Geostel Vision Ltd v Oraka Technologies Ltd

IN THE COURT OF APPEAL OF NEW ZEALAND

I TE KŌTI PĪRA O AOTEAROA

CA254/2018
[2020] NZCA 256

BETWEEN	GEOSTEL VISION LIMITED First Appellant
AND	PAUL DAYNES AND THE ESTATE OF GORDON ROBERTSON Second Appellant
AND	ORAKA TECHNOLOGIES LIMITED First Respondent
AND	ORAKA GRADERS LIMITED Second Respondent
AND	MICHAEL WILLIAM SCHWARZ Third Respondent

CA261/2018
BETWEEN	NAPIER TOOL & DIE LIMITED Appellant
AND	ORAKA TECHNOLOGIES LIMITED First Respondent
AND	ORAKA GRADERS LIMITED Second Respondent
AND	MICHAEL WILLIAM SCHWARZ Third Respondent

Hearing:	2 March 2020
Court:	Courtney, Ellis and Brewer JJ
Counsel:	K T Glover and P R Casey for Appellants in CA254/2018 P G Skelton QC and C D Herbert for Appellants in CA261/2018 B P Henry and A R Kenwright for Respondents
Judgment:	25 June 2020 at 11.30 am

JUDGMENT OF THE COURT

AThe appeal is allowed. The High Court’s damages award of $510,000 is quashed and we substitute damages in the sum of $47,000.
BThe costs award in Technologies’ favour in relation to the 2017 High Court hearing is set aside, with the issue of costs for that hearing remitted back to the High Court to be reassessed in light of this Court’s judgment.
CThe appellants are entitled to costs in this Court for a standard appeal on a band A basis plus any disbursements.

____________________________________________________________________

Para No.
BACKGROUND
Facts	[6]
Background to the infringement	[11]
Procedural history	[20]
THE USER PRINCIPLE	[21]
THE SECOND DAMAGES HEARING	[23]
The plaintiffs’ evidence	[24]
Mr Schwarz	[25]
Mr Black	[26]
Mr Davies	[30]
The defendants’ evidence	[35]
Mr Beach	[36]
Mr Yorke	[43]
Mr Hussey	[55]
THE HIGH COURT DECISION	[63]
Approach	[64]
Factual findings	[66]
The damages calculation	[67]
Interest and costs	[75]
THE ISSUES ON APPEAL	[76]
APPROPRIATE APPROACH ON APPEAL	[78]
WAS THE HIGH COURT WRONG TO FIND THAT THE REASONABLE LICENCE FEE WAS $6.00 PER CUP ASSEMBLY?
Was Napier the appropriate notional licensee?	[84]
Discussion	[92]
Exclusivity	[103]
Discussion	[106]
Weight given to the relative bargaining power of the notional licensor and licensee	[109]
Alternatives available to the parties	[110]
Competition in the market for asparagus grading machines	[124]
The summary judgment against Technologies and its state of insolvency	[132]
The level of originality in the cup assembly design and its overall importance to the grading machine	[136]
What is the reasonable licence fee payable in respect of each infringing use?	[138]
Preliminary observation	[141]
Disregarding evidence about comparable licence fees	[145]
Calculations based on Geostel’s anticipated – rather than actual – profits	[148]
Rejection of the 25 per cent rule	[150]
Conclusions	[153]
Result	[157]

REASONS OF THE COURT

(Given by Ellis J)

Following appeals to this Court in 2010 and again in 2013, Napier Tool & Die Limited (Napier), Geostel Vision Limited (Geostel) and Geostel’s owners (Mr Paul Daynes and Mr Gordon Robertson[1]) were found to have infringed the copyright in the designs for a cup assembly used in a machine which sorts and grades asparagus.[2] Put simply, the infringing cup assemblies based on the designs had been made by Napier for Geostel.
[1]Mr Robertson has since died.
[2]Oraka Technologies Ltd v Geostel Vision Ltd [2010] NZCA 232, (2010) 88 IPR 227 [2010 CA decision]; Oraka Technologies Ltd v Geostel Vision Ltd [2013] NZCA 111 [2013 CA decision].
The protected designs were authored by Mr Michael Schwarz, who had also designed the grading machine in which the cup assemblies were first used (the Oraka Grader). At the time of infringement, the copyright in the designs for the cup assemblies belonged to Mr Schwarz’s company, Oraka Technologies Ltd (Technologies). At that time, however, the business relating to the Oraka Graders and the cup assemblies was conducted by another company run (but not owned) by Mr Schwarz, Oraka Graders Limited (Graders).
Following its finding of liability in 2013, this Court referred the matter back to the High Court for an inquiry into damages. In 2016 the High Court awarded Technologies $4.1 million in damages (including interest).[3] That amount constituted the loss of profits suffered by Graders as a result of the copyright infringement.
[3]Oraka Technologies Ltd v Geostel Vision Ltd [2016] NZHC 1188 [2016 HC Decision].
But on a further appeal to this Court in 2016 it was held that damages suffered by a third party could not be awarded for a copyright breach.[4] Instead, they were to be measured on the basis of a notional licence fee — the amount that the parties would have agreed should be paid in return for permission to use the designs if they had negotiated at the time of the first breach. The matter was once more referred back to the High Court so that damages could be determined on the basis of this “user principle”. In so doing, the Court noted counsel’s acceptance that the task was not a major one and that the relevant principles were well settled.[5]
[4]Napier Tool & Die Ltd v Oraka Technologies Ltd [2016] NZCA 554, [2017] 2 NZLR 611 [2016 CA decision].
[5]At [77].
Hinton J’s application of the user principle yielded a notional licence fee of $6.00 for each infringing cup assembly sold, giving a total damages award of $510,000.[6] As a result of the earlier liability finding, Napier, Geostel and Messrs Daynes and Robertson are jointly and severally liable to Technologies for that sum. In this further appeal they challenge the amount of this award.[7]

BACKGROUND

Facts

[6]Oraka Technologies Ltd v Geostel Vision Ltd [2018] NZHC 769, [2018] 131 IPR 363 [2018 HC decision].

[7]The respondents filed cross-appeals but advised in December 2019 that these would not be pursued.

We take our factual narrative largely from the earlier decisions of this Court, and the 2013 decision in particular. It is useful to begin with the Court’s description of the Oraka Grader itself:[8]
[7]       The Oraka grader is essentially a conveyor belt with a specialised cup assembly. For the conveyer belt the grader uses a standard, commercially available, chain made by Flextrak Enterprises Ltd (Flextrak). The cups each receive an individual asparagus spear and carry it under a camera/illumination station at the rate of about 12 spears per second.
[8]       The image of the individual spear is captured by a camera, processed by a computer and each spear is graded according to its length, colour and quality, depending on the demands of the market for which the crop is destined. When the asparagus spear reaches the designated collection chute for its grade, the cup assembly is electronically tripped to tip the cup. The spear falls into a collection chute where it is packaged with other spears of the programmed weight of spears into a chute specific to the grade.
[9]       At the beginning of the grading process, a small header belt pushes the asparagus spear lengthwise into the cup. The tip of the spear faces away from the chute so that when the cup trips and inclines and the spear slides out, the butt end of the spear impacts on the chute door. When the computer registers that the right number of spears are in the chute, the chute door opens and an asparagus bunch slides down to the bottom half of the chute ready for bundling.
[8]2013 CA decision, above n 2.
Mr Schwarz’s evidence was that he had worked on developing an asparagus grading machine since the early 1980s. After a long development period, a grader was up and running by early 1993. It was at this stage he approached Napier to have a mould for the cup assembly made.
Napier prepared preliminary drawings in February 1993 which Mr Schwarz approved. Napier then prepared tooling drawings based which it used to make a mould and, so, to begin manufacturing what has been termed the “Schwarz” cup assembly for Technologies. It is the February 1993 preliminary drawings that are the relevant copyright works.
By 1996, Technologies had run into financial difficulties. Mr Schwarz then formed a new company, Oraka Technologies Holdings Ltd (Holdings). Holdings was, from that point until 2001, used as the vehicle for any commercial activity associated with the Oraka Grader and the associated cup assemblies.
In 2001, Graders was incorporated by Mr Schwarz’s son and daughter. Graders (run by Mr Schwarz) took over the business previously run by Holdings. Today, Graders remains the entity responsible for any commercial activity associated with the Oraka Grader.

Background to the infringement

At around the time of Graders’ incorporation, Napier was approached by a Mr Andrew Armstrong, about the possibility of designing a competing cup assembly.[9] Mr Armstrong was associated with the Flextrak company which, as noted in the quotation at [6] above, was the manufacturer of the belt mechanism used in the Oraka grader.
[9]It was Mr Armstrong who had initially introduced Mr Schwarz to Napier, in around 1993. Napier had initially been a shareholder in Flextrak.
Flextrak had an arrangement with Mr Schwarz whereby Flextrak supplied Oraka Grader customers directly with replacement Schwarz cup assemblies for their grading machines. As the cups were subject to wear over time, there was an ongoing market for the provision of replacements. To facilitate supply of these, Mr Schwarz had given Mr Armstrong a customer list.
Mr Armstrong’s evidence was that he made the approach to Napier because he saw an opening in the market for a more robust cup assembly. He also had concerns about the financial position of Mr Schwarz’s companies and said there had been a number of complaints from customers regarding the after-market service they had received.
On 5 April 2001, Mr Armstrong and Napier’s Mr Witham met to discuss the details of the proposal. There were several pages of drawings made by Mr Armstrong and Mr Witham at the meeting. The High Court subsequently found that certain aspects of these drawings replicated aspects of the Schwarz cup assembly in sketch form, and that they contained references to the actual dimensions of the Schwarz cup.[10]
[10]Oraka Technologies Ltd v Geostel Vision Ltd HC Hamilton CIV 2005-419-809, 7 April 2011 [2011 HC decision].
After Napier had completed the design work for the competing cup, it referred Mr Armstrong to Mr Daynes who had recently severed his own business ties with Mr Schwarz. Mr Daynes’ company, Kamber Electronics Ltd (Kamber) had worked with the Oraka interests by supplying hardware and software packages associated with the electronic aspects of the Oraka Grader. Mr Daynes had developed much of the computer software incorporated in the grader.
Mr Armstrong met both Mr Daynes and his business partner, Mr Robertson, (formerly Mr Schwarz’s Greek representative) in Greece. Mr Daynes and Mr Robertson had already been working together on a new asparagus grading machine to compete with the Oraka Grader. The outcome of a subsequent meeting back in New Zealand was that Mr Armstrong transferred the “competing cup” project to Mr Daynes and Mr Robertson. The idea was that the competing cups wold be able to be used on both the Daynes/Robertson grader as well as on the Oraka Grader.
Geostel was incorporated by Messrs Daynes and Robertson in July 2001.
In November 2001, before proceeding with manufacture of the competing cup assembly, written advice was obtained from a patent attorney. The advice was that there was no copyright breach as the two assemblies expressed the same concept, but in a different manner. That advice turned out to be wrong.
Napier began making the competing cup assembly for Geostel. Geostel began to sell both the Daynes/Robertson Grader and the cup assemblies to clients in Greece, Australia and the United States.

Procedural history

The claim for breach of copyright was first filed in 2005. Its progress over the subsequent 15 years can relevantly be summarised as follows:
(a)On 18 February 2009, Allan J dismissed the claim on the grounds that the plaintiffs (Technologies, Graders and Mr Schwarz) did not own the copyright in the cup assembly.[11]
[11]Oraka Technologies Ltd v Geostel Vision Ltd HC Hamilton CIV 2005-419-809, 18 February 2009 [2009 HC decision].
(b)The plaintiffs successfully appealed. This Court held that the appellants owned the copyright in the designs for the cup assembly and remitted the case back to the High Court to determine liability.[12]
[12]2010 CA decision, above n 2.
(c)On 7 April 2011, Allan J found that the defendants had not infringed the plaintiffs’ copyright in the designs for the cup assembly.[13]
[13][2011] HC decision, above n 10.
(d)Again, the plaintiffs successfully appealed. On 18 April 2013 this Court entered judgment for Technologies, saying:[14]
[14]2013 CA decision, above n 2 (footnotes omitted, emphasis added). The Court of Appeal did not specify which of the appellants was the owner of the copyright.
[146] …We have held that the respondents copied a number of elements that, while not functional in the true sense, nevertheless possessed low originality. These elements are still protected, although this protection is not great. In this case it was not one such element alone that was copied but a number and it is significant that both experts agreed that the Geostel cup appeared to be a second generation of the Schwarz cup. Further, there is the requisite degree of originality in the dimensions A, B and C and they are central enough to the cup assembly for the copying of those dimensions to be the copying of a substantial part of the first appellant’s work. Our conclusions regarding the copying of a number of elements of lower originality support this conclusion. We therefore hold that the respondents have infringed the copyright of the first appellant.
(e)The matter was remitted back to the High Court for an inquiry into damages.
(f)On 5 November 2013 the plaintiffs filed a memorandum in the High Court which (as we understand it) addressed the matters that would be in issue in that inquiry. Napier filed a memorandum dated 3 December 2013 indicating it was largely content with the plaintiffs’ articulation of the issues.
(g)A six-day damages hearing was scheduled to commence on 1 December 2014. The close of pleadings date was 13 June 2014.
(h)On 20 May 2014 the plaintiffs served particulars of damage adopting a “loss of business” methodology rather than a “loss of sales” methodology. The particulars characterised the claimed loss as “losses suffered by all three plaintiffs”.
(i)On 22 May 2014, Faire J scheduled a judicial settlement conference to take place on 1 August 2014.
(j)On 6 June 2014 Napier filed an application for particulars of lost sales, to which the plaintiffs filed a notice of opposition on 13 June 2014. The application was scheduled for hearing before Faire J.
(k)In a minute dated 24 July 2014, Faire J recorded that it was apparent to him from the judgments of Allan J in 2011 and the Court of Appeal in 2013 that Technologies was the only plaintiff to the damages claim. He vacated the judicial settlement conference and hearing dates, as well as the hearing of the application for particulars.
(l)The plaintiffs nonetheless maintained their position that because Graders and Mr Schwarz remained parties to the proceeding they were entitled to argue that they have suffered losses as a result of breaches of Technologies’ copyright. They sought to amend their pleadings to support such a damages claim.
(m)The amendment was opposed by Napier on the grounds that:
(i)Technologies was attempting to claim Graders’ lost revenue which involved an impermissible piercing of the corporate veil;
(ii)neither Graders nor Mr Schwarz had the benefit of a judgment against the defendants for breach of copyright and the inquiry remedy should be for Technologies alone;
(iii)Graders had no standing to sue because it was neither the owner of the copyright nor did it have a written exclusive licence;[15] and
[15]The Courts had found that Graders had been making the cup assemblies under an informal, oral licence from Technologies. But Napier argued that Graders itself could not sue for breach because ss 120 and 123 of the Copyright Act 1994 only allow the owner of the copyright or the holder of a written exclusive licence to do so.
(iv)in any event, the heads of loss sought were so radically different from that originally pleaded that they constituted fresh causes of action which were statute barred.[16]
[16]More particularly, Napier said that for a decade, the plaintiffs had pleaded that their claim was for either an account of profits or, in the alternative, damages for lost sales. By contrast, the proposed new pleading involved two different and novel heads of damage giving rise to new considerations and requiring a new factual inquiry.
(n)Napier also said that because Technologies was not trading at the time of the infringement, it had suffered no loss.
(o)On 12 May 2015, Toogood J issued a judgment in which he said it was not clear (without further investigation) that Graders and Mr Schwarz were precluded from pursuing an inquiry into their claims to recover losses flowing from the breaches.[17] He therefore permitted the plaintiffs to amend their statement of claim.
(p)Napier maintained its position that third party losses were not recoverable at the subsequent damages hearing in the High Court in 2016.[18] It did not contest the plaintiffs’ quantification of those losses.[19]
(q)The High Court Judge found that Technologies could recover the losses suffered by Graders (but not Mr Schwarz) on the basis that those losses were, in effect, also Technologies’ losses.[20] Damages totalling $4.1 million were awarded.
(r)Napier successfully appealed that decision.[21] This Court held that the general rule that compensation for loss can only be awarded to one who has suffered that loss applied and set aside the High Court award. The Court divided, however, on whether that should be the end of the matter:
(i)The majority was of the view that the claim should be remitted to the High Court for quantum to be retried on the basis of “a notional licence fee payable in respect of each infringing use”[22] during the infringement period.
(ii)The minority view was that Technologies should be held to its unequivocal election not to pursue an account of profits or a claim for a notional licence fee and the litigation should end there.[23]

THE USER PRINCIPLE

[17]Oraka Technologies Ltd v Geostel Vision Ltd [2015] NZHC 991.

[18]Geostel took no active part in the hearing.

[19]The amended statement of claim dated 18 January 2016 claimed losses totalling $4,453,661. That figure comprised Graders’ expected revenue of $3,603,027 and loss of salary for Mr Schwarz of $850,634.

[20]2016 HC decision, above n 3.

[21]2016 CA decision, above n 4.

[22]At [80].

[23]At [83]–[86].

Before turning to consider the second damages hearing in the High Court and the decision under appeal it is useful to say a little more about the nature of the exercise in which it was required to engage and the more specific features of the user principle itself. For that, there is no better stepping-off point than Lord Wilberforce’s judgment in General Tire and Rubber Co v Firestone Tyre and Rubber Co Ltd, from which we relevantly extrapolate the following:[24]
[24]General Tire and Rubber Co v Firestone Tyre and Rubber Co Ltd [1976] 93 RPC 197 at 211–215.
(a)Infringement of the kind presently in issue[25] constitutes a tort, for which the infringer is answerable in damages.
[25]General Tire was concerned with the assessment of damages for the breach of a patent, but there is no dispute that the principles from it apply equally to a case of copyright breach.
(b)As with any other tort the object of damages is to compensate for loss or injury and — insofar as economic torts are concerned — the measure of damages is to be, so far as possible, that sum of money which will put the injured party in the same position as he would have been in if he had not sustained the wrong.
(c)There are two essential principles in valuing such a claim for damages:
(i)the plaintiffs have the burden of proving their loss; and
(ii)damages should be liberally assessed but that the object is to compensate the plaintiffs and not punish the defendants.
(d)Although the fact-specific nature of the exercise makes it dangerous to generalise, the reported cases can be usefully grouped to exemplify the approaches of courts to two typical situations:
(i)first, where the intellectual property (IP) relates to an invention which is exploited by the patent owner to make products which are sold at a profit, the effect of an infringement will be to divert sales from the owner of the patent to the infringer and the measure of damages will then normally be the profit which would have been realised by the owner of the patent had the sales been made by him.
(ii)secondly, where the IP is exploited by its owner through the granting of licences for royalty payments, the measure of damages an infringer must pay will be the sums which he would have paid by way of royalty if he had obtained such a licence.
(e)In the latter kind of case, quantifying a hypothetical royalty can be difficult but, ultimately, will be a matter of evidence.
(f)Where there is evidence of the royalties paid by other users of the same invention as a result of a free bargain between the patentee and those users, the price so paid can — all other things being equal — generally be taken as being the price or royalty that presumably would have to be paid by the infringer (the “going rate” approach).
(g)In some cases where it is not possible to prove either that there is a normal rate of profit, or that there is a normal, or established, royalty, it is for the plaintiff to adduce evidence which will guide the court. This evidence may consist of:
(i)the practice, as regards royalty, in the relevant trade or in analogous trades; and/or
(ii)expert opinion expressed in publications or in the witness box; and/or
(iii)the profitability of the invention; and/or
(iv)any other factor on which the judge can decide the measure of loss.
(h)Since evidence of this kind is by its nature general and also probably hypothetical, it is unlikely to be of relevance, or if relevant, of weight, in the face of the more concrete and direct type of evidence referred to at (f), but:[26]
… there is no rule of law which prevents the court, even when it has evidence of licensing practice, from taking these more general considerations into account. The ultimate process is one of judicial estimation of the available indications.
(i)The “true principle” which covers both kinds of case (those where there are comparable licences, and those where there are not) is that where the right holder grants permission to make the infringing article at a fixed price, the court then takes the number of infringing articles, and multiplies that by the sum that would have had to be paid in order to make the manufacture of that article lawful, and that is the measure of the damage that has been done by the infringement.
[26]General Tire, above n 24, at 214.

It is, of course, the second class of case referred to in General Tire which involves resort to the user principle. And, as this Court recently noted in New Zealand National Party v Eight Mile Style, LLC (Eight Mile), the more specific principles developed by the subsequent English authorities in that kind of case have been usefully synthesised in Copinger and Skone James on Copyright as follows:[27]
[27]New Zealand National Party v Eight Mile Style, LLC [2018] NZCA 596, [2019] 2 NZLR 352 [Eight Mile] at [29], quoting Gillian Davies, Nicholas Caddick and Gwilym Harbottle (eds) Copinger and Skone James on Copyright (17th ed, Sweet & Maxwell, London, 2016) vol 1 at [21‑292] (footnotes omitted).
(i) The hypothesis is that the actual licensor and the actual infringer are willing to negotiate with each other as they are, with their strengths and weaknesses, in the market as it existed at the date of the infringement. Accordingly the task of the court is to assess the value of the use to the defendant, not to a hypothetical person.
(ii) However, any impecuniosity on the part of the notional licensee is to be disregarded, as are the personal characteristics of the parties (e.g. an easy-going or aggressive nature), as opposed to the objective factors with which they were faced. Such matters are not considered to provide any guidance as to what the right is worth.
(iii) The fact that one or both parties would not in fact have reached an agreement is irrelevant.
(iv) The terms of any notional licence must reflect the actual infringement. Accordingly, where only part of a copyright work has been infringed, the notional licence is a licence to carry out the infringing act and not a licence to use the whole of the copyright work. The period of the licence is the period of actual infringement. Where the infringer enjoyed exclusivity in practice, the notional licence should accord with the reality. Likewise, the licence should permit the infringer to contract with others on the terms on which it actually contracted.
(v) Where there has been nothing like an actual negotiation between the parties, it is reasonable for the court to look at the eventual outcome and to consider whether or not that is a useful guide to what the parties would have thought at the time of their hypothetical bargain.
(vi) The court can take into account other relevant factors, and in particular delay on the part of the claimant in asserting its rights.
(vii) It is relevant under this head that the defendant could have arrived at the same result by lawful means if the parties can be expected to have taken that fact into account in their hypothetical negotiation. That alternative need not have had all the advantages or other attributes of the infringement.
(viii) In the absence of comparable licences or other compelling evidence the royalty may be based on the “available profits” method: the defendant’s actual profits are calculated; it is assumed that the parties would have accurately predicted these profits when negotiating; the profits are then (in effect) divided between the claimant and the defendant.
(ix) In some cases it may be appropriate to award as damages the cost of producing or commissioning the material in a form which did not infringe copyright.
(x) In a case within this group, the court may have to call into play “inference, conjecture and the like”, and apply “a sound imagination and the practice of the broad axe”.

THE SECOND DAMAGES HEARING

Because of the nature of the points raised by the present appeal, it will be necessary to set out the evidence called by each side following the remittal back in some detail.

The plaintiffs’ evidence

The plaintiffs called evidence from Mr Schwarz, Mr William Black, and Mr Paul Davies (in reply).[28]

Mr Schwarz

[28]The evidence of Mr Caraccioli is discussed separately, and later, below.

Mr Schwarz gave evidence about matters he (as the effective notional licensor) regarded as relevant to the notional licence negotiation. By way of summary, he said:
(a)he would not have agreed to a licence;
(b)the cup assembly design was unique and was the key to an accurate high speed asparagus grader;
(c)Technologies was a one-product company, selling only grading machines (albeit that Graders was in fact the seller);
(d)there should be a royalty payable on the sales by Geostel of the grading machines incorporating the infringing cup mechanism
(e)the sale by Geostel of its grading machines would have derived a profit of greater than 50 per cent;
(f)the royalty payable in relation to the sales of the grading machines should equate to 50 per cent of the retail value of the grading machines sold because of the importance of the cup assemblies to those machines; and
(g)the royalty payable on the cup assemblies sold separately should be between 75 and 100 per cent of their retail price.

Mr Black

Mr Black is a chartered accountant, specialising in restructuring and insolvency. He had no specific IP licensing expertise.
After observing that the defendants appeared to have consciously set out “to enrich themselves at the expense of the plaintiffs” and that the market for the Oraka Grader was “significantly disrupted” by the copyright infringement, Mr Black said:
30. In setting a notional licence fee to apply in the circumstances of this case I am firmly of the view that applying an “industry standard” is not the correct approach. I note that one of the references cited at paragraph [74] of the Appeal Court judgement [sic] is to General Tire. In that case Lord Wilberforce stated that there could be “any other factor on which the judge can decide the measure of the loss”. He added that “the ultimate process is one of judicial estimation of the available indications”.
31. In my view the “other factors” and “available indications” that need to be taken into account and afforded due significance in this case are that a family business has been by all objective measures significantly damaged by the wrongful conduct of the defendants. This was a business with a sole product which had significant growth prospects until the infringing activity took place. The growth prospects of the business have been extensively articulated in the evidence of an expert, Mr Marnus Beylefeld, Chartered Accountant, in his Brief of Evidence dated 18 January 2016 in the earlier High Court damages enquiry.[29] I do not repeat Mr Beylefeld’s reasoning here but note that in my view this is a significant factor that needs to be taken into account in arriving at an appropriate notional royalty to apply to the infringing sales.
[29]Mr Beylefeld had calculated the loss to Graders, not Technologies.
Mr Black said notional licence fees of between 30 and 40 per cent of infringing sales were “not uncommon” and referred to an Australian paper which suggested that “damages awarded under the user principle have a restitutionary aspect”.[30]
[30]David Yates, Judge of the Federal Court of Australia “Recent developments in IP remedies” (30th Annual Conference of the Intellectual Property Society of Australia and New Zealand, Sydney, 9–11 September 2016). This proposition was rejected by Hinton J (2018 HC decision, above n 6, at [18]) and, subsequently, by this Court in Eight Mile, above n 27, at [39].
Ultimately, Mr Black calculated the value of Geostel’s sales of graders and cup assemblies during the infringing period and then applied the royalty percentages contended for by Mr Schwarz (without further specific analysis)[31] to those figures. He arrived at a damages sum of $5.1 million.[32]

Mr Davies

[31]Mr Black did seek to justify the application of the higher, 75 per cent, royalty to the cup assemblies by saying “[g]enerally speaking profit margins on consumables are significantly higher than profit margins on the underlying equipment/machinery in which they are used. For this reason it would be appropriate to apply a higher notional licence fee percentage than that which is applied to the grader machine sales.”

[32]As Mr Hussey (one of the defendants’ experts) noted this amount would have comprised a royalty of $5.06 per cup where they were sold separately (as replacements) and $127.31 per cup where they were sold as part of a grading machine.

Mr Paul Davies was Head of Intellectual Property at EverEdge Global (NZ) Limited (EverEdge). EverEdge provides business advice, particularly in the areas of assessing and monetising IP. Mr Davies was only called by the plaintiffs in reply to the evidence given for the defendants, including by Mr Yorke and Mr Beach (discussed later, below).
Mr Davies opined that the more likely notional licensee here was Geostel, rather than Napier (or both).
After echoing the views of Mr Schwarz and Mr Black that the licence fee should apply not only to the cups themselves but also to the graders sold by Geostel, Mr Davies said:
(a)it was most likely Technologies would not have granted a licence: “an opportunity denied in the circumstances”;
(b)the only notional licence fee likely to make the licensor “willing” was one that compensated for the entire loss of business “by the licensor”;
(c)a willing buyer would require a much lower licence fee which is often why intellectual property rights are infringed: “when caught the offender must take the consequences”;
(d)it is likely that the parties in this instance would have been unable to bridge the gap had there been a negotiation in 2001; and
(e)a true arm’s length negotiated position is unlikely to have resulted in this instance.
Then, he concluded:
73.      On that basis, in my opinion margins beyond those of normal manufacture of commodity products should attribute to the intellectual property in the solution. However there is no rational commercial basis for a licensing expert such as myself to provide an opinion as to the quantification of the notional license fee to be applied.
74.      The license fee as I understand the authorities is to provide the owner of the intellectual property with fair compensation for infringement of its rights notionally assessed not commercially assessed.
75.      The notional assessment in my opinion is a matter for the court not an expert, however the principles I have set out hopefully assist the court with what should in a commercial situation be considered. …
(Emphasis added).
Like Mr Schwarz and Mr Black, therefore, Mr Davies rejected, and did not attempt to apply, the orthodox “user principle” test of the hypothetical willing licensor/willing licensee.

The defendants’ evidence

The key witnesses called by the defendants were Mr Nigel Beach, Mr Scott Yorke and Mr Shane Hussey.

Mr Beach

Mr Beach had significant experience in licensing IP and had (until recently) headed a significant research and development team at Compac, a world-leading produce grading company.[33]
[33]Compac is an Auckland-based company which designs and builds machines to weigh and sort fruit, using software, computers, electronics and video cameras. Computer software is used to detect the colour, weight, blemishes, size and shape of fruit at high speed. Mr Beach’s evidence was that Compac is the number two supplier in the world of this type of machinery.
Mr Beach’s evidence was that, in his experience, the factors ordinarily relevant to Compac’s negotiation of a licence fee of this kind were:
(a)the territorial scope of the relevant IP protection;
(b)the strength of the IP protection; and
(c)the availability of alternatives, including developing a non-infringing alternative.
The point about territorial scope is that the copyright in Technologies’ designs is geographically very limited. Few countries in the world afford copyright protection to such designs; in this New Zealand is an outlier, internationally.[34] So Mr Beach’s evidence was that the amount paid for a licence to manufacture components such as the cup assemblies would have been around three to four per cent of gross sales, but only in the event of worldwide IP protection. Without such protection, he said that he would have offered a nominal licence fee (“substantially less” than three per cent of gross sales) to avoid the inconvenience of manufacturing overseas.
[34]It seems that the Schwarz interests were aware of this territorial limitation more or less from the outset. For example, in 2004, Graders sought and obtained preservation orders in this Court in relation to the tooling held by Napier in order to prevent the cup moulds from being sent overseas, because the cups could then be manufactured there without any copyright infringement.
In terms of the strength of protection, the point made by Mr Beach was that patents confer the strongest IP protection on the holder, followed by registered design rights, with copyright affording the lowest protection. So the fact that the licence here related to IP that was protected only by copyright meant that a lower notional licence fee should be arrived at.
And as for the other options open to Geostel in 2001, Mr Beach:
(a)disagreed with Mr Schwarz’s assertion that the cup assembly was novel in 2001, given its similarity to what Compac was doing with other types of produce and also based on his direct knowledge of cups for asparagus graders made in Australia at around the relevant time, known as GP Graders;
(b)said he did not think that it would have been difficult for Compac to produce asparagus-grading machines including cup assemblies for carrying asparagus spears; and
(c)agreed with Mr Daynes that, by comparison with the software, reliability and service the cup assembly was relatively unimportant; it was simply a functional transport system.
Mr Beach also gave evidence, based on his experience at Compac, about the possibility that a competing and non-infringing cup assembly could have been developed by the defendants. He said he was “confident” that the development process would take up to nine months overall, with the tooling completed after seven months followed by eight weeks for testing and “tweaking” the design.
In terms of cost, Mr Beach estimated $225,200 for the development process, which comprised:
(a)labour cost of $95,200;
(b)tooling cost of $80,000;[35] and
(c)test machine cost of $50,000 if a test machine was not already available.

Mr Yorke

[35]Napier had in fact charged Geostel $65,000 for tooling, which seems to confirm Mr Beach’s view that his time and costs estimates were “generous”.

Mr Yorke is a solicitor with specialist expertise in IP licensing. His evidence covered (inter alia):
(a)the identity of the appropriate notional licensee, which he thought it would be Napier, rather than Geostel;
(b)the most commonly used methods of determining a licence fee, and which method should be used in the present case;
(c)the factors that would influence any royalty rate set under the licence and the circumstances in which that royalty would normally be payable; and
(d)the appropriate rate for a notional licence fee.
In opining on these matters, Mr Yorke made a number of assumptions, none of which are now really in dispute. They were that:
(a)the notional licence related solely to the subsisting copyright under New Zealand law in the design drawings for the cup assemblies;
(b)Technologies was the owner of the copyright in those design drawings;
(c)the designs were protected by New Zealand copyright, but there was no enforceable copyright subsisting outside New Zealand;
(d)none of the plaintiffs have any other relevant IP rights;
(e)the relevant period for the notional licence is mid-2001 to March 2009;
(f)Napier manufactured the cup assembly components in New Zealand and supplied them to Geostel;
(g)Geostel did not itself manufacture the cup assembly components in New Zealand;[36]
(h)Geostel did not supply the cup assemblies to any New Zealand customer during the relevant period;
(i)Geostel supplied the cup assemblies outside New Zealand, either incorporated with grader machines or as spare parts for use with such machines; and
(j)Geostel had a standard price for supplying a set of 1000 cup assemblies as consumables.
[36]This is the one matter listed which requires further discussion, as to which see [92]–[98] below.
Mr Yorke’s evidence addressed the appropriate royalty rates on the alternative bases that Napier was the notional licensee and that Geostel was the notional licensee although, as we have said, his view was that it should be Napier.[37]
[37]He was also unclear whether the Court would regard the supply and sale of cup assemblies overseas as separate infringing acts, so again covered both alternatives in his evidence.
Mr Yorke noted that there are many ways of calculating royalties, with varying degrees of complexity and application. He said negotiating such fees is an art rather than a science and it would be dangerous to adhere strictly to any one methodology. In most cases, the best that can be achieved is to give the parties a starting point. He said the three most commonly used methodologies are:
(a)the cost approach;
(b)the comparable market approach; and
(c)the income approach.
Mr Yorke explained that under the cost approach the licence fee is set at a level aimed at either reimbursing the licensor for its costs in developing and protecting the relevant IP or at a level reflecting the costs the licensee would have incurred either in creating the relevant technology itself or in creating a non-infringing version.[38] The cost approach takes no account of the actual market value of the IP.
[38]He said sometimes the royalty will be set at cost plus an agreed margin.
The comparable market approach involves setting the licence fee by reference to the fees charged by others for licensing comparable technologies or IP. The use of this approach is dependent on sufficiently similar comparable deals and the availability of the relevant data. Mr Yorke said that although various royalty rate databases exist, they are of limited use.
So Mr Yorke’s view was that the income approach — whereby a licence fee is based on the licensee’s expected profits from using the relevant IP — would be the most appropriate in this case, albeit that this is expressed in the licence agreement as a percentage of sales. In the absence of any long-term pre-tax operating profit projections dating back to 2001 (which he said were unlikely to be available) he proposed to use the actual long-term pre-tax operating profit of the notional licensee. The availability of this methodology was accepted by Mr Davies and Mr Black.[39]
[39]Mr Schwarz also acknowledged that notional licence could not exceed actual profits.
Mr Yorke opined that the income approach involved the use of the “25 per cent rule”, whereby the starting point is that a licensee is assumed to pay 25 per cent of its long-term operating profits (Earnings Before Interest, Tax and Depreciation/Amortisation or EBITDA) as a licence fee. He said the adoption of such a starting point reflected his own experience of royalty rates as often being in the range of one to 10 per cent of net sales and noted that the rule was used by others for guidance, including Inland Revenue.[40]
[40]The 25 per cent rule is a proxy for a five per cent royalty on sales.
The 25 per cent starting point should then be adjusted (up or down) to take into account the various relevant factors in the individual case which, Mr Yorke said, included:[41]
[41]Messrs Davies and Black agreed these were relevant factors.
(a)the type of IP being licensed;
(b)the strength of that IP;
(c)the territorial scope of that IP;
(d)the cost and effort that would be required to make a non-infringing version;
(e)the extent to which the licensed technology is ready for market or requires further development or investment;
(f)the amount of effort and expenditure the licensee would be required to invest before launch;
(g)the state of the relevant market, including the existence or otherwise of competitors, and any barriers to entry by potential competitors;
(h)the availability of alternative products;
(i)whether the licence would be exclusive;
(j)the extent to which the licensee’s profit was attributable to the IP being licensed; and
(k)any pressure to do the deal quickly, for example because one of the parties is under financial pressure or is working to a tight deadline.[42]
[42]We note that the second Copinger principle expressly excludes any impecuniosity suffered by the licensee as a relevant factor.

After taking into account the relevant factors in play in the present case, Mr Yorke concluded that, in the event that the notional licensee was Geostel, the appropriate licence fee for the cup assemblies alone would be somewhere between three and five per cent of Geostel’s projected or — if the relevant data was not available — its actual, long term pre-tax operating profits.[43] Although that could be converted into a percentage royalty based on Geostel’s net sales, that step was unnecessary as a royalty on profits could simply be used. He gave a figure of between five and 10 per cent if the licensee was found to be Napier.
[43]This estimate was based on the assumption that Geostel’s proposed sales of the cup assemblies to its overseas customers would not be infringing acts
Mr Yorke said that where the cup assemblies were supplied by Geostel together with a grader machine for one price, the simplest way to proceed would be to work out how many cup assemblies were supplied with each machine, and then calculate the licence fee for that machine sale based on the licence fee that would have been charged on the sale of the cup assemblies had those cup assemblies been sold separately. He said that given his understanding that Geostel had supplied cup assemblies as a consumable at a standard price per thousand units, it should be simple to work this out.
Mr Yorke was not challenged in cross-examination on the appropriateness of the adoption of the “income” or “available profits” method and nor was any question raised with him about his use of the 25 per cent starting point.

Mr Hussey

Mr Hussey is an accountant with specialist expertise in business valuations and the assessment of economic loss. He has presented papers to the Intellectual Property Society of Australia and New Zealand on the topic of Brand Valuation Assessment of Damages in Intellectual Property. In his evidence he assessed the sales and (anticipated or actual) profit figures available from Napier and Geostel and then applied Mr Yorke’s proposed percentages to them.
As far as Napier was concerned, his starting point (based on Napier’s anticipated and actual costings in 2001) was that the cup assemblies were manufactured at a cost of $2.19 and sold for $4.25, giving a gross margin of 48 per cent of sales. After making various further adjustments (for reasons he explained) Mr Hussey concluded that, as of mid-2001:
(a)Napier’s expected gross margin from selling the cup assemblies would be around 50 per cent of sales (which was around 19 per cent higher than its overall gross margins as reported in its formal financial statements); and
(b)its expected operating profits/EBITDA arising from the manufacture of the cup assemblies would be around 47.5 per cent of sales.[44]
[44]In other words, Mr Hussey assessed Napier’s operating costs as constituting 2.5 per cent of sales.
Mr Hussey said that applying Mr Yorke’s five to 10 per cent royalty rate to Napier’s anticipated EBITDA would result in the royalty being between 2.38 to 4.75 per cent of infringing cup assembly sales, which (according to Napier’s evidence) totalled $560,846. That yielded a royalty of between 16 and 32 cents a cup or total royalties of between (approximately) $13,000 and $27,000. Mr Hussey noted that if Mr Black’s 75 per cent figure were adopted then the royalty would be $5.06 per cup (which would have been 158 per cent of Napier’s EBITDA or 1.58 times any profit made by Napier).
As far as Geostel was concerned, Mr Hussey began by noting that:
(a)Geostel bought the cup assemblies from Napier for $6.75;
(b)Geostel’s average selling price for the cup assemblies was $23;
(c)Geostel would therefore have been anticipating a gross margin of around $16.25 (or 70 per cent) per unit;
(d)Each grading machine sold came with 1000 cup assemblies, which would be worth $23,000;
(e)Mr Yorke’s position was that:
(i)a royalty would be paid on the cup assemblies but not the machines themselves;
(ii)an appropriate royalty rate would be three to five per cent of Geostel’s long term profits if the notional licence was for manufacture only (not sale); and
(f)after allowing for overheads of 3.5 per cent the anticipated long-term profit would be 66.5 per cent of sales revenue attributable to the cup assemblies which, based on Mr Yorke’s three to five per cent figure would yield a royalty rate of between two and 3.325 per cent of sales.
Mr Hussey then noted that Geostel’s financial statements revealed that its operations had not been nearly as successful as anticipated and, in fact, over the infringement period:
(a)its gross profit (after deducting direct expenses) was $981,000, or 11.5 per cent of total sales (including graders) of $8.504 million; and
(b)after deducting overheads, its operating profit (EBITDA) was only $294,000 or 3.5 per cent of sales.
Based on Mr Black’s evidence, Mr Hussey noted that $7.384 million of total sales related to the sale of 29 grading machines so the remaining $1.12 million (Mr Hussey assumed) related to the sale of cup assemblies.
Next, Mr Hussey calculated the sales value of the cup assemblies sold together with the grading machines as $774,221.[45] Adding this to the $1.12 million of other cup assembly sales meant that the total amount of sales subject to the royalty was $1,894,000.[46]
[45]This figure suggested a sale price of $29.77 per cup which was higher than the $23 anticipated sale price he had earlier mentioned.
[46]This figure is predicated on the sale value of the snap on chassis being included in the calculations (Mr Hussey did alternative calculations based on it being excluded). No issue is taken on appeal with the Judge’s decision that the snap on chassis should be included.
And last, an application of Mr Yorke’s royalty bands (between three and five per cent of EBITDA or between two and 3.325 per cent of sales) yielded total royalties of between $37,880 and $62,976 or between 60 and 99 cents per cup.

THE HIGH COURT DECISION

In the section which follows we summarise those parts of the 2018 High Court judgment which are necessary in order to understand and determine the matters raised by this appeal.

Approach

After reviewing the relevant authorities, the Judge extrapolated the following principles governing the assessment of user principle damages:[47]
[47]2018 HC decision, above n 6, at [33].
(a)The hypothetical negotiation is a construct used to identify a reasonable price that the defendant would pay for the use of the plaintiff’s property.
(b)The task is to identify the fee that would have been agreed — not what ought to be imposed.
(c)The notional licensee should be the party who is the main beneficiary of the breach.
(d)The parties are assumed to be willing, regardless of whether they would in fact have ever agreed to grant or take a licence.
(e)The parties are assumed to have their actual characteristics; but they are assumed to make reasonable use of their respective bargaining positions.
(f)The date of assessment is the date that infringement began.
(g)The reluctance of the licensor to assist a potential new competitor is relevant.
(h)Alternatives are relevant — at least to the extent that they were known at the time and are available on the facts.
(i)Events after the date of assessment are irrelevant. But they may assist the Court in determining what the parties would have known or thought at the time.
Then, she said a three-step process would be required:[48]
(a)to ascertain the total number of infringements, which she took to be the total number of cup assemblies manufactured by Napier for Geostel;
(b)to estimate the notional licence fee to be applied to each item; and
(c)to fix loss by multiplying the number of infringing items by the notional licence fee.

Factual findings

[48]At [34].

The Judge next turned to determine the further facts she saw as relevant to fixing the notional licence fee. By way of summary she found that:[49]
[49]We examine the findings in more detail, as necessary, later.
(a)Geostel, not Napier, was the appropriate notional licensee;[50]
[50]At [55].
(b)the notional licence would have been exclusive;[51]
[51]At [64]
(c)the business of asparagus graders is, or was, a niche market;[52]
[52]At [35].
(d)“Oraka” had no viable competitor at the time of breach because other grading machines were still at the prototype or trial stage;[53] but
(e)there was nonetheless a large potential overseas market and commercial competition overseas was looming;[54] and
(f)Geostel had alternatives available — either designing around the cup or manufacturing the cup overseas (where the copyright in the designs was not recognised) but:
(i)any “design around” would have been less desirable and more financially burdensome to Geostel than obtaining a licence;[55] and
(ii)cups produced overseas could not be sent to New Zealand for attachment to the machines without breaching Technologies’ copyright;[56]
(iii)the cup assembly was a valuable — though not essential — part of the machine;[57]
(iv)a debt owed by Geostel to Kamber was not material to the negotiation of the notional licence fee;[58] and
(v)there had been a total of 85,000 infringing cup assemblies.[59]

The damages calculation

[53]At [35]. It is unclear who the Judge meant by “Oraka” here.

[54]At [37].

[55]At [83].

[56]At [85]. The cups and machines would either have to be sent to buyers separately to be assembled without testing, or the machines would have to be sent to — and fully assembled in — China before being re-exported to buyers. The Judge said both possibilities were unrealistic.

[57]At [92].

[58]At [66].

[59]At [44]. No issue is taken with this on appeal.

Early in the judgment the Judge recorded the competing contentions of the parties as to the appropriate methodology for calculating damages.[60] More particularly, she noted:
[60]At [50]–[52].
(a)Technologies contended that:
(i)where the cups were provided as part of a grader sale, the licence fee should be calculated at 50 per cent of the invoice price of the whole grader;[61]
(ii)where the cups had been sold as spare parts, the notional licence fee should be 75 per cent of the invoice price of the cup; and
(iii)by reference to calculations done on that basis by a chartered accountant (Mr Black) the notional licence fee totalled $5,009,691.[62]
(b)The defendants contended that the best means of assessing the notional licence fee was by reference to the “income approach” and the use of Mr Yorke’s “25 per cent rule” as a starting point. The Judge noted that an application of Mr Yorke’s approach in this case meant that the notional licence fee would comprise:
(i)between three and 10 per cent of Geostel’s actual long-term pre‑tax operating profit; or
(ii)between five and 10 per cent of Napier’s expected profit from manufacture and sale of the cup assembly units to Geostel.
[61]This was premised on the cup being a pivotal part of the grader, and Mr Schwarz’s evidence that the Oraka grader had a 50 per cent profit margin.
[62]The Judge noted Technologies’ submission that this figure was “consistent with the loss of $4.1 million found to have been suffered by Graders as a result of the defendants’ infringements” and that the “question of Graders’ lost profits is a valid and important consideration in the present exercise” (at [51]).
Having already concluded that Geostel, not Napier, was the notional licensee, Hinton J declined to apply either the plaintiffs’ or the defendants’ proposed methodologies. Specifically, she rejected:
(a)the plaintiffs’ submission that:
(i)Graders’ loss of profits was a useful guide for the hypothetical negotiation;
(ii)the licence fee should be assessed by relevance to the value of the grading machine, rather than the cup assemblies; and
(b)the defendants’ submissions that:
(i)Geostel’s reported actual profit was a useful guide for the hypothetical negotiation; and
(ii)Mr Yorke’s 25 per cent “rule of thumb” was of assistance.
No specific mention was made of Mr Beach’s evidence about comparable licences.
The Judge’s reasons for rejecting the 25 per cent rule were that:^[63]
All witnesses accepted it represented at best a fallback position and at worst was of little utility, and I see no need to adopt a fallback position.
[63]At [133]. The judge also noted that when the “rule” was applied to the profit anticipated from the cup assembly, it yielded a result “not so very far removed from [her] assessment”.
The reasons given for expressly rejecting Mr Yorke’s proposed approach were that:[64]
[64]At [99]–[109].
(a)what actually happened should be relevant only to the extent it confirmed the original position;
(b)the Judge in Kohler Mira Ltd v Bristan Group Ltd (a case relied on by Geostel) looked at actual profits only because there was no more compelling evidence available to determine what a notional licence fee might have been;[65]
(c)the reasoning in Ultraframe (UK) Ltd v Eurocell Building Plastics Ltd (another case relied on by Geostel) was brief and unclear;[66] and
(d)in the case before her there was “more compelling” evidence than Geostel’ s final profit figures, namely the evidence that, as at mid-2001:
(i)the second and third defendants (Messrs Daynes and Robertson and Napier) were very optimistic about the profitability of their venture;
(ii)Geostel had two orders (for graders) in place and, given its state of knowledge at the time, would have expected a high potential profit on each sale and (so) would be prepared to pay a “proportionately high, but still numerically small licence fee for the one part — the cup assembly”;[67]
(iii)Geostel knew that the cost to it per cup assembly supplied by Napier was $6.75; and
(iv)Geostel would have contemplated a profit per cup assembly of about $15.[68]
[65]Kohler Mira Ltd v Bristan Group Ltd (No 2) [2014] EWHC 1931 (IPEC), [2015] FSR 9.
[66]Ultraframe (UK) Ltd v Eurocell Building Plastics Ltd [2006] EWHC 1344 (Pat), 2006 WL 1635019.
[67]2018 HC decision, above n 6, at [105].
[68]The Judge noted that that Mr Hussey’s evidence was that Geostel would have anticipated achieving a gross margin of around $16.25 per unit.
The Judge said:
[108] I therefore do not consider the quantum of the licence fee is limited by the actual (very modest) final operating profit made by Geostel. That, like many other points, is a “benefit of hindsight” argument. As Mr Hussey said, presumably the business was a disappointment to its owners. That would appear to be borne out by the finding that Graders would have made a significant profit over that same period were it not for the defendants’ infringement. That also has to be put to one side, but if I were considering one “outcome”, I clearly would have to consider both. That would hardly assist the defendants.
[109] I proceed on the basis that Geostel would have considered, as at mid‑2001, there would be ample room for a fee on the cups that was high relative to the value of the cup, the per cup cost being a very small item relative to the value of the machine. A licence fee would have seemed relatively inconsequential as a line item.
At [120] the Judge recorded her view that the most relevant factor was the known competition from Graders for an exclusive licence. She said that at the date of breach, Graders would have been prepared to pay a “significant” licence fee to hold on to its business. Conversely, it was reasonable to assume that Geostel would have been prepared to pay a significant licence fee to secure the cup assembly and oust Graders from the market. But equally, she acknowledged that Geostel would not have agreed to a fee that might put its expected profits at risk.[69]
[69]At [121]–[123].
After then referring again to the other matters she regarded as relevant (summarised at [66] above), the Judge concluded:
[129]    I assess a reasonable licence fee to be $6.00 per cup assembly, which represents about 40 per cent of Geostel’s anticipated profit margin on the non‑machine cup assemblies.
[130]    I consider that, as at mid-2001, Geostel would have agreed to pay that amount and would not have considered that any alternatives were preferable. In terms of the two immediate machines, it would have meant an outlay of $12,000 by early 2002 at the latest (when the first graders were delivered), as compared to its next best alternative, which would have involved a payment of $225,000 within approximately the same time period.
[131]    In this notional negotiation, I have to proceed on the basis that for the same reasons, and acting reasonably, Technologies would have accepted a licence fee of $6.00 per cup assembly.
[132]    The total licence fee for the 85,000 cup assemblies supplied over the period 2001 to 2009 is therefore $510,000.

Interest and costs

In a later judgment:[70]
(a)The plaintiffs were ordered to pay Napier’s costs and disbursements totalling $72,661.99 in relation to the 2016 hearing in the High Court (the result of which was reversed in this Court on 28 November 2016).
(b)The defendants were ordered to pay costs on a 2B basis (in relation to the 2017 damages hearings) and interest on the judgment sum, running at the applicable Judicature Act 1908 rate from 1 May 2005 (the mid‑point of the breach period) to the date of judgment.

THE ISSUES ON APPEAL

[70]Oraka Technologies Ltd v Geostel Vision Ltd [2019] NZHC 1325.

Prior to the hearing before us the parties filed an agreed statement of issues raised by the present appeal. They are:
(a)What is the correct approach to determining an appeal from a judgment fixing the quantum of damages? Must the court apply the approach laid down by the Supreme Court in Austin, Nichols & Co Inc v Stichting Lodestar or the narrower approach described in Flint v Lovell?[71]
[71]Austin, Nichols & Co Inc v Stichting Lodestar [2007] NZSC 103, [2008] 2 NZLR 141; and Flint v Lovell [1935] 1 KB 354 (CA).
(b)Did the High Court err in determining that the reasonable licence fee to use Technologies’ copyright in the cup assembly was $6.00 per cup assembly, being a total of $510,000? In particular, did the High Court err in:
(i)identifying Geostel rather than Napier as the licensee for the purpose of the notional negotiation;
(ii)finding that the licence fee should be assessed as if Geostel or Napier were an exclusive licensee and that a premium should be assessed on that basis; and
(iii)the weight it gave to the relative bargaining power of the notional licensor and licensee having regard to:
·alternatives available to the parties;
·competition in the market for asparagus grading machines; and
·the debt owed to Kamber by Technologies and Technologies’ general state of insolvency at the relevant time?
(c)What is the reasonable licence fee payable in respect of each infringing use?[72]
(d)Should the High Court’s costs award in relation to the 2017 hearing be set aside and reassessed by the High Court in light of the Court of Appeal's judgment?
[72]Although not specifically referred to in the agreed statement of issues, the arguments under this heading included (a) the use of anticipated, rather than actual, profits in the relevant calculations; (b) the relevance of competition from Graders and (c) the overall importance of the cup to the grader.

We address each issue in turn.

APPROPRIATE APPROACH ON APPEAL

As noted earlier, in General Tire the House of Lords described the process of determining damages in a case such as the present as ultimately being one of “judicial estimation of the available indications”.[73] Under this traditional approach, an appellate court will afford a considerable margin of appreciation to the trial judge. That is a reflection of the orthodoxy that, because determining damages is essentially a question of fact, a trial judge’s assessment will only be disturbed on appeal if there has been some error of principle or the amount of the damages awarded was so high or so small as to make it, in the court’s judgment, an erroneous assessment.[74]
[73]General Tire, above n 24, at 214.
[74]Flint v Lovell, above n 72, at 360.
On behalf of Napier, Mr Skelton QC (supported by Mr Glover for Geostel) urged us to adopt the broader approach espoused in Austin, Nichols & Co Inc v Stichting Lodestar. He submitted that adopting the Austin Nichols approach would mean that the Court could conduct the damages assessment “afresh” and make its own award. A similar submission was recently made to this Court in Eight Mile. But, the Court said:[75]
[110] In light of the approach followed in New Zealand on appeals by way of rehearing as explained in Austin, Nichols & Co Inc v Stichting Lodestar an argument might be advanced that the traditional constraints on appeals against damages awards should no longer be maintained in New Zealand. However we did not hear full argument on that issue and it is unnecessary for us to entertain it in the circumstances of this case where we are satisfied that … the High Court’s assessment of damages involved a number of errors of law.
[75]Eight Mile, above n 27, at [110] (footnote omitted).
We begin by noting that, as high as the traditional threshold might seem, it has not prevented appellate courts from intervening in appropriate cases, including on the grounds that the first instance judge erred in his or her assessment of the evidence. By way of example:
(a)in Eight Mile this Court overturned the first instance damages award on the basis that “[b]y, in effect, giving little if any recognition to the territory factor in her analysis, we consider that the Judge did depart from the recognised process of judicial estimation of the available indications”;[76] and
(b)in General Tire, the House of Lords overturned a first instance damages award (which had been upheld in the Court of Appeal) on the basis that the trial Judge had wrongly:
(i)assessed a (lump sum) licence fee by reference to the value of the use by the notional licensee rather than what the notional licensor had lost;[77] and
(ii)ignored evidence of the fees paid under other licences entered into by the plaintiffs in relation to the use of the relevant invention.
[76]At [65].
[77]General Tire, above n 24, at 220–221.
In Eight Mile the errors found to warrant intervention included giving undue or insufficient weight to certain evidence[78] and the Judge’s preference for the evidence of one witness over another.[79]
[78]It was accepted by counsel for both the appellant and the respondents in Eight Mile that “act[ing] on a wrong principle of law” could include matters of weight (Eight Mile, above n 27, at [108]–[109]).
[79]For example at [65] the Court said (emphasis added):
Given the evidence of not only Mr Gough and Ms Hellriegel but also Mr Donlevy on the relevance of the New Zealand territory, we do not consider that it was open to the Judge to adopt the international baseline figure of Ms Zamoyska and Mr Martin.
In General Tire Lord Wilberforce described the specific evidence of the other licence fees as being “of such a different and superior order of concreteness” as to make recourse to the vaguer evidence put forward by General Tire inappropriate.[80] He said:[81]
In the end it comes to the failure of [General Tire] either to discredit or discount [Firestone’s] evidence or to bring forward satisfactory evidence of their own.
[80]General Tire, above n 24, at 223.
[81]At 223.
As it happens we have concluded that there were material errors in aspects of the High Court’s assessment of the evidence here that warrant intervention on appeal. Like the Court in Eight Mile, we have concluded that those errors have resulted in a departure from the recognised process of judicial estimation based on the available indications. We proceed on that basis; there is no need to consider the threshold issue further.

WAS THE HIGH COURT WRONG TO FIND THAT THE REASONABLE LICENCE FEE WAS $6.00 PER CUP ASSEMBLY?

Was Napier the appropriate notional licensee?

Mr Hussey’s evidence made it clear that the identity of the notional licensee makes quite a difference to the quantum of the notional licence fee and, so, to the damages award.
As noted earlier, the Judge found that the notional licensee was Geostel. She began her discussion of this topic by saying:[82]
[54] The plaintiffs and Geostel both proceed, or argue, on the basis that Geostel should be the notional licensee. Mr Skelton says the hypothetical negotiation would be between Technologies and Napier.
[82]2018 HC decision, above n 6, at [54].
It is not now disputed that this statement was incorrect. For reasons that are obvious (in light of the difference in outcome noted above) both Geostel and Napier contended then — as they contend now — that Napier was the appropriate licensee.
We do not think that this, by and of itself, could be a vitiating error. But the Judge’s belief that Geostel itself accepted that it was the appropriate “notional licensee” no doubt fortified her final view on the matter, which she expressed as follows:[83]
… Napier is in breach by manufacturing the cup assembly, but it is in effect a middle-man. It manufactured the cups for Geostel. Any licensing fee would ultimately be borne by Geostel. Mr Skelton acknowledges in his submissions that Napier would be passing on any licence fee to Geostel. In these circumstances, it seems to me that Napier’s profits, or otherwise, are not relevant. The question is, what would Geostel reasonably pay? This approach is consistent with the course adopted in Force India where the main beneficiary of the breach was treated as the notional licensee.
[83]At [55] (footnote omitted).
Substantive issue is taken with this analysis by the appellants on two related grounds. They say that it involves a misreading of the critical passage in Force India Formula One Team Ltd v 1 Malaysia Racing Team SDN BHD and also fails to reflect the practical reality of the matter, in this case.[84]
[84]Force India Formula One Team Ltd v 1 Malaysia Racing Team SDN BHD [2012] EWHC 616 (Ch), [2012] RPC 29.
Force India was a case involving both breach of confidence and copyright. The Force India Formula 1 racing team had developed a half-size model of a Formula 1 race car for testing in a specially designed wind tunnel. Force India successfully claimed that, in the course of developing a car for Force India’s Formula 1 competitor, Lotus, Aerolab wrongfully used Force India’s confidential information and downloaded computer files that contained copyright designs. As here, damages fell to be assessed on a user principle basis. And as here, the Court was therefore required to decide between two potential notional licensees — Lotus or Aerolab. The Judge in that case concluded:
(435] In my judgment the parties to the hypothetical negotiation would be Force India and Aerolab/FondTech, since it is the latter that misused the confidential information. Lotus did not misuse the confidential information. Lotus neither instigated the misuse nor was it was aware of it at the time. Furthermore, the nature of the misuse was primarily Aerolab/FondTech CAD draftsmen taking a short cut. Still further, as I have pointed out, relatively little of the confidential information found its way into parts for the actual Lotus racing car. Thus the main beneficiaries of the misuse were Aerolab and FondTech rather than Lotus. On the other hand, I consider that, even on the footing that the negotiation was between Force India and Aerolab/FondTech, the fact that Aerolab/FondTech’s client was a prospective competitor to Force India would be a relevant factor to take into account.
(Emphasis added).
Here, the appellants submitted that the Judge wrongly focused only on the second part of these dicta, namely that Aerolab was the “main beneficiary” of the breach. They say that she ignored the prior and more fundamental point, which is that it was also Aerolab — and not Lotus — who misused the confidential information in the first place. So (they say) the obvious analogy here is said to be with Napier, not Geostel.
As to practical reality, the appellants submitted that the focus must be on what activity it was that required a licence. They say that here, the infringing act was the copying inherent in the manufacturing, which was carried out by Napier and so it was Napier who required a licence.

Discussion

The difficulty with both these submissions is that they ignore the basis upon which Geostel was found jointly and severally liable for the breach. That is not a matter referred to directly in the High Court judgment and it became apparent at the hearing before us that — strangely — it is still a matter of contention. And that this was so is confirmed by the fact that Mr Yorke and Mr Hussey felt obliged to deal with the question of damages on alternative bases — more specifically that Geostel’s infringement involved the manufacture of the cups or their manufacture and sale.[85]
[85]Their evidence made it clear that liability for sales would potentially have increased the quantum of any damages award.
It seems clear from the second appeal judgment that, as at 2013, the alleged infringement by Geostel was both manufacturing and sale. At [3] the Court said:[86]
It was alleged in the High Court that, from 2001 onwards, Geostel … manufactured and sold a cup assembly … that was a substantial copy of the appellants’ copyright works. Napier Tool was alleged to have infringed the appellants’ copyright in the Schwarz cup assembly by manufacturing infringing works both for Geostel and on its own account.
[86]2013 CA decision, above n 2 (emphasis added). This is consistent with Allan J’s summary of the pleading at [18] of the 2009 HC decision, above n 11. No mention is made in either judgment of the specific pleading against Messrs Daynes and Robertson individually, although that is of no present moment.
It is worth reiterating at this point that “manufacturing” is not by and of itself an infringing act under the Copyright Act 1994. Rather, it is the copying of Technologies’ designs inherent in manufacturing the cup assemblies that is the relevant restricted act.[87]
[87]Copyright Act, ss 16(1)(a), 29 and 30.
Sale is, however, also potentially a discrete infringing act.[88] And it is not disputed that Geostel did sell the cup assemblies. But as we understand it, any contention that those sales constituted an infringement turned on whether it could be said that those sales — which were all to overseas entities — could be said to have occurred in New Zealand. That is because, as noted at [38] above, Technologies’ designs essentially have no copyright protection overseas.
[88]Sections 16(1)(b) and 36(c).
We do not need to get into the merits of the “sale” argument here other than to note it may not be straightforward. The signal point is that on our reading of the many judgments in this matter there has never been express consideration of the question, let alone a positive conclusion that the sales by Geostel constituted an infringement of the Technologies’ copyright. It was never considered by the High Court in 2009 and 2011 because that Court was of the view that there had been no infringement by copying — so the possibility of the sale of infringing copies did not exist.[89] And on our reading of this Court’s 2013 judgment, the finding that there had been an infringement was based on the identification of the relevant infringement as copying.[90] Although not made explicit there, we think a fair reading of the judgment suggests the conclusion that Geostel, Napier and Messrs Daynes and Robertson were jointly and severally liable for copying the designs was based on the Court’s view either that Napier was Geostel’s agent or, perhaps, that all parties had acted in concert in the infringement, pursuant to a common purpose.[91]
[89]2009 HC decision, above n 11; and 2011 HC decision, above n 10.
[90]2013 CA decision, above n 2. See in particular the discussion of the law at [83]–[92].
[91]As discussed in CBS Songs Ltd v Amstrad Consumer Electronics Plc [1988] AC 1013 at 1057; Universal Music Australia Pty Ltd v Cooper [2005] FCA 972, (2005) 150 FCR 1 at [135]; and National Rugby League Investments Pty Ltd v Singtel Optus Pty Ltd [2012] FCAFC 59, (2012) 201 FCR 147 at [27]–[30].
We also note that Mr Henry’s continued suggestion to us that Geostel’s infringement included sales is contrary to the plaintiffs’ latest pleading (filed in April 2017, after the remittal back) where the allegation simply was:
The defendants infringed the first plaintiff’s copyright
6. The manufacture of each infringing cup is a breach of the first plaintiff’s copyright.
We therefore proceed to consider the appellants’ two criticisms of the High Court’s conclusion that Geostel was the appropriate notional licensee on the basis that Geostel’s liability — like Napier’s — is founded on its involvement in copying Technologies’ designs through the manufacture of the cup assemblies.
As to the first criticism, we think any analogy with Force India is imperfect. In that case, Lotus (unlike Geostel) had been found not to be liable at all — either for breach of confidence or for copyright infringement. By and of itself that suggested that the notional licensee would be Aerolab, who had been found liable (albeit in a limited way) for breach of confidence.[92] Moreover, in light of our conclusions above, there is no meaningful or logical basis to say that it was Napier, rather than Geostel, who “misused” the designs. And if that is so, we think the Judge’s conclusion that Geostel was the main beneficiary of the misuse cannot be impeached.
[92]Another party (1 Malaysia UK) was found liable for breach of copyright.
For essentially the same reason, the second criticism must also fall away. Geostel’s liability is founded on the same infringing act as Napier’s — manufacturing (or the copying that is inherent in the manufacturing process). And if that is so, then the distinction sought to be drawn by the appellants cannot be sustained.
For completeness, we also note there was a sound evidential foundation for the Judge’s preference. While Mr Yorke expressed the view that Napier was the appropriate licensee, he accepted that it could have been Geostel. And while Mr Davies also accepted that the notional licensee could be either party, he expressed a preference for Geostel. After noting the fact that — as at the time of the notional negotiation (mid-2001) — there was already an informal or implied licence between Technologies and Graders (rather than Technologies and Napier), he said:
19. If Oraka Technologies were to appoint any party to be a licensee, it is likely to be a party that similarly benefits from the end sale to the consumer. This is the party receiving the greatest benefit from the sale and hence has the greatest margin that may be applied.
Accordingly we consider it was entirely open to the Judge to hold that Geostel was the appropriate notional licensee here. This ground of appeal fails.

Exclusivity

Whether or not Geostel’s notional licence would have been an exclusive one is plainly relevant to setting the notional fee. That is clear from the fourth of the Copinger principles set out at [22] above. Included in that principle is the statement that “[w]here the infringer enjoyed exclusivity in practice, the notional licence should accord with the reality”.[93]
[93]Davies, Caddick and Harbottle, above n 27.
Here, the Judge rejected the appellants’ argument that any license granted would not have been exclusive. She said:[94]
[62]     The first plaintiff submits that Geostel’s business plan was to destroy Graders, including by denigrating the Oraka grader and business. While there may be something in that, all that is relevant for present purposes is that Geostel was clearly in the market to compete head-on with Graders and indeed, as at mid-2001 had already obtained at least fairly firm orders from the Greek market where one of the second defendants had been or still was an agent for Graders. There was no indication Geostel intended to limit its sales area and every indication it expected to supersede Graders.
[63]     As Newey J observed in 32Red, the hypothetical licence should be taken to have permitted the defendant to use the terms and conditions it in fact used.
[64]     It would follow that the notional negotiation would be on the basis of an exclusive licence.
[94]2018 HC decision, above n 6, (footnote omitted).
We note that the observation from 32Red Plc v WHG (International) Limited reflects of the fourth Copinger principle.[95]

Discussion

[95]32Red Plc v WHG (International) Ltd [2013] EWHC 815 (Ch) at [58].

In our view the Judge was right to reject the possibility of a geographically split licence. But we are unable to agree with her reasons or her ultimate conclusion, that the licence would have been exclusive.
The reason why the licence would not have been divided territorially is because the concept of exclusivity in this case did not logically relate to geography. That is because the subject of the notional licence (manufacturing the cups) occurred in New Zealand and, as already noted, sale overseas was not the relevant infringing act here. Moreover — and as we have also mentioned earlier — Technologies’ designs were only the subject of domestic copyright; they had no protection in the countries to which the cups were exported. So (as Mr Yorke made clear during his oral evidence) Technologies had no IP rights to grant or licence overseas. Territorial scope was a factor that should have favoured Geostel in any notional negotiation and we think that was wrongly ignored by the High Court.
In our view, the proper focus of the exclusivity question should have been the possible existence of another, competing, licensee.[96] And such a licensee did, in reality, exist here — in the form of Graders. Moreover, Graders did, in fact, compete with Geostel throughout the notional licence period. If (as the fourth Copinger factor indicates) the notional licence should accord with reality, then the fact that the infringer (Geostel) did not enjoy exclusivity in practice also required a downward adjustment to any notional fee. In our view the High Court’s failure to recognise this also constitutes a material error.

Weight given to the relative bargaining power of the notional licensor and licensee

[96]Indeed, this was the focus of the expert evidence of (for example) Mr Yorke on this issue.

Geostel attacked a number of the Judge’s findings relating to the factors relevant to a notional licence negotiation. More particularly, Mr Glover was critical of the High Court’s assessment of:
(a)the alternatives available to the parties;
(b)the competition in the market for asparagus grader machines;
(c)Technologies’ financial position in 2001; and
(d)the level of originality in the cup assembly design and its overall importance to the grading machine.

Alternatives available to the parties

In the judgment under appeal the Judge noted that that in the Force India case the English and Welsh Court of Appeal had said:[97]
In any negotiation the parties to the negotiation will be considering what their alternatives are to doing the deal. There is no reason why a hypothetical negotiation should be any different in that respect. It is, of course, different from a real negotiation in one respect because in the hypothetical negotiation not doing the deal at all is not an alternative.
[97]2018 HC decision, above n 6, at [28], citing Force India Formula One Team Limited v Aerolab SRL [2013] EWCA Civ 780.
That relevance of hypothetical (but realistic) alternatives is also confirmed in the seventh Copinger principle, which we repeat for convenience here:[98]
It is relevant under this head that the defendant could have arrived at the same result by lawful means if the parties can be expected to have taken that fact into account in their hypothetical negotiation. That alternative need not have had all the advantages or other attributes of the infringement.
[98]Davies, Caddick and Harbottle, above n 27 (footnotes omitted).
Here, the Judge addressed the question of alternatives not just from the perspective of the defendant (Geostel) but also from the perspectives of Technologies.[99]
[99]She explained this approach in the 2018 HC decision, above n 6, at [74], where she said:
Although the discussion in the cases has been around alternatives for the licensees, this must equally apply to alternatives available to the licensor. The existence of a real alternative for the licensor is of particular significance in this case. It has not arisen on the facts of the other cases, although the language used, for example by the Court of Appeal in Force India, clearly extends to alternatives available to both parties.
As to Technologies, she began by recording the plaintiffs’ principal submission that, because the granting of a licence would effectively transfer the whole of the plaintiffs’ asparagus grading business to Geostel, the licence fee should be, at a minimum, in the amount of Graders’ lost profit (namely $4.1 million).[100]
[100]At [70].
The Judge quite rightly rejected this contention as being a barely disguised rerun of the plaintiffs’ failed loss of profits claim.[101] But then she said:
[73] In this connection however, it is of particular relevance that, as matters stood at the time of the notional negotiation, Graders had an (informal) exclusive licence from Technologies. Although Graders had not been paying for the licence, it would clearly be prepared to do so, and the licence fee would be materially affected by which party would pay more, albeit that Graders has no chance of winning this particular negotiation.
…
[76] … Graders would obviously have been prepared to pay a material licence fee to hold onto the exclusive licence for the cups.
[101]At [72].
For reasons related to the point made at [108] above we find this analysis problematic. We consider that the possibility of competition for the (notional) licence from Graders should not have been regarded as a relevant factor here. Graders were, in fact, already in possession of an implied licence, for which they were paying Technologies nothing. In our view, any notional negotiation with Geostel should logically have taken this status quo as its stepping-off point. It is counter-intuitive (given the non-arm’s-length relationship between Graders and Technologies) to hypothesise (as the High Court analysis does) that Technologies would have revoked Graders’ existing licence in order that it could engage in a bidding war with Geostel. Rather, the more apt and obvious premise is that Graders’ existing licence would continue, as before. This would logically diminish the price Geostel would be willing to pay for a notional licence.
In terms of alternatives from Geostel’s perspective, the Judge referred to Geostel’s submission that it had two principal options open to it: designing around the cup or (given that copyright only protected the designs in New Zealand) manufacturing the cup very cheaply overseas.[102]
[102]At [78].
The “design-around” possibility relied on the evidence of Mr Beach, an acknowledged expert in the design of produce grading machines, although he had no specific experience of asparagus graders. Mr Beach’s evidence was that a cup could have been developed which fitted the Geostel grader, but did not infringe Technologies’ copyright, within a period of about seven months and at an estimated cost of about $225,000.
But the Judge was sceptical, saying:[103]
… Napier, who were accepted to have both expertise and direct experience with the cup, seem to have taken from about July 2001 until after April 2002, under some pressure from Geostel, trying to design or redesign a cup for the Geostel grader to Geostel’s satisfaction. In the end, what Napier came up with infringed Technologies’ copyright.
[103]At [81].
Rather, she said:
[82]     I agree with Mr Davies, who is the head of intellectual property at EverEdge Global (NZ) Ltd and gave expert evidence for the first plaintiff, that there could not be any guarantee of success at the end of Mr Beach’s seven‑month design period, and I consider in any event that as at mid-2001 Geostel would not have wanted to wait that out and accept that uncertainty, or would have happily paid a licence fee to avoid doing so. That is illustrated by its exasperation with Napier. It would have been prepared to pay for what it would have considered at the time to be a proven product that would enable it to secure the immediate order stream it had already accessed. …
[83]      It is also particularly relevant that the design-around alternative involved a large capital outlay in a short timeframe, capital which Geostel clearly did not have, whereas a licence fee involves a small outlay over a long period.
[84]      I accept, however, that there would be a limit to the quantum of any fee, bearing in mind the alternative design route.
As to the possibility of non-infringing manufacture overseas, the Judge said:
[85] I do not consider the argument that copyright could have been circumvented by manufacturing overseas, is viable. There was some vagueness as to which country was contemplated, with focus mainly on China. The part could not be manufactured in China and transported to New Zealand to complete the assembly of the machine, as the importation of it would breach the copyright. Geostel’s business plan (as indeed Oraka’s) was based on manufacture, pre-assembly and testing in one place before dismantling and export. It was therefore not realistic for a New Zealand-made machine and Chinese-made cups to be separately transported to the buyer and assembled there without prior testing, or for the New Zealand machine to be transported to China for assembly with the cups and then re-export[ed].
[86] There was insufficient evidence for me to conclude that as at mid‑2001, there was any realistic option other than for the cups to be manufactured in New Zealand. I note that was the manufacturing process followed throughout by the defendants. It did not, for example, seek to reduce its costs by manufacturing part of, or the whole machine elsewhere, although on the surface there would be significant cost-savings by doing so. In my view, Geostel did not do that because, like Graders, its assessment was that it would not work for a machine such as this.
The Judge concluded that she would proceed on the basis that:[104]
… the defendants were neither in a high-stress position, nor had strongly viable alternatives as at mid-2001, although the ability to design an alternative cup placed some curb on the licence fee that would be agreed by Geostel.
[104]At [88].
On this point we are unable to agree with the appellants that these conclusions were not open to the Judge. They were based on her assessment of evidence which she was entitled to accept. Her reasoning was firmly founded in the realities with which Geostel was faced at the time of the notional negotiation.
Our conclusions under this heading are, therefore, that:
(a)the High Court was wrong to accept the hypothesis that Technologies had an alternative (exclusive) licensee to Geostel in the form of Graders and that competition from Graders for the licence would have pushed up the price of the notional licence; but
(b)there was no error in the assessment of the alternatives open to Geostel.

Competition in the market for asparagus grading machines

This point appears to be related to the last. As we understood the argument, Geostel says there was a third “alternative” that would have been open to it. This was the possibility that Geostel might have been able to obtain cups that would have worked in its grader from some other, existing, asparagus grader manufacturer.
This issue was not squarely addressed in the course of the High Court judgment. Rather, it was implicitly rejected as a consequence of the Judge’s earlier factual findings, where she said:
[35] The market for asparagus graders is, or was, a niche market. I accept the evidence of Mr Schwarz and Mr Caraccioli (a Californian asparagus grower who was not called for cross-examination) that as at mid-2001, Oraka had no active commercial competitor. Mr Daynes said he had seen another machine, but had not seen it working. Mr Beach, a software engineer who gave expert evidence for the defendants, said he had seen an alternative “GP Graders” machine, but it was before his time and so he did not know much about it. He had never been involved in the asparagus grading industry, which he described as a boutique market. It seems any other machine in operation in mid-2001 was a prototype or trial machine. According to Mr Schwarz, these trial machines were failing.
The short and inferential point is that if, as the Judge found, there was no commercial competition in the asparagus grader market in 2001, there could be no possibility of sourcing cups from another grader manufacturer.
On appeal, Geostel submitted that this conclusion was at odds with the relevant evidence, including:
(a)the evidence of Mr Daynes that there were at least three other asparagus-grading machines which had been sold worldwide prior to July 2001, manufactured by GP Graders, Sammo and Strauss;[105]
(b)the evidence of Mr Beach who (as the Judge noted):
(i)expressly disagreed with Mr Schwarz’s evidence that the Oraka cup was unique in the world as at 2001;
(ii)knew of the GP Graders cup which Compac had worked on in 1994;[106]
(c)Mr Daynes’ evidence that the cup assembly on the Sammo machine had been modified from a design originally used for kiwifruit and apples, and that it was common practice to use a standard cup assembly and simply to change the top part of the carrier (depending on the type of produce);
(d)the evidence that off-the-shelf cup designs were available from KVP Falcon and Habasit;
(e)Mr Schwarz’s acceptance under cross-examination that it was possible to operate a grader without using Oraka’s cups; and
(f)the evidence that the Sammo and GP Graders cups were suitable for transporting asparagus spears through the grading machines in a way which did not damage the spears, and that Mr Daynes had adapted a GP Graders machine to work with Geostel’s software.
[105]Mr Daynes had sent Mr Schwarz a report on the Sammo machine in April 2001.
[106]Compac designed the software and electronics for the asparagus grading machine and GP Graders designed and manufactured the cup used on it.
We are inclined to agree with Mr Glover that the Court’s preference for Mr Caraccioli’s evidence over that of Mr Beach and Mr Daynes, is difficult to justify. Mr Caraccioli’s brief of evidence was prepared for the purposes of the first damages hearing (it is dated 18 January 2016). Most of his brief was concerned with supporting Mr Schwarz (from whom he had purchased five Oraka Graders in 2003) and opining on the damage caused to Mr Schwarz’ business by Mr Robertson/Geostel. That was, at least, arguably relevant to the position Technologies was trying to advance in 2016. All he appears to have said about competition in the market at the relevant time was:
In about 2003 I heard about the Oraka grader. From recollection another grower in the Stockton area was putting in an Oraka machine and I went over and had a look at it. At that time the only other asparagus grading machinery that I had knew about was the Geostel grader which was being put in for neighbours of mine, the Violinis.
Ultimately, Mr Caraccioli was not required to attend the first hearing in person and his brief was taken as read without the need for cross-examination. That is, of course consistent with defendants’ wider position that the lost profits claim was misconceived as a matter of law. On that analysis (which later proved to be correct) his evidence was simply irrelevant.
Less than a week before the second damages hearing, however, the plaintiffs advised they intended to rely upon Mr Caraccioli’s brief for the purposes of that hearing. Mr Skelton advised that it was, by then, too late to require him to attend the hearing or to facilitate his cross-examination.[107] Instead, in closing submissions, Geostel made the obvious points that he was not called as an expert, his evidence was vague in nature and limited to saying what he could recall from his time as an asparagus grower in California in the early 2000s. Geostel submitted the Court should prefer the clearer and better evidence of Messrs Beach and Daynes that alternative cup assemblies were, indeed, available at that time.
[107]Although the respondents say this could and would have been facilitated had an objection been raised at the time, the necessary arrangements would — as a matter of the court’s processes — almost certainly have required an adjournment of the scheduled four (and ultimately five) day hearing.
In light of these matters we have reservations about the weight apparently given to the evidence of Mr Caraccioli. His evidence was not the best evidence on the issue. The Judge’s comment about the absence of cross-examination and the weight she afforded his evidence took no account of that. While it is not ultimately a point that weighs heavily with us in the overall context of the appeal, it fortifies us in the conclusion we reach.

The summary judgment against Technologies and its state of insolvency

In the High Court Geostel had submitted it was relevant that, at the time of the notional negotiation, Technologies owed a debt of approximately $107,000 to Kamber (another company owned by Mr Daynes). Geostel referred to a summary judgment obtained by Kamber against Technologies for that amount and said that it was relevant that, in the hypothetical negotiation, Kamber could have assigned the debt to Geostel who could have then used it to liquidate Technologies. This would have placed Geostel in a strong bargaining position.
In rejecting this submission, the Judge said:[108]
[66] In that hypothetical scenario, Technologies would no doubt have disputed the debt, which Mr Schwarz said it had grounds to dispute, but chose not to. Further, Mr Glover's argument ignores the fact that Technologies had a significant asset by way of the copyright which is the subject of this litigation. Even if Technologies were to be notionally liquidated, it would not follow that the copyright would fall into the hands of Geostel. Clearly Graders would have competed for the purchase of the copyright, or more likely been called on to pay a licence fee to enable the debt to be paid.
[108]2018 HC decision, above n 6.
There is now no argument that the Judge’s recorded understanding of the position in relation to the summary judgment proceeding was mistaken. We were taken to the relevant court documents which reveal that the summary judgment was entered after hearing from both parties and despite opposition from Technologies. So the Judge’s reliance on Mr Schwarz’ evidence here was misplaced; the debt was undoubtedly owed and it was, we think, undoubtedly a bargaining chip which could have been utilised by Mr Daynes and Geostel in any negotiation. Indeed, Mr Schwarz accepted in cross-examination that this would have created substantial leverage for the licensee.[109]
[109]Geostel also referred us to Mr Schwarz’s acceptance that Technologies was not in a good financial position at the time of hypothetical negotiation.
We agree with the appellants that the High Court was in error on this point.

The level of originality in the cup assembly design and its overall importance to the grading machine

We have noted Mr Beach’s evidence on this topic earlier. Geostel submitted that it should have been adopted.
Notwithstanding his undoubted expertise, however, we consider the Judge was entitled to take a somewhat (although not wholly) different view. Her conclusion that the cup assembly was a valuable part of the machine and her preference for Mr Davies’ evidence on this point was open to her.

What is the reasonable licence fee payable in respect of each infringing use?

As well as attacking on the Judge’s assessment of the factors just discussed, the appellants also separately attack the methodology adopted in the High Court for quantifying the licence fee.
We have set out the principles relevant to assessing damages in a case like this earlier in this judgment. What is most directly engaged by this ground of appeal is the eighth Copinger principle which, to reiterate, is that:[110]
In the absence of comparable licences or other compelling evidence the royalty may be based on the “available profits” method: the defendant’s actual profits are calculated; it is assumed that the parties would have accurately predicted these profits when negotiating; the profits are then (in effect) divided between the claimant and the defendant.
[110]Davies, Caddick and Harbottle, above n 27.
More specifically, the issue is whether it can be said that, in light of that principle, the Judge erred in:
(a)disregarding Mr Beach’s evidence about comparable licence fees in the produce grading industry; and/or
(b)preferring to base her calculations on Geostel’s anticipated, rather than actual, profits; and/or
(c)rejecting Mr Yorke’s recommended use of the 25 per cent rule.

Preliminary observation

Before turning to consider those questions we think it is important to reiterate that, as is made quite clear in General Tire and all subsequent cases, the burden was on the plaintiffs to prove their loss and (so) to adduce evidence which will guide the court in its assessment of a notional licence fee.
In our view the plaintiffs failed to do that here. Instead, they reran (in thinly disguised form) Graders’ lost profits claim, despite the earlier finding by this Court that those profits could not be the proper foundation for a damages claim. That is made quite clear from:
(a)Mr Black’s proposition that “the plaintiffs needed to be effectively compensated for the entire loss of the business” and his effective treatment of the Oraka companies as a single entity (directly contrary to the findings of this Court in 2016); and
(b)the statement by Mr Davies that:
The only notional licence fee likely to make the licensor “willing” is one that compensates for the entire loss of business by the licensor with the licensee being directly competitive.
We consider the High Court Judge was right to put this evidence to one side. We agree with her that:[111]
… the “negotiation” is between Technologies and Geostel, not Graders and Geostel. Technologies was not trading. I agree with the defendants that the focus has to be on what is a fair licence fee that the defendants would pay, not to be confused with a re-run of a loss of profit claim by the plaintiffs.
[111]2018 HC decision, above n 6, at [72]. It seems plain from the omission of any substantive reference to Mr Black in the course of the judgment that his evidence did not assist the Court in any material respect.

The plaintiffs proffered no alternative methodology. As noted earlier, Mr Davies simply said that that it was not a matter on which expert licensing evidence could assist. In light of that we think a conclusion that the plaintiffs failed to prove their loss might well have been available, particularly given the history of this matter. But the appellants did not ask us to proceed on that basis and so we do not do so. Instead, we turn now to consider the three issues noted at [140] above.

Disregarding evidence about comparable licence fees

As noted earlier, the House of Lords in General Tire overturned the decisions below on the grounds that the lower courts had wrongly ignored the evidence about the actual fees charged for licences entered into by the plaintiff for the same product.
The only witness who spoke about comparable licence fees was Mr Beach. His evidence was that based on his experience, the likely royalty charged for permission to make a direct copy of component such as the cup assemblies would be around three to four per cent of gross sales (but only in the event of worldwide IP protection). In the event that the protection was not worldwide, he said the fee would have been nominal. No contradictory evidence was given for the plaintiffs and there was no cross-examination of Mr Beach on the point.[112]
[112]Although the plaintiffs’ evidence about the appropriate royalty fee was couched in percentage terms, that percentage was not based on industry practice. Rather (and as noted above) it was predicated on the unsupportable assumption that the percentage should result in a fee that bore some relationship to Graders’ lost profits.
Once account is taken of the fact that the notional licence could not have conferred worldwide protection, Mr Beach’s evidence did, to some extent, tie in with and support the evidence of Mr Yorke and Mr Hussey.[113] That said, however, his evidence was not of the kind that the House of Lords found had been wrongly ignored in General Tire. It was not evidence about the fees charged by Technologies for other cup assembly licences — it was general evidence about componentry licences based on his experience in the produce grading industry. We do not think it could be said that the licences referred to by Mr Beach were sufficiently specific or sufficiently similar to the notional licence at issue here that the High Court was required to take his evidence into account.

Calculations based on Geostel’s anticipated — rather than actual — profits

[113]Namely that that an appropriate licence fee would be in the vicinity of two to 3.325 per cent of sales.

The Judge used Geostel’s anticipated, rather than actual, profits because she considered that to do otherwise would be to assess the notional licence fee with the benefit of hindsight.[114] There was inferential support for that approach in the evidence of Mr Yorke, who explained the use of actual profits is simply a proxy for anticipated profits which may be difficult to ascertain some time after the event. And here, the Judge did have the evidence of Mr Hussey as to the price Geostel paid Napier for the cup assemblies and the price for which Geostel on-sold them.
[114]2018 HC decision, above n 6, at [108].
That said, however, the “available profits” method articulated in the eighth Copinger principle (and adopted in the Kohler-Mira and Ultraframe decisions) uses actual profits. As Copinger makes clear, that is because it is assumed that the parties would have accurately predicted these profits when negotiating.[115] And here, the evidence was that Geostel’s overall actual operating profits (i.e. for the sales of the grading machines and the cup assemblies over the entire infringement period) was only $294,000 — considerably less than the $510,000 damages sum awarded. So while the Judge’s use of anticipated profits did not, by and of itself, involve an error of principle we think it was wrong to ignore Geostel’s actual profits in this case.

Rejection of the 25 per cent rule

[115]See also Kohler-Mira, above n 65, at [55].

As well as being based on anticipated rather than actual profits, the Judge arrived at the $510,000 licence fee figure by using a notional royalty rate of 40 per cent.[116] Self-evidently, 40 per cent is two thirds higher than the 25 per cent starting point proposed by Mr Yorke, and it was eight times more than his most generous (five per cent) end point.
[116]2018 HC decision, above n 6, at [129].
Although it is clear from the judgment under appeal that the Judge intended to (and did) reject Mr Yorke’s 25 per cent rule, her reasons for then adopting the 40 per cent figure are less obvious. We have noted her rejection of the plaintiffs’ evidence on the question of how the royalty should be set (earlier our agreement with it) so that cannot be the basis for the 40 per cent. And the only other evidence on the topic was Mr Beach’s evidence about broadly comparable licence fees, which we have also said the Judge was entitled to disregard.[117]
[117]Although Mr Yorke also mentioned the cost approach, that was discounted by him. We would think that its potential application would require considerable further evidence which was not before the Court.
It would, in theory, have been open to the Judge to regard the 25 per cent rule as a starting point rather than an upper limit and to move upwards from there based on her assessment of the relevant factors.[118] We note in passing that this is how the rule was applied in the more recent High Court decision of Dodson Motorsport Ltd v Logiical Performance Ltd.[119] But that is not a point we can take much further. Because the Judge expressly rejected the use of the 25 per cent rule, the choice of 40 per cent appears to have been impressionistic; we are unable to discern an evidentiary basis for it. Moreover, and in light of our different conclusions on a number of the relevant factors (summarised below) we do not think even a generous application of the 25 per cent rule could result in a licence fee that was 40 per cent of either anticipated or actual sales.

Conclusions

[118]It was not said to be an upper limit by Mr Yorke.

[119]Dodson Motorsport Ltd v Logiical Performance Ltd [2019] NZHC 918. In that case, the expert witness Mr Beylefeld (who was called by the plaintiffs in the present case at the 2016 damages hearing) proposed using the 25 per cent rule, which was described by the Judge (at [257]) as “a presumption that the parties would agree that 25 per cent of [the notional licensee’s] profits would be paid to [the] licensor” but which could then be either increased or decreased depending on relevant factors.

We have concluded earlier that there were material errors in and omissions in the High Court’s assessment which warrant a considerable reduction in damages here. To reiterate, we consider that:
(a)the notional licence would not have been exclusive;
(b)Graders would not have competed for the licence but, rather, have continued to operate (as it in fact did) under its implied licence at no cost;
(c)the fact that the IP which was the subject of the notional licence (Technologies’ copyright in the cup assembly design) was not protected overseas was a relevant factor which should have been taken into account;
(d)the debt owed by Technologies to Kamber was relevant and placed Technologies in a weak notional bargaining position;
(e)the licence fee should not have exceeded Geostel’s actual operating profits; and.
(f)there was no obvious basis for the use of the 40 per cent royalty figure and, in the absence of any other evidence, the 25 per cent rule should have been used as a starting point.
These are important matters. In particular, the Judge expressly noted that the notional competition for the licence from Graders was (in her view) the “most relevant factor” in the notional negotiation.[120] We also regard the territorial limits of the right being licensed was a particularly significant factor in terms of setting the appropriate fee. Viewed collectively, we consider that these matters have resulted in an award of damages which is so high as to make it an erroneous assessment.
[120]2018 HC decision, above n 6, at [120].
Once that point is reached, and bearing in mind our earlier agreement with the High Court that the plaintiffs failed to discharge the evidentiary burden on them, the only evidence upon which a notional fee could be based is the evidence given by Mr Yorke and Mr Hussey, namely that the 25 per cent starting point should be adjusted down to between three to five per cent of Geostel’s projected profits or between two and 3.325 per cent of sales. This yields a sum of:
(a)between $38,250 and $63,750, if the Judge’s method of assessing gross profits is used;[121] or
(b)between $37,880 and $62,975 if sales are used.[122]
[121]The Judge’s method of calculating anticipated profits yields gross profits of $1,275,000 (85,000 cup assemblies x $15 per cup assembly).
[122]Using Mr Hussey’s $1,894,000 figure.
Given the very limited territorial protection that the notional licence could afford, and our conclusion that the licence would not have been exclusive, we are unable to place the licence fee above the middle of these ranges. We consider a reduction in the damages award from $510,000 to $47,000 — or approximately 55 cents per cup — constitutes an appropriate application of the 25 per cent rule in this case.

Result

The appeal is allowed. The High Court’s damages award of $510,000 is quashed and we substitute damages in the sum of $47,000.
The costs award in Technologies’ favour in relation to the 2017 High Court hearing is set aside, with the issue of costs for that hearing remitted back to the High Court to be reassessed in light of this Court’s judgment.
The appellants are entitled to costs in this Court for a standard appeal on a band A basis plus any disbursements.

Solicitors:
Malloy Goodwin Harford, Auckland for Appellant in CA254/2018
Hudson Gavin Martin, Auckland for Appellant in CA261/2018
Shanahans Law Ltd, Auckland for Respondents

[2020] NZCA 256

25 June 2020