating to category 1, a bigger proportion of the costs were relevant to that than just the extra costs. If category 1 had very little economic importance, one might say the right thing to do was just examine the extra costs it cost. However, when category 1 has such economic importance then perhaps that is not the only way to look at it.

Mr. Howe also submits that if the claimants had been prepared to accept that the defendant was right about category 1 in the first place, then the way in which the arguments would have been advanced would have been different and less costly, because one would then have been starting from an agreed position on the application of the law, at least in relation to category 1, and one could examine how that differed in relation to categories 2, 3 and 4. He argues that this is another matter to take into account.

Standing back, I am of the view that I should deprive the claimants of a higher proportion of their costs than 5%, for the reasons advanced by Mr Howe, but I am quite sure that to deprive them of 75% would be grossly unfair and bear no relation to the reality of the way this case has gone, taking account of all these matters and including the importance of category 1.

Doing the best I can in the circumstances, it seems to me that a fair, proportionate and reasonable deduction, taking all this into account, is 25%. That is what I will do.