Research In Motion UK Ltd v Visto Corporation

On 28^th February 2008 I handed down my judgment following the trial of this action. The issues decided at trial concerned the validity of Visto’s 905 patent and its infringement by a product called the Blackberry Mail Connector. RIM was the overall winner of the issues decided at trial, because it succeeded in revoking the 905 patent on the grounds of obviousness and because it related to a computer program as such. Other attacks launched by RIM, added matter and insufficiency failed.

Visto did however win on the issue of infringement: on the view which I took of the construction of the claims, which was a broad one, there was infringement. However Visto advanced a narrower construction which I rejected. Had I accepted Visto’s narrower construction, Visto would have failed on the issue of infringement. There was however no true “squeeze” between infringement and validity: the principal item of prior art, the Lotus Domino Defence had the feature brought in by the narrower construction of the claim.

Earlier in the proceedings two RIM patents had been attacked by Visto. But (see my judgment at paragraph 3) those patents were no longer in issue by the time of the trial because RIM had consented to their revocation. So Visto won on two out of the three patents in dispute.

There was also an action for a declaration of non-infringement (“DNI”) which RIM discontinued in April 2007.

The case now comes before me now to decide what costs order should be made.

The approach to costs orders under the CPR is set out in CPR 44.3. That approach allows for an issue-based costs order in appropriate cases. In patent cases, an issue-based approach to costs is now the norm: see Smithkline Beecham v Apotex [2004] EWCA Civ 1703; Actavis v Merck [2007] EWHC 1625 (Pat); Generics v Lundbeck [2007] EWHC 1606 (Pat) and Monsanto Technology v Cargill International [2007] EWHC 3113 (Pat).

Where the court decides to disallow the costs of a successful party in relation to a particular issue, it can sometimes go further and order the successful party to pay the unsuccessful party’s costs of that issue. It is not necessary to find unreasonable conduct on the part of the successful party in order to make such an order. In Monsanto v Cargill Pumfrey LJ (in a case where he was the trial judge) approved of a submission in the form of three questions asked by counsel in Actavis v Merck as demonstrating the correct approach:

“Have you won, have you lost a suitably circumscribed issue, so you should be deprived of your costs [of that issue], and is this an exceptional case such as to lead to an adverse costs order on an issue in favour of the overall loser?”

That seems to me to be a concise statement of the correct approach.

When the court takes an issue-based approach and concludes that the case justifies an award of costs on an issue against the overall winner, it is generally convenient to allow for those costs by making an appropriate percentage deduction from the winning party’s costs which are to be assessed (“the percentage set-off approach”). The advantage of proceeding via the percentage set-off approach is that it avoids the need to carry out an assessment of both parties’ costs. That in itself can be a considerable saving. A situation where it is easy to proceed in this way is where both parties overall costs are similar, and both parties estimate that they have spent a similar percentage on the issue. To give an illustration, if the winning party has lost on an issue which accounted for 10% of its costs, and it is concluded that it should pay the losing party’s costs, then one can safely deduct 20% from the winning party’s costs. Even if the party’s estimates are different as to the percentage spent on the issue, it may still be appropriate to take this course, by deducting the sum of the winner’s estimate and the loser’s estimate. The parties may in fact have spent different percentages on the same issue: on some issues one party may need to make more of the running than the other. As Pumfrey LJ said in Monsanto v Cargill at [5] the court is not in a position to adjust the parties’ estimates unless it has a really good reason to do so.

There are, however, cases where the percentage set-off approach gives rise to difficulties. The simple case discussed above assumes that both parties’ costs are the same. In that case one can allow for the unsuccessful party’s costs by a percentage deduction because X% of the loser’s costs are the same absolute amount as X% of the winner’s. If there is a disparity in the parties’ costs the arithmetic may become more complicated. One way of proceeding in an appropriate case would be to apply a conversion factor to the loser’s percentage. That approach may be appropriate where there is no reason to suppose that one side’s costs will be assessed down more severely than the other.

There is no dispute that RIM are the overall winners in relation to 905. They succeeded in revoking the patent. So there was nothing to infringe.

There is no dispute that RIM should pay the costs of that part of the action which concerned the RIM patents which it did not ultimately defend.

There is also no dispute that RIM’s costs should be reduced by a percentage to reflect the fact that it lost on infringement, added matter and insufficiency.

Visto say that the court should go further and award Visto its costs of the infringement issue and added matter/insufficiency, although it does not press that order so strongly in relation to added matter and insufficiency if the court adopts its approach to percentages as discussed below.

Normally it would be convenient to approach all these costs by applying the percentage set-off approach. But for reasons which I will now explain I do not think that course will achieve a just result in the present case.

There is one very striking feature of this case. Visto’s estimated costs for the whole of these proceedings are some £1.6 million. Yet RIM’s costs are estimated at nearly £6 million.

This is a quite staggering disparity. But apart from the disparity, it is astonishing that a sum of the order of £6m can have been spent by RIM’s advisors in fighting these disputes. It is all the more astonishing because RIM have throughout this litigation sought to emphasise that the Blackberry Mail Connector, which is the subject of the allegation of infringement, was not important commercially, and indeed was in the process of being phased out.

The Court’s function at this stage is not to stop parties spending money on litigation: it has no power to prevent expenditure, although it will use its powers to ensure that the litigation is run in a way that does not cause parties to expend unnecessary sums: see per Laddie J in Merck & Co v Generics [2003] EWHC 2842 (Pat); [2004] RPC 31 at 82-93. The court does however have a function in preventing unnecessary and unreasonable costs being recovered by a party from its opponent. In countries where the recovery of litigation costs from an opponent is not provided for, these problems do not arise. But parties can be discouraged from enforcing or defending their rights if they face claims for unreasonable and unnecessary costs being allowed against them. This was recognised in the Gowers Review of Intellectual Property published in December 2006: see paragraph 3.21.

This case was not a particularly heavy patent action. There was no disclosure, and the technology was not of the most complex kind. The trial itself only lasted about 5 days.

How is it that one side (RIM) can have outspent its opponent (Visto) in the same action by a factor of 4 or 5 to 1? I should say straight away that there cannot be the slightest suggestion that Visto’s advisors have adopted an unduly parsimonious approach. Visto employed well known specialist city solicitors, and an expert witness of a very high calibre. Their witness statements and expert reports were thoroughly and conscientiously prepared; and Visto employed highly distinguished specialist leading and junior counsel. So we know that it was possible to run one side of this case through to the end of the trial for an expenditure of Visto’s order without cutting any corners. Mr Moss, Visto’s solicitor, says in his witness statement that he would expect his firm, which is a City firm, to be at the upper end of the charging spectrum for work of this kind. Yet, as he points out, RIM’s solicitors profit costs were £5.18 million as compared with his firm’s profit costs of £1 million. So the disparity, on that basis is 5 to 1. One must also bear in mind that Visto’s costs may be assessed down significantly as well.

I have been shown a costs breakdown prepared on behalf of RIM. Over the 15 months between claim form and trial one partner, three senior associates and three associates have been involved as well as an unspecified number of trainees and paralegals.

I have to say that the breakdown reveals some really shocking statistics. One of the senior associates has spent 2252 hours on this case in the course of 15 months. Another associate has spent 2291 hours. These hours each represent the best part of the working day for the whole 15 month period, assuming a 35 hour week. For just these two associates, the charges amount to nearly £2 million. For these sums of money one would be entitled to expect each of them to be able to recite all the documents in the case by heart.

The single partner has spent 1387 hours on the case. The remaining associates have spent 3500 hours or so between them. Perhaps most surprising of all is that another £1 million odd has gone on trainees and paralegals. They have spent over 5000 hours on the case. I was told they had been reading the results of prior art searches. This is surprising given that the prior art cited in this case was high profile material from Lotus Notes and Microsoft. A version of Lotus Notes had been relied on in the United States before this action started.

If one adds up all the hours spent by RIM’s solicitors, one finds that some 9 man years have been spent over 15 months. All for a trial with no disclosure which lasted about 5 days.

The picture summoned up by this bill of costs is one which is totally unfamiliar to anyone who has been involved in economically conducted patent litigation. Ms. Dagg, the partner at Allen & Overy with conduct of the case seeks to explain these high levels of costs by reference to the fact that the present dispute forms part of a global battle between RIM and Visto. In paragraph 23 of her witness statement she says:

“The press releases attached to the Particulars of Claim … make it clear that Visto’s overall goal was to shut down RIM’s Blackberry system entirely. Therefore, although it might be thought that this action was not particularly commercially important to RIM (the alleged infringement being the little used and now discontinued Mail Connector), it was an important part of a much larger global battle. Obtaining a timely decision of this Court on the validity of the 905 Patent was important for RIM in this overall battle”

I have to say that I have great difficulty with the logic of this paragraph. Insofar as the patent represented any threat at all to RIM’s business it has only been, so far as the evidence goes, a threat to the Mail Connector product which RIM have consistently maintained was of little significance. In any event Visto gave an acknowledgement of non-infringement in respect of the Blackberry system as a whole (omitting to mention the Mail Connector) some two months or so after it was requested in early December 2006. The reason proffered by RIM as to why the Mail Connector was not mentioned was that it was:

“of commercial insignificance to RIM in the UK as well as in other jurisdictions”

Mr Watson QC, who appeared for RIM, faintly suggested that the reason for needing to pursue the proceedings so vigorously was to preserve the position for the future with regard to other uses of HTTP to traverse firewalls. There was no hint of any of that in the evidence, either at trial or on the application before me.

Ms Dagg explains that her instructions were to leave “no reasonable stone unturned”. Beyond this, there is not really any explanation of what the extra hours spent by RIM’s solicitors were occupied with. Certainly there is no material before me to lead me to suppose that RIM can expect to recover as reasonably incurred costs anything significantly in excess of those which Visto found it necessary to spend.

Mr Watson at one point suggested that having made deductions from RIM’s costs, I should allow for any costs awarded to Visto by applying a conversion factor of ¼. So if Visto had spent 20% of their time on infringement, I could allow for that by deducting 5% from RIM’s costs. I have no hesitation in rejecting that approach in the present case. As Mr Watson I think accepted, if RIM’s costs are subsequently cut down to the level of Visto’s, then the conversion factor will have been shown to be unnecessary. The appropriate deduction from the RIM bill (reduced by assessment) would be 20%, not 5%.

Mr Carr QC, who appeared for Visto, said that I should approach this case on the basis that RIM’s costs, when subject to detailed assessment, will be cut down to the level of Visto’s costs. He submitted that I could apply a percentage set off approach to RIM’s costs, confident in the knowledge that those costs will be assessed down to levels similar to Visto’s. Or I could make an order now capping off RIM’s costs at some figure.

Mr Carr supported that submission as follows. He said that proportionality is fundamental to the approach to costs under the CPR. Moreover proportionality was not just a factor to be taken into account by a judge or costs judge when assessing costs: it was a factor to be taken into account when making the costs order.

Put in that broad way, those submissions are correct. They receive support from the judgment of Lord Woolf in Lownds v Home Office [2002] EWCA Civ 365; [2002] 1 WLR 2450 at [3] and [10]. Thus a court may deploy the notion of proportionality in deciding whether a party should recover costs of particular issues. That decision may depend, in part, on how much is as stake. Mr Carr sought however to take this further by referring to what Lord Woolf said at [31]:

“In other words what is required is a two-stage approach. There has to be a global approach and an item by item approach. The global approach will indicate whether the total sum claimed is or appears to be disproportionate having particular regard to the considerations which CPR r 44.5(3) states are relevant. If the costs as a whole are not disproportionate according to that test then all that is normally required is that each item should have been reasonably incurred and the cost for that item should be reasonable. If on the other hand the costs as a whole appear disproportionate then the court will want to be satisfied that the work in relation to each item was necessary and, if necessary, that the cost of the item is reasonable. If, because of lack of planning or due to other causes, the global costs are disproportionately high, then the requirement that the costs should be proportionate means that no more should be payable than would have been payable if the litigation had been conducted in a proportionate manner. This in turn means that reasonable costs will only be recovered of the items which were necessary if the litigation had been conducted in a proportionate manner.”

I think it is clear that in referring to the two-stage approach, Lord Woolf was referring to the assessment of costs. Nevertheless it is right to say that in deciding what order to make about costs, considerations of proportionality are important. Moreover a trial Judge who sees a figure for costs before the process of assessment is embarked on by a costs judge is in a good position in many cases to express a view as to whether the bill represents disproportionate expenditure, and is clearly entitled to express that view for the benefit of the costs judge.

Mr Carr also referred me to the decision of the Court of Appeal on an appeal from the judgement of Lewison J. on costs in Ultraframe (UK) Ltd v Fielding (costs) [2006] EWCA Civ 1660. In that case the claimants had expended some £2.5 million in costs in order to prove a dishonest conspiracy in respect of which it had suffered no loss. Lewison J. had held that to be a disproportionate expenditure on costs for such a limited result. He allowed the claimant 10% of its overall costs to reflect a proportionate approach to the proof of that conspiracy, which he reflected in a deduction from the winning party’s (the Burnden defendants’) costs. It appears that this resulting figure of £250,000 was the same as had been spent in an earlier action in which the conspiracy allegations were made, the costs of which were being dealt with by the judge in this action. So, despite the fact that there had been no loss, there was a live issue as to £250,000 worth of costs of the earlier action.

At paragraph 9 of the judgment of the Court of Appeal, Waller LJ described the judge's order as "capping the costs that Ultraframe would be entitled to recover at £250,000". At paragraph 36 Waller LJ said this:

"On that question the judge took the view that Ultraframe's expenditure was disproportionate. He would only have allowed expenditure of £250,000 as compared with the £2.5 million Ultraframe say that they incurred. The effect of the judge’s order, as now properly understood, is likely to lead to the costs, which the Burnden defendants would otherwise have been entitled to being reduced by significantly more than £250,000. There is no general rule that the losing party who can establish dishonesty must receive all his costs establishing that dishonesty, however disproportionate they may be. Indeed cases in which no order for costs has been made demonstrate there is no such rule. Proportionality and conduct of the paying party is clearly something to which a trial judge is entitled to have regard.”

Mr Carr submits that these passages show approval by the Court of Appeal to a process of capping a party’s costs on the grounds of proportionality at the stage of making the costs order. I agree.

Mr Watson submitted that the case only established that costs could be refused where they had been expended on a pointless issue. Where as here the relevant costs had been expended in successful pursuit of the revocation of a patent it would be wrong to cut down the costs summarily in the way Lewison J proposed.

I do not think that the power to cut down costs on the grounds of proportionality at the costs order stage is restricted in the way suggested by Mr Watson. But Mr Watson is right to draw attention to the distinction between a pointless issue and one where there is a limited commercial purpose. That is a factor which the court will bear in mind in any given case. Moreover, in cases where the court can see precisely what the financial value of success has been, it may be much easier to form the view not only that the costs are disproportionate, but also what a proportionate expenditure would be. One can see that principle illustrated in Home Office v Lownds where costs of the order of many times the sum in issue were expended. In Ultraframe, the Judge’s view was plainly that spending more than £250,000 in pursuit of an equivalent sum was disproportionate. By contrast it is not so easy for a trial judge to see precisely what the value of a claim to revoke a patent will be.

I am however entirely satisfied that the sums expended by RIM are disproportionate in relation to what was at stake here. RIM’s repeated description of the issues as having no real significance commercially mean that there is no justification for costs rising above what would commonly be expected for an action of this nature.

Whilst it undoubtedly has attractions, I think that there are dangers in the present case in accepting Mr Carr’s invitation to assume or declare that RIM’s costs will be no more than Visto’s. Whilst I have concluded that, viewed globally, RIM’s costs are disproportionately high, I do not feel that I am in a position to form a sufficiently precise view of where, between stratospheric heights of RIM’s actual costs and those of Visto, a proportionate expenditure on the part of RIM would fall. It is one thing to be able to say that a party’s costs are disproportionate, it is entirely another to be able to specify at precisely what level the costs would become proportionate. Suppose that RIM’s costs are only assessed down by 50%. In those circumstances it would not be fair to allow for Visto’s costs on the assumption that both sides costs were the same. For the same reason I do not consider it would be just to take the approach suggested by Mr Carr of capping RIM’s costs at the level of Visto’s costs.

It follows that with some regret I shall not be able to apply the percentage set-off approach in the present case. Nevertheless I direct the Costs Judge in this case to proceed with the assessment on an item by item basis and on the basis that items claimed by RIM are to be allowed only if necessary and reasonable. In this connection I should make it clear that I do not regard endless searching for prior art as necessary. Searching is an activity which experience shows can be endless and usually brings diminishing returns. I see no reason why Visto should pay for prior art searches except to the limited extent that they uncovered cited prior art not known to RIM when they started the action.

I have already recorded the fact that it is common ground that the issue of infringement is sufficiently circumscribed issue for its costs to be dealt with separately. Is this a suitably exceptional case for an order requiring RIM to pay Visto’s costs of that issue? Or is it enough if RIM simply do not recover their own costs?

Mr Carr relies on RIM's failure to produce an adequate PPD, the number of non-infringement defences unsuccessfully run by RIM and the fact that RIM failed to make any proper admissions even on issues that which were not contentious. This included insisting on seeing repeats of experiments even though there was ultimately no dispute as to what those experiments showed.

I think RIM's conduct in relation to the issue of infringement does deserve to be marked by an award of costs in Visto's favour. Equally I think Visto are entitled to their costs of added matter and insufficiency. The former, at least, took up some considerable time at the trial.

RIM's and Visto's solicitors’ estimates of the amounts deductible are as follows.

These estimates are on different bases. RIM’s on the basis of their 905 costs alone; Visto’s on the basis of total costs.

I regard it as preferable to express the percentages on the basis of total costs, as this will greatly simplify the assessment.

RIM’s estimate of the proportion of its total costs attributable to the RIM patents is 20%. Its total costs also include 10% attributable to liaising with foreign advisors. It follows that the 905 costs are 70% of the total. So the 20% deduction from the 905 costs is a 14% deduction from the total.

RIM should recover, on their figures, their total costs, less 20% for the RIM patents and a further 14% for the 905 deductions i.e. 66% of their assessed costs.

I see no reason why I should not adopt Visto’s figures for the costs which they are entitled to on the RIM patents, infringement, added matter and insufficiency and the DNI. Visto allot 21.6% of total costs to RIM patents. Coupled to the costs I consider they should recover from RIM, these add up to 50.7%, but I will round it up to 51%.

The order will be

I should record that RIM requested an interim payment on account of nearly £1 million. This sum is of the order of Visto’s entire profit costs of the actions.

If RIM’s bill is assessed down to the level of Visto’s then, on the order I am making, RIM will recover 15% of its assessed costs. The most I would have been prepared to contemplate by way of interim payment would have been 7.5%, to take account of the need to be cautious in ordering an interim payment.

In a normal case the effect of an interim payment is that the amount in dispute is reduced, with the result that a detailed assessment can be avoided by settlement. It seems to me that if RIM are seriously pursuing their bill on the scale presented to me, then the parties will be so far apart that the fact of an interim payment will not have this effect at all. Although parties should not be kept out of their money for longer than necessary, no factor suggesting that RIM would suffer any particular hardship by having to wait for the result of the assessment was relied on. I am comforted also by the fact that, had RIM’s costs not been so obviously disproportionate, I would have been able to apply the percentage set off approach with the likelihood that I would have ordered an interim payment.

In the unusual circumstances of this case I decline to make any order for an interim payment.