LION - BEER, SPIRITS & WINES (NZ) LTD V PERNOD RICARD NEW ZEALAND LTD CA771/2012 [2013] NZCA 625

LION - BEER, SPIRITS & WINES (NZ) LTD V PERNOD RICARD NEW ZEALAND LTD CA771/2012 [2013] NZCA 625 [6 December 2013]ORDER LIMITING DISTRIBUTION OF THIS UNREDACTED VERSIONOF THE JUDGMENT TO THE PARTIES AND THEIR LEGALREPRESENTATIVESTHIS REDACTED VERSION IS FOR PUBLIC RELEASEIN THE COURT OF APPEAL OF NEW ZEALANDCA771/2012[2013] NZCA 625BETWEEN LION - BEER, SPIRITS & WINES (NZ)LTDAppellantAND PERNOD RICARD NEW ZEALANDLTDRespondentHearing: 17, 18 and 19 September 2013Court: O'Regan P, Stevens and White JJCounsel: J A Farmer QC, M R Crotty and S P Pope for AppellantA R Galbraith QC, D J Cooper and S V A East for RespondentJudgment: 6 December 2013 at 3 pmJUDGMENT OF THE COURTREASONS OF THE COURT(Given by O'Regan P)Table of contentsPara NoIntroduction [1]Issues [2]Appeal standard [8]Sale and purchase agreement [9]Retailer margins [16]Guaranteed Margin Agreement [18]Alleged breach of warranties [25]Contractual context [35]Materiality [39]Warranty claim based on written answer [42]What was disclosed (and warranted) in the TTF? [45]What Lion did with the disclosed information [54]Margins on the Lindauer brands [64]Why were lower margins assumed for non-Lindauer brands? [65]Why were retail prices raised in the integration model? [66]The effect of the GMA on volume mix [67]Other information available to Lion [78]Breach of warranty 3? [86]Breach of warranty 4? [105]GMA applied for limited periods of time [111]The GMA was not binding on Lion [112]The GMA applied across Pernod's range [114]Lion had other information available to it [115]Pernod was entitled to expect Lion to consider other information [116]Conclusion: warranty 4 [118]Causation [120]Loss [124]Result [130]Costs [131]Introduction[1] The appellant (Lion) and the respondent (Pernod) entered into a sale and purchase agreement under which Pernod sold a number of wine brands to Lion.1Various disputes arose and High Court litigation followed. In the High Court, Pernod claimed against Lion for reimbursement of excise duty on stocks supplied by Pernod to Lion. Lion counterclaimed for breach of warranties and for recovery of amounts paid by Lion to retailers in respect of discounts on sales stock supplied byPernod to the retailers. The High Court hearing took 28 sitting days. Pernod's claim for excise duty succeeded and Lion's counterclaims for breach of warranty and forrecovery of discounts failed.2 Lion now appeals against the decision of Allan J to dismiss its breach of warranty claim. It does not appeal against the excise duty or discount payments aspects of the High Court decision.1 The other party to the agreement, Indevin Group Ltd, is not involved in the present litigation.2 Pernod Ricard New Zealand Ltd v Lion - Beer, Spirits & Wine (NZ) Ltd [2012] NZHC 2801["High Court judgment"]. Because of commercial confidentiality in relation to some of theevidence before the High Court, the published version of the judgment has a number of details redacted from it.Issues[2] Under the warranty claims Lion contended that Pernod had breached two of the warranties set out in schedule 1 of the sale and purchase agreement. The first of these was in paragraph 3 of the schedule, and was a warranty to the effect that the information disclosed during the sale process was materially accurate and not materially misleading in its context. We will call this warranty 3. The High Court Judge found that Pernod had breached this warranty. Pernod filed a notice supporting the High Court judgment on another ground, seeking the reversal of that finding. The first issue before us is, therefore, whether there was a breach of warranty 3.[3] The second warranty said to have been breached was a warranty contained in paragraph 4.1 of schedule 1 of the sale and purchase agreement. We will call this warranty 4. This contained a warranty that Pernod was not aware of any circumstances not disclosed to Lion which, if disclosed, might reasonably be expected to lead a purchaser to reduce materially its assessment of the value of the assets subject to the sale and purchase agreement. The High Court Judge found that there was no breach of this warranty. Lion appeals against that finding. The second issue is, therefore, whether warranty 4 was breached.[4] The Judge found that, although warranty 3 had been breached, the breach had not caused any loss. The third issue is, therefore, if either warranty 3 or warranty 4 is breached, did the breach cause any loss to Lion?[5] If we find a breach of warranty and causation, the next issue is whether Lion suffered loss and, if so, what the quantum of loss was. Because the High Court Judge found there was no causation, he did not make findings on either of theseissues. Lion's claim is now limited to what it says is the impact of its future (lowerthan anticipated) return from sales to Progressive Enterprises Ltd, the owner of Countdown supermarkets, on the value of the brands. It originally claimed on a wider basis, but now accepts that non-disclosure of a guaranteed margin agreement (GMA) with Progressive cannot be said to affect its assessment of future margins on sales to other retailers.[6] Lion claims damages of $8,962,000 based on data provided by Progressive on the margins during the relevant period, or, alternatively, $6,249,000 based on data provided by Pernod about those margins.[7] Before addressing these issues, we will set out the background to the transaction evidenced by the sale and purchase agreement.Appeal standard[8] Counsel for Pernod, Mr Galbraith QC, argued that this Court should be cautious in re-evaluating factual findings made by the Judge at first instance. He relied for that submission on a recent decision of the United Kingdom Supreme Court, McGraddie v McGraddie.3 We do not consider that the approach of the United Kingdom Supreme Court is applicable in New Zealand. Rather, we apply the standard adopted by the Supreme Court of New Zealand in Austin, Nichols & Co Inc v Stichting Lodestar that "those exercising general rights of appeal are entitled tojudgment in accordance with the opinion of the appellate court, even where thatopinion is an assessment of fact and degree and entails a value judgment".4Sale and purchase agreement[9] Pernod is the largest wholesale wine distributor in New Zealand. Lion is the leading New Zealand beer distributor. Under a sale and purchase agreement dated 18 October 2010, Pernod sold wine brands, vineyards, stock, plant and equipment intended to be used for the ongoing manufacture and sale of the wine brands to Lion, and another company called Indevin Group Ltd. Indevin would own the infrastructure assets, own and grow fruit, and process the wine, providing these services at commercial rates to Lion. Lion would be responsible for bottling, packaging, marketing and sales of the wine. Prior to this agreement Lion had a relatively small wine business, but it is now in substantial competition with Pernod. Under the sale and purchase agreement Lion acquired about one-third of Pernod'swine brands.3 McGraddie v McGraddie [2013] UKSC 58, [2013] 1 WLR 2477.4 Austin, Nichols & Co Inc v Stichting Lodestar [2007] NZSC 103, [2008] 2 NZLR 141 at [16];Kacem v Bashir [201] NZSC 112, [2011] 2 NZLR 1 at [31] [33].[10] The parties also entered into a distribution agreement, under which Lion wasappointed as Pernod's distributor from 1 November 2010 until the transaction settledfollowing approval from the Overseas Investment Office.[11] During negotiations on the sale and purchase agreement Lion was allowed access to a virtual data room set up by Pernod for the purpose of conducting due diligence. The virtual data room included a mechanism for Lion to submit questions to be answered by Pernod. During this process, Lion wanted to find out about the terms on which Pernod sold the brands Lion was acquiring to its major customers.[12] We were told that there were two reasons why Lion sought this information. First, Lion would be taking over the relationship with those customers and it needed an indication of the terms upon which Pernod had been contracting with them. Second, Lion needed that information to validate its financial models about the value of the assets it was purchasing. Lion had conducted a modelling exercise to estimate the revenue Lion was likely to derive from the brands, and it used these estimates to determine the value it placed on the brands it was purchasing from Pernod. In order to validate the modelling exercise, Lion wanted some indication of the terms on which Pernod had supplied its major customers in the past.[13] Pernod was reluctant to disclose its trading terms to Lion because that information was highly confidential and valuable to Pernod, given that Lion was already a competitor in the wine distribution business. Moreover, the sale and purchase agreement did not involve the sale of all of Pernod's wine business, butonly part of it. Pernod was concerned not to compromise the brands it retained. The position was that Lion was already a competitor of Pernod and would, after the settlement of the transaction, be a much more significant competitor. Pernod also said it considered there may be competition law implications of competitors sharing pricing information in the manner requested by Lion.[14] A stalemate was reached in negotiations. Lion described the issue of therelease of Pernod's trading terms as a "deal breaker".5 A compromise was reached. Pernod provided limited information, subject to confidentially undertakings, to5 High Court judgment at [312].Mr Sutton, Lion's Commercial Manager of Marketing, and Mr Hutton, Lion'sPricing Manager. This information became known as the "trading terms file" (TTF).[15] The TTF contained information about Pernod's trading terms withProgressive for two of the brands being purchased by Lion: Lindauer and Lindauer Special Reserve. These were the two largest brands by revenue that Lion was purchasing, accounting for approximately [ ] per cent of the revenue from all the brands sold by Pernod to Lion. Progressive and the other major supermarket chain, Foodstuffs, accounted for about 80 per cent of sales of those brands.6 The TTFprovided information about agreed promotional prices, Progressive's margin and theresulting price payable to Pernod. With this information, and public information about overall pricing, Lion completed its model to determine the likely margin it would receive on sales. Following receipt of this information, Lion was satisfied that it was likely to receive value from the brands justifying the proposed contract price and the sale and purchase agreement was executed.Retailer margins[16] Most supermarket wine sales occur at discounted prices, supported by advertising and in-store displays. The supplier and the supermarket will generally agree to the terms of the promotion, and this will normally contain an agreed level of contribution, or "price support", that the supplier will give to the supermarket tofacilitate the agreed promotional price. An important part of these arrangements is the level of margin (that is, the difference between the price at which the wine is sold to the supermarket and the retail price) to which the supermarket is entitled in respect of the brand concerned.[17] The margin required by the supermarket is expressed as a percentage of the retail price excluding GST. When the retail price for a particular brand is fixed for a period of time (for example, a promotional week) there will be an agreement as to the margin to be derived by the supermarket. Generally, supermarkets would require a higher percentage margin on undiscounted sales, a lower margin on moderately6 Foodstuffs comprises three co-operatives but these are seen collectively as a single competitor to Progressive.discounted promotions, and an even lower margin on deep discount promotions. It is these percentage margins that are at the heart of the dispute in this appeal.Guaranteed Margin Agreement[18] Pernod was, at the time the sale and purchase agreement was entered into, party to a GMA with Progressive. Under the GMA, Pernod and Progressive agreed the overall average margin that Progressive would receive for all Pernod wine sold to Progressive in a particular period. In the year prior to execution of the sale and purchase agreement, Pernod and Progressive had entered into four GMAs applying for periods of one to three months as follows:(a) 7 December 2009 to 3 January 2010: agreed margin [ ] per cent;(b) 19 April 2010 to 27 June 2010: agreed margin [ ] per cent;(c) 28 June 2010 to 26 September 2010: agreed margin [ ] per cent;(d) 27 September 2010 to 1 January 2011: agreed margin [ ] per cent.[19] Pernod's witnesses explained that the increasing level of margin reflected theincreasing buyer power of Progressive because of an oversupply of wine for domestic sale, consequent upon falling exports and increasing production.[20] The GMAs were informally documented, usually in emails between Pernod and Progressive. The essence of the agreed arrangement was that, if the margin received by Progressive on all wines purchased by it from Pernod during a particular period was less than the agreed margin, Pernod would make a payment to Progressive to put Progressive in the position it would have been in if the agreed margin had been achieved. It was unclear if Progressive had to make a refund if the agreed margin were exceeded, but the evidence was that Pernod ensured that this did not occur.[21] Provision was made in the GMA for Progessive to retain the flexibility to discount unilaterally retail prices of wines supplied by Pernod, on the basis that such discounts were excluded from the GMA calculations.[22] As noted earlier, at the time the sale and purchase agreement was executed, the agreed margin under the GMA was [ ] per cent. Progressive had signalled that it intended to seek an increase in the agreed margin from 1 January 2011 to [ ] per cent.[23] Mr Sexton, Progressive's Merchandising Manager, who was called as awitness under subpoena by Pernod and was described by the Judge as "animpressive, articulate witness ... plainly concerned only to provide the Court with asmuch assistance as he could",7 explained the nature and purpose of the GMA as follows:Since late 2009, Progressive and Pernod Ricard have agreed a margin. This provides an agreed overall margin between Progressive and Pernod Ricard for the wine it supplies. Under this arrangement, individual promotions may be at different margin levels provided that the overall margin which Progressive receives over all of Pernod Ricard's products delivers the agreedmargin level. Progressive has always reserved the right to change retails where they see fit and on these occasions the products in question are removed from the margin calculation.The margin arrangement does not result in Pernod delivering a significantly higher overall margin than other vendors.If there was no margin arrangement in place, we would require the same overall margins from Pernod Ricard and so they would receive the same overall price. The difference would be that, without a margin arrangement,Pernod Ricard's business would be a lot more complex to manage and whenyou complicate a business you invariably sell less product and therefore make less money.This is because the margin arrangement provides many practical benefits to Pernod Ricard and allows Pernod Ricard to maximise promotional and non promotional sales volumes. These are:(a) It gives Pernod Ricard much greater flexibility in agreeing promotions to suit its sales objectives for different brands at different times.(b) It takes the complexity out of the process of agreeing promotional slots and margins.7 High Court judgment, above n 2, at [347].(c) It gives Pernod Ricard better access to Progressive'smanagement and creates a better overall relationship.For Progressive, the advantage of the margin arrangement is that it is much easier to administer. We do not need to monitor margins on individual brands and promotions as closely, because the arrangement ensures we receive our overall margin requirements regardless.[24] Pernod did not disclose the existence of the GMA to Lion. At the heart ofLion's claim is its contention that this omission rendered misleading the information Pernod provided to Lion in the TTF. Pernod's reluctance to provide Lion withpricing information (commercial confidentiality and competition law concerns) applied also to its non-disclosure of the GMA. However, its sensitivity was not shared by Progressive. Progressive disclosed the existence of the GMA to Lion on the day after the sale and purchase agreement was signed.Alleged breach of warranties[25] Lion claimed that the non-disclosure of the GMA breached warranty 3 and warranty 4.[26] Under warranty 3, Pernod warranted that:The information in the Disclosure Information and the Trading Terms File is, taken as a whole so that all relevant information in the Disclosure Information and the Trading Terms File is taken into account with respect to the particular subject matter, materially accurate and is not materially misleading in its context, in each case subject to any limitations or qualifications contained in the relevant information. The warranty contained in this paragraph 3 does not apply to:(a) the Financial Information;(b) any information that is a budget, forecast, estimate, projection, model, yield projection, price forecast, or any assumption, statement of opinion or statement of intent;(c) the information memorandum relating to the Sale Assets contained in the Data Room;(d) any information that has been obtained from any of the public databases referred to in clause 15.4(c) of this Agreement.[27] The "Disclosure Information" included all the documents and information inthe virtual data room, information contained in a disclosure letter, and the TTF.[28] Under warranty 4, Pernod warranted that:[Pernod] is not aware of any circumstance which has not been disclosed in writing to [Lion] or [Indevin] and which if disclosed to a proposing purchaser for value of the Sale Assets might reasonably be expected to lead that purchaser to reduce materially its assessment of the value of the Sale Assets as a whole.[29] Lion claims that the non-disclosure of the GMA breached both of these warranties because:(a) in breach of warranty 3, without knowledge of the GMA the pricing information in the TTF was materially inaccurate or misleading; and(b) in breach of warranty 4, the GMA was a circumstance that would have caused Lion, acting reasonably, to materially reduce its assessment of the value of the sale assets as a whole.[30] The giving of warranties by Pernod was governed by clause 14 of the sale and purchase agreement. Clause 14.1 provided that Pernod gave the warranties to Lion on the date of the agreement, subject to the qualifications and limitations in clause 14.[31] Under clause 14.3, Lion acknowledged and agreed as follows:(a) No additional warranties[Pernod] has not made any representation or warranty (express or implied) in connection with the Agreement or the Sale ... other than [the warranties contained in Schedule 1] ...(b) Relied on own judgementin entering into this Agreement and proceeding to Completion [Lion] relies solely on its own judgement, investigations and professional advice received and does not rely on any statement, undertaking, representation or warranty of any kind, other than [the warranties in schedule 1].(c) Damages sole remedyto the fullest extent permitted by law:(i) No other cause of actionthe only relief available to [Lion] arising from any breach of a [warranty] is the right to the purchase price paid for the Sale Assets adjusted (by way of payment to the relevant party of a refund) under clause 15.9 ... and [Lion] is not entitled to any other or separate cause of action for damages or other relief ...[32] Clause 15.9 provided that any monetary compensation received by Lion as a result of any breach by Pernod of any warranty was deemed to be in reduction and partial refund of the purchase price.[33] The parties were agreed that the claim was a claim for expectation damages. The parties agreed that the correct measure of expectation damages was the amount that represented the difference between the value of the assets being purchased assuming no breach of warranty (that is, assuming the warranted information was true) and the value of the assets as represented (that is, no GMA).8 This focuses on value at the date of the sale and purchase agreement, because any breach of the saleand purchase agreement would have occurred on that date. As Lion's written submission put it: "Pernod effectively agreed to sell (and Lion agreed to buy) theassets on the basis that there was no GMA. In fact, there was a GMA. The loss isthe difference between what Lion agreed to buy and what was actually delivered".[34] While we agree that is the correct approach to the assessment of expectation damages, there is a need for caution in its application in this case because of the nature of the transaction. That measure is reasonably easy to apply when the warranty is about the state of repair of a chattel, for example. But where the warranted information is about past trading terms in circumstances where the value ascribed to the assets is based on future cash flows from trading on terms set by the buyer that are different from those applied in the past by the seller, the exercise of measuring the loss may become more problematic.Contractual context[35] Counsel for Lion, Mr Farmer QC, argued that the contractual context also included the confidentiality undertaking given by the two Lion executives,8 See Burrows, Finn and Todd, Law of Contract in New Zealand (4th ed Lexis Nexis 2012) at [21.2.2(a)].Mr Hutton and Mr Sutton, to whom the pricing information at issue in this case was to be disclosed. Mr Farmer referred to the fact that, under the terms of the confidentiality undertaking, Pernod agreed that one of the purposes for disclosure ofthe confidential information was "confirming [Lion's] key pricing and margin assumptions." He said this made it clear that Pernod was aware that Lion was using the information for the purposes of validating its model.[36] We disagree. While Pernod did agree that this was one of two purposes for which the information was disclosed, the wording of the confidentiality undertaking makes it clear that it was disclosed only for the two specified purposes and the Lion executives involved then undertook that they would not use the confidential information applied to them by Pernod for any other purpose and, in particular, not for pricing purposes.[37] It seems to us, therefore, that the reference to the purpose for the disclosure of the information and the confidentiality undertaking was a limitation on its use, rather than an acknowledgment by Pernod of the significance of the information to Lion. This is confirmed by the fact that the information disclosed in the TTF demonstrably could not validate the model, because it was so selective and did not claim to be representative of sales of all brands to all retail outlets.[38] Another important point of context was that the disclosure was made in October, and related to intended margins for Lindauer in the period up to Christmas. Lindauer was by far the most important of the brands being sold, and it has great popularity in the Christmas season. The period covered by the disclosure (October to December) straddled the date of the sale and purchase agreement. Sales after thedate of the sale and purchase agreement were to Lion's account but at prices alreadyagreed between Pernod and Progressive. This context lends support to Pernod'sposition that the disclosure related only to Lindauer and did not purport to give any information about other brands or any basis for assumptions about other brands.Materiality[39] Both warranty 3 and warranty 4 include the qualifying adverb "materially".In the High Court, the Judge said that information that Lion said should have been,but was not disclosed by Pernod, would be "material" if Lion, acting reasonably,would be likely to attach some significance to it and would therefore have taken it into account.9 He drew on a number of cases in the insurance context in order to reach that conclusion.10[40] We put it to counsel that, in the present context, the test adopted by Cooke J in Coleman v Myers was more appropriately applied, given the similarity of the subject matter.11 The test adopted by Cooke J was as follows:As a broad test of materiality, then, one may speak of "those considerationswhich can reasonably be said, in the particular case, to be likely materially toaffect the mind of a vendor or of a purchaser." The same idea is expressedmore fully by Marshall J in delivering the opinion of the United States Supreme Court in TSC Industries Inc v Northway Inc 426 US 438 (1976), a case under the Securities Exchange Act, of 1934 and concerning proxy solicitation:The general standard of materiality that we think best comports with the policies of rule 14a-9 is as follows: An omitted fact is material if there is a substantial likelihood that a reasonable shareholder would consider it important in deciding how to vote ...What the standard does contemplate is a showing of a substantial likelihood that, under all the circumstances, the omitted fact would have assumed actual significance in the deliberations of the reasonable shareholder. Put another way, there must be a substantial likelihood that the disclosure of the omitted fact would have been viewed by the reasonable investor as having significantly altered the'total mix' of information made available.[41] Counsel accepted that this was the appropriate test. We do not see the outcome of applying it as altering the conclusion on materiality reached by the Judge.12Warranty claim based on written answer[42] Before turning to the principal plank of Lion's case, we will briefly deal witha preliminary matter. This concerns a written answer given by Pernod to a question9 High Court judgment, above n 2, at [360].10 At [358] and [359].11 Coleman v Myers [1977] 2 NZLR 225 (CA) at 334.12 See also the judgment of Heath J in R v Moses HC Auckland CRI-2009-004-1388, 8 July 2011 at [45]–[51].from Lion about "below the line" commitments that would have to be honoured. Pernod's answer was that the only such commitments were the "standard promotional slots". This was a reference to promotions that had been agreed withretailers for the period after settlement of the sale and purchase agreement. These were honoured by Lion.[43] Mr Farmer argued that this answer was incorrect or at least incomplete because it did not refer to the GMA. Mr Galbraith responded that this had not been pleaded and, in any event, the answer was correct. There was no need to refer to the GMA because the question was directed at commitments that would have to be honoured by Lion. Lion did not have to honour the GMA.[44] Mr Farmer's answer to that was that the standard promotional slots did not have to be honoured by Lion either. It is not clear to us whether Lion's decision tohonour those promotional slots was voluntary or not. But even if it was, that would indicate that the standard promotional slots did not need to be mentioned in theanswer, not that the GMA needed to be. We accept Mr Galbraith's point that thequestion was limited to binding arrangements. The GMA was not binding on Lion, so the answer did not need to refer to it.What was disclosed (and warranted) in the TTF?[45] As mentioned earlier, the negotiations between Lion and Pernod had reachedsomething of an impasse because of Pernod's refusal to disclose pricing information to Lion. The interaction between Lion and Pernod and their advisers in the period leading up to the development of this impasse provides some understanding of the impasse and of the action taken by Pernod to resolve it.[46] Pernod issued an information memorandum to potential bidders in late April2010. Lion's Board approved Lion making a non-binding indicative offer on 27 May 2010. At that time the transaction was to be structured as the sale of a company (Lindauer Wine Company), but ultimately became a sale of assets rather than of a newly-established company.[47] In early June 2010, Lion was invited by Pernod to participate in stage 2 of thesale process, which provided for Lion to undertake due diligence. Pernod'sinvestment bankers, Cameron Partners, set up a "virtual data room", on whichinformation about the brands to be sold was posted. However, this information did not include the margins on sales of the brands to be sold. The absence of this data was a matter of concern to Lion and was the subject of ongoing discussions between representatives of Lion and Pernod. It was also identified by Lion as one of the risks in the transaction from its point of view in its due diligence sales team report.[48] Lion and Indevin submitted a revised non-binding indicative offer on 22 July 2010, but the offer was stated as being subject to due diligence, particularly, confirmation of key commercial trading terms with domestic and international customers. The non-disclosure of margin and volume data by Pernod continued to be a sticking point between the parties. The negotiation of the price was occurring in tandem with those discussions.[49] A paper presented to the Board of Lion on 23 September 2010 sought approval to proceed with the transaction subject to satisfactory resolution of duediligence matters. The paper recorded Pernod's unwillingness to disclose trading terms information. Lion's Board approved the transaction, subject to a number ofmatters, including satisfactory completion of outstanding due diligence.[50] Further meetings occurred between Lion and Pernod representatives in late September and early October, but no resolution was reached. Then, on 5 October 2010, it was agreed that Mr Sutton and Mr Hutton of Lion would sign confidentiality undertakings and would then be given access to certain commercially sensitive information on the basis that they would keep the material confidential. Lion thensought "net-net pricing" for four brands: Lindauer, Lindauer Special Reserve, Saints and Corbans White Label, which were the four largest brands that would be sold to Lion. The net-net price was the recommended retail price, minus GST, minus theretailer's margin. The significance of this was that it allowed Lion to obtain anunderstanding for those brands of the margin that Progressive required in relation to the relevant products from retail price points.[51] Eventually, it was agreed that data would be disclosed only in relation to the two biggest brands, Lindauer and Lindauer Special Reserve. Information was disclosed showing the margins on those two products for the previous two year period. After further discussion, Pernod agreed to disclose pricing information for Lindauer and Lindauer Special Reserve for the period between the date of disclosure and Christmas 2010. The information was the maximum retail price in the relevantweek, the retailer margin and the "net net price". No mention was made of theGMA. The tables that were disclosed were as follows:LINDAUER 750 ML PELRRP Retailer Margin Net Net Price4-Oct-1011-Oct-10 $9.99 [ ]% $[ ]18-Oct-10 $9.99 [ ]% $[ ]25-Oct-101-Nov-10 $10.99 [ ]% $[ ]8-Nov-1015-Nov-10 $9.99 [ ]% $[ ]22-Nov-1029-Nov-10 $10.99 [ ]% $[ ]6-Dec-1013-Dec-10 *** *** ***20-Dec-10 *** *** ***27-Dec-10 *** *** ***LINDAUER SPECIAL RESERVE 750MLPELRRP Retailer Margin Net Net Price4-Oct-1011-Oct-1018-Oct-10 $14.99 [ ]% $[ ]25-Oct-10 $14.99 [ ]% $[ ]1-Nov-108-Nov-10 $14.99 [ ]% $[ ]15-Nov-1022-Nov-10 $14.99 [ ]% $[ ]29-Nov-106-Dec-10 $14.99 [ ]% $[ ]13-Dec-1020-Dec-10 *** *** ***27-Dec-10*** pricing yet to be submitted[PEL is an abbreviation of the name of Progressive;RRP is "recommended retail price".][52] The evidence was that the margin for both Lindauer and Lindauer Special Reserve in the weeks for which no entry appears in the above tables (that is, the margin that applied when those products were sold at their full retail price) was over [ ] per cent. However, only very small quantities were sold at that price.[53] It is notable that the information disclosed in relation to Lindauer Special Reserve relates only to the price point of $14.99, at which the retailer margin is [ ] per cent. This can be contrasted with the information disclosed in relation to Lindauer, where the retailer margin differs substantially between that applying at the higher price point, $10.99, ([ ] per cent) and that applying at the lower price point of $9.99 ([ ] per cent).What Lion did with the disclosed information[54] Under the confidentiality undertaking, only two Lion employees, Mr Sutton and Mr Hutton, were allowed to see the information disclosed in the TTF about Lindauer and Lindauer Special Reserve.[55] Lion's financial modelling in the early part of due diligence was largelyconducted by Mr Julian Waller and Mr Sutton. One of the initial models was known as the PMC file, which was a model that took pricing and volume data to forecast the net revenue that Lion could generate from the brands. The PMC file contained three price points for each of the brands: EDP (everyday price), Promo (standard promotional price), and KSP (key selling period, that is, a deep discount price). For each of those prices the model contained an estimate of the retailer margin, the retail price, and the net-net price (that is, the price the supplier would receive, called the"landed price" in the PMC file). In order to determine accurately the revenue Lioncould expect to make on the brands, Lion wanted to verify its assumptions aboutPernod's and Progressive's trading terms, and in particular to verify the marginsrequired by Progressive.[56] After receiving the TTF, Mr Hutton took the PMC file and created a new model incorporating the TTF information, referred to as the "pricing acquisition model" or the "integration model". In this integration model Mr Hutton forecast themargins that Progressive (and other customers) would require from each brand. From that he calculated an overall margin across all brands. Then, using the publicly available AC Nielson data on volumes of each brand of wine sold at each price, he could calculate a forecast revenue figure. Incorporating the TTF information into the integration model resulted in a forecast net revenue of $[ ] million. This was [ ] less than earlier modelling prior to the TTF had produced but this was not considered ofsufficient magnitude to affect Lion's most recent indicative offer price.[57] The PMC file before receipt of the TTF contained the following assumptions for Progressive on retail prices and retailer margins:[58] [REDACTED][59] Following receipt of the TTF, Mr Hutton made the following changes in his integration model (the changed figures are underlined and in bold type):[60] [REDACTED][61] The changes Mr Hutton made from the PMC file in the integration model can be summarised as follows:(a) No change to retail prices and retailer margins for EDP price point.(b) At the Promo price point:(i) Retail prices of Lindauer, Lindauer Special Reserve and Bernadino were raised by $1.00. The new retail prices for Lindauer and Lindauer Special Reserve were taken from the TTF, although the TTF did not expressly state that those prices were for the Promo price point. The retail prices for other brands were not in the TTF.(ii) Retailer margin for Lindauer at $10.99 was adjusted from [ ] per cent to [ ] per cent, in line with the TTF.(iii) Retailer margin for Lindauer Special Reserve at $14.99 was adjusted from [ ] per cent to [ ] per cent, in line with the TTF.(iv) Retailer margins for Corbans White Label, Saints and Bernadino were adjusted significantly upwards to [ ] per cent. There was no information in the TTF on these brands.(v) Retailer margins for Corbans Homestead and Timara were adjusted significantly upwards to [ ] per cent. There was no information in the TTF for these brands.(vi) No change to retail price or margins for Aquila (remained at [ ] per cent margin). There was no information in the TTF for this brand.(c) At the KSP price point:(i) Retail prices of Lindauer, Lindauer Special Reserve, Corbans Homestead and Bernadino were raised by $1.00. The new retail price for Lindauer was taken from the TTF, although the TTF did not expressly state that that price was for the KSP price point. The new retail prices for the other brands, including Lindauer Special Reserve, were assumptions and not contained in the TTF.(ii) Retailer margin for Lindauer at $9.99 was adjusted from [ ] per cent to [ ] per cent, in line with the TTF.(iii) Retailer margin for Lindauer Special Reserve at $11.99 was adjusted from [ ] per cent to [ ] per cent. Information for this price point was not contained in the TTF.(iv) Retailer margins for all other brands (except Aquila, which remained the same) were adjusted significantly upwards to [ ] per cent. These were not in the TTF.[62] All of these changes, and the final verification of the model, happened within a short timeframe. The TTF was first received by Mr Hutton at 2:27 pm on Thursday 14 October 2010. At 3:18 pm on Friday 15 October 2010 Mr Hutton informed Mr Sutton and Mr Craig Baldie that he and Mr Glass (Lion's SalesDirector) were comfortable with their model and the forecast net revenue to Lion. The sale and purchase agreement was executed on Monday 18 October 2010.[63] Mr Sutton stated in his brief that Lion used the TTF data to confirm its assumption that retailers obtained different margins at different price points, and to confirm the assumption that there was a tiered pricing offer. He said:6.8 The Trading Terms File also confirmed our assumption that the retailers obtained different margins at different price points. It also confirms that there was a tiered pricing offer, ie that there were different retailer margins at different price points. We assumed these included:(a) when sold at "every day" prices, [ ] per cent;(b) when sold at standard promotional prices, [ ] to [ ] per cent; and(c) when sold at "key selling period" promotional prices, [ ] per cent.Margins on the Lindauer brands[64] The margins used at the Promo price points for both Lindauer and Lindauer Special Reserve were taken from the TTF. The Lindauer margin at KSP was taken from the TTF. The margin on Lindauer Special Reserve at KSP was not taken from the TTF. Mr Hutton said that he did not regard the assumption that the margin on Lindauer Special Reserve at KSP would be [ ] per cent as speculative. The historical margin data in the TTF showed that Special Reserve had been sold at $11.99 or lower in previous peak summer trading periods and the [ ] per cent margin was simply applied from the equivalent discount price for Lindauer. Mr Hutton regarded this to be a reasonable assumption given his belief that the TTF was representative, and that [ ] per cent was within the range of margins provided during peak periods in the historical TTF data. We agree that the assumption of a margin of [ ] per cent for Lindauer Special Reserve at KSP was reasonable, although we find it surprising that this was not confirmed with Pernod.Why were lower margins assumed for non-Lindauer brands?[65] Mr Hutton explained that only two margins were used for Corbans White Label and Bernadino because Lion was planning to offer only a Promo discount for those brands. For Saints, Mr Hutton stated that Lion planned to offer only a [ ] per cent margin, and if the retailer wanted to run a Promo slot, then the retailer coulddecide to increase the retail price in order to achieve a retailer margin of [ ] per cent. In cross-examination he was asked why he had assumed margins of [ ] per cent at the Promo price point for Corbans White Label, Saints and Bernadino (and [ ] per cent for Aquilla) when the evidence was that supermarkets would likely accept lower margins on brands that drove foot traffic such as Lindauer. Mr Hutton stated that for these brands he had assumed that Progressive could achieve a [ ] per cent margin by raising the retail price at its end. Mr Sutton confirmed this explanation. We do not consider that there is anything in the TTF information that would give a basis for the assumption that Progressive would act in this manner.Why were retail prices raised in the integration model?[66] Mr Hutton was asked why he had increased the retail price on some brands in the integrated model, but assumed that the volume sold would stay the same, when elasticity modelling showed that a $1 change in retail price would alter sales volumes. Mr Hutton said he and others working on the due diligence had arrived at a set of pricing that they believed could apply in the market. The retail price changes in the model to Lindauer and Lindauer Special Reserve were consistent with the retail prices shown in the TTF. He indicated that he considered retail price increases for other brands to be consistent with the changes made to the Lindauer brands. Again, this assumes the TTF information about Lindauer was representative of all brands and the assumption about what pricing would apply in the market does not appear to be based on information in the TTF.The effect of the GMA on volume mix[67] After the TTF information was incorporated into the integration model, Mr Sutton calculated the weighted average margin that Progressive would earn on all the brands. Mr Sutton calculated weighted average margins for Lindauer at [ ] per cent, for Lindauer Special Reserve at [ ] per cent, and for all the brands purchased at [ ] per cent. Mr Farmer pointed out that Pernod pleaded that the average margin for the Lindauer brands was [ ] per cent, which is still lower than the agreed margin under the GMA.[68] An expert witness called by Pernod, Mr Hagen, made the point that, using the TTF information, Lion had forecast average margins for the Lindauer brands that were above that required by the GMA (that is, the GMA produced more favourable margins for Lion than they had predicted using the TTF). Mr Hagen's opinion evidence was that while Mr Sutton's weighted average margin across all the brandswas lower than that required by the GMA, this was because of Lion's unreasonableassumptions about the margins Progressive would require on the non-Lindauer brands, not because of the non-disclosure of the GMA in the TTF.[69] Mr Sutton's response was that the GMA changed Lion's volume mixassumptions, and that when the effect of volume mix is taken into account the weighted average margins in the integration model reduce. Mr Sutton said that Mr Hagen did not appreciate that the assumptions underlying Lion's integration model could not simply be applied to a margin guarantee arrangement. Mr Sutton'sevidence was that had Lion known about the GMA, its volume mix assumptions would have changed resulting in the following weighted average margins:(a) For Lindauer the weighted average margin would have reduced from [ ] per cent to [ ] per cent.(b) For Lindauer Special Reserve the weighted average margin would have reduced from [ ] per cent to [ ] per cent.(c) For all brands the weighted average margin would have reduced from [ ] per cent to [ ] per cent.[70] Mr Sutton explained in his evidence in chief how the GMA impacted on this modelling. In essence, Lion assumed that retailers would undertake their own discounting of strong brands like Lindauer. He said Lion assumed that where, for example, Lion discounted to an intended retail price of $10.99, Progressive would, in relation to half of the supplies, further discount to $9.99. This would meanProgressive's actual margin would reduce from about [ ] per cent to about [ ] per cent. The purpose of this additional discounting would be to generate greater foottraffic in Progressive's supermarkets. The lower retail price would drive increasedretail sales. The assumption was based on Lion's own experience and on its analysisof the historic price points in the AC Nielson data. However, because the GMA meant Progressive would expect a margin of [ ] per cent on all supplies, the incentive to engage in this further discounting was diminished or eliminated.[71] Mr Galbraith pointed out that Lion's concerns about the impact of the GMAon volume mix was notably absent from its pleaded claim. He also questioned its logic, given that Lion did not intend to agree to a GMA. Why, he asked rhetorically, would Lion model its future cash flows on the basis of trading under a GMA when that was not going to occur?[72] Mr Galbraith also argued that the volume data on which Lion relied (obtained through AC Nielson) reflected actual sales during periods in which a GMA was in force, so if the GMA affected volume mix as Lion claimed, the data would have included that effect.[73] Mr Galbraith said Lion's volume mix argument did not take into account that the GMA excluded promotions that were not agreed between the supplier andretailer. Mr Sutton explained Lion's position. His evidence was that a unilateraldrop in the retail price by the retailer would harm the supplier because it would increase the volume sold. For example, if the supplier and retailer agreed to discount to $9.99, but the retailer then further discounted to $8.99, that would increase the volume of wine sold on discount. The supplier would have planned that X bottles of wine would be sold on discount at a lower profit to the supplier, in a GMA environment but because of the further drop in price X+Y bottles of wine would be sold at the discounted price.[74] Mr Sutton then went on to say that in this situation it meant that the supplier had to top up the retailer on the additional volume sold to the guaranteed margin. So, where there was a unilateral discount, even though the supplier did not have to fund the extra discount in retail price, the supplier would still have to pay the guaranteed margin on a greater volume of sales. However, it is unclear on the evidence whether this was more than a theoretical problem or, if it was, how big a problem it was in practice.[75] Mr Sutton's evidence was supported by Lion's expert witness, Mr Wilson.He said that Mr Hagen's analysis assumed the same sales volumes at particular pricepoints whether the margins were as disclosed in the TTF or in a GMA environment. Mr Wilson said that was an invalid assumption because in a GMA environment the retailer had an incentive to push up sales at lower price points, reducing thesupplier's margin. He said this meant that Lion's model would have had differentassumptions and risk factors if Lion had known of the GMA. He said incomplete information meant the modeller was unable to make the best selection of model or ofvariables to use in the model. He concluded that under a TTF scenario, "Lion maywell have viewed the future of the brands more optimistically than was possible under a [GMA] where the retailer's incentives around price points are quite different".[76] Mr Wilson's evidence was not accepted by the High Court Judge.13 Rather,he accepted Pernod's position that, under a GMA, it is the supplier that is in effectivecontrol of promotional slots, rather than the retailer. This was said to be so becausethe retailer had the comfort of the agreed margins under the GMA. Pernod's positionwas confirmed by Mr Sexton of Progressive and the Judge said he placed reliance on Mr Sexton's independent evidence.[77] Mr Farmer argued that volume mix was at the heart of the case. In effect, Lion believed it was buying brands from a trader that was operating on essentially the same basis as it was, with margins being agreed with customers on a product by product, price point by price point basis. That is why it made the assumptions it did about margins. If it had known about the GMA, it would not have done this. The fact that Pernod had a GMA with Progressive made it inevitable that the assumptions made by Lion were distorted. We will come back to that contention later.Other information available to Lion[78] During the due diligence period, Lion also undertook what is called a "Sales Due Diligence" exercise. The Lion executives responsible for the exercise submitted a report in June 2010 called the Sales Due Diligence Report. One of the findings in13 High Court judgment, above n 2, at [416].this report related to the margin Progressive was likely to require after the settlement of the sale and purchase agreement. It said:Customer margins – based off our PMC's & forecast we can meet the [ ] percent margin requested by Progressive (this increase or change to our current terms is expected now).[79] Surprisingly, Mr Sutton did not read the Sales Due Diligence Report and Mr Hutton was unaware of its existence. Allan J found that it ought to have beentaken into account by those responsible for Lion's modelling and that, if it had, themargin assumptions made by Lion may well have been modified.14[80] Lion also received emails from Progressive to wine suppliers because it was a supplier of wine to Progressive before the sale and purchase agreement was signed. An email sent by Progressive to Lion and others in August 2010 advised that Progressive required margins of [ ] per cent to [ ] per cent for pending promotions. A similar email sent in September 2010 indicated similar margin requirements. These emails were made available to Mr Sutton and Mr Hutton when they werefinalising Lion's model. Although the information in these emails was factored intoLion's "worst case" model, Allan J found that it had been discarded in the course of Lion's decision in the final model to assume revenue from the transferred brands of$[ ] million,15 which was the basis on which Lion confirmed that its price for the brands was sustainable.[81] As the Judge noted, these published margins were closely aligned to the assessment made in the Sales Due Diligence Report. The emails were also expressed in terms that contemplated no negotiation or horse-trading. The Judge rejectedLion's contention that the margins in the emails were a "wish list". Mr Sexton of Progressive made it clear that this was not the case. The Judge found that Lion'sNational Account Manager for Progressive, Mr Hoffman, had treated them as actual requirements by Progressive.16 The margins Lion agreed with Progressive in February 2011 for the non-Lindauer brands were very similar to those indicated in these emails.14 At [386].15 At [391]. The Judge did refer to this figure but presumably meant $[ ] million, which was the revenue figure derived from the integration model.16 At [395].[82] The Judge also noted that Mr Hoffman could have been consulted by Mr Sutton and Mr Hutton without their breaching their confidentiality obligations to Pernod. This did not happen. Mr Hoffman said in evidence that he would have advised against some of the low margin assumptions in the Lion model in relation to non-Lindauer brands if he had been consulted.[83] Although the margins Progressive was achieving on wine supplies made by Lion before the sale and purchase agreement was signed appeared lower than those it was signalling in the emails referred to earlier, the Judge considered the emails werea better guide to Progressive's likely future requirements.[84] Overall, the Judge considered that Lion's approach to its modelling exercisewas too narrowly focused. In preparing the integration model, Mr Sutton and Mr Hutton should have taken into account the information from the Sales Due Diligence Report, the Progressive emails and the knowledge Mr Hoffman had aboutProgressive's likely expectations. The PMC model, which became the integrationmodel following changes made after receipt of the TTF, forecast annual revenue from the brands of $[ ] million. When Lion did take into account the informationfrom the above sources, in its "worst case" model, the forecast revenue was $[ ]million lower.[85] As we have noted at [4] above, the Judge found against Lion in relation to whether breach of warranty 3 caused any loss to Lion. The above facts supported theJudge's finding on causation. We address causation further below: see for example,[115] and [120] [123].Breach of warranty 3?[86] Lion's case is that the non-disclosure of the GMA meant that the information disclosed about retail margins in the TTF was not materially accurate and was materially misleading in its context.1717 The text of warranty 3 is set out at [26] above.[87] The substance of Lion's position is that the existence of the GMA changedthe dynamics of the relationship between Pernod and Progressive in a manner which was not apparent to Lion because of the non-disclosure of the GMA. Lion thought it was purchasing brands that were sold on a product-by-product and promotion/price point by promotion/price point basis, which was consistent with the way Lion itself operated and also the basis on which it had constructed its model. Lion relied on the evidence of Mr Wilson and Mr Sutton, summarised above, to the effect that its model would have been constructed in quite a different way had it known about the GMA.Lion's claim is that it would have valued the business purchased from Pernod differently and either would not have gone ahead with the purchase or would have done so, but at a reduced price.[88] Allan J determined this issue in Lion's favour.18 He concluded that the information about Lindauer margins for sales to Progressive in the TTF wasmaterially misleading in its context because the existence of the GMA was of "actual significance to Lion" and would have been of significance to any person in Lion'sposition, acting reasonably. He concluded this was the case because it was"inevitable that the existence of the GMA ... must have had some impact on Progressive's approach to negotiations over pricings and margins."19[89] Pernod said that the finding in the High Court judgment that a breach of warranty 3 had occurred was wrong. It argued that the relevant information in the TTF was materially accurate and not materially misleading in its context. In this respect it supported the orders made in the High Court but on a different ground than those appearing in the judgment.[90] Much of Pernod's case focuses on the limited nature of what was disclosed,and the limited purpose of that disclosure. Its counsel, Mr Galbraith, pointed out thatLion's written submissions in support of its appeal repeatedly stated as fact that thelimited information disclosed by Pernod about the Lindauer brands was representative of all the brands sold by Pernod to Lion, when, in his submission, it was not. He pointed out that Lion itself did not treat the information about the18 At [361].19 At [362].Lindauer brands as representative when it recalibrated its model. We accept that submission. Lion assumed margins for non-Lindauer brands that can be explained only by the fact that Lion did not treat the information disclosed by Pernod about Lindauer margins as representative of margins on the sales of all brands.[91] In our view the record of what Lion did with the information it received in the TTF establishes clearly that it made a number of assumptions and deductions from the information it received that could not fairly be said to arise from the information received. Our analysis at [54]–[63] above establishes that.[92] Mr Galbraith said that Lion knew that Pernod was not disclosing its overall trading terms for brands other than Lindauer, and Lion had simply made its own assumptions, for which it had to take responsibility, in relation to other brands.[93] Mr Galbraith submitted the Judge's conclusion that the GMA impacted onLion did not reflect the fact that the GMA would not be binding on Lion after thesale. In fact, the GMA applied to all of Pernod's brands, not just those being sold toLion, and was an arrangement that Pernod saw as advantageous from its point of view, just as Progressive did. But Lion was under no obligation to agree to a GMA with Progressive after the sale, and the events subsequent to the sale showed that Lion did not, in fact, enter into a GMA with Progressive. Rather, it continued the product-by-product and promotion/price point-by-promotion/price point method of operation that it had previously adopted in relation to its sales of product to Progressive and which it had used as the foundation for its model.[94] Mr Galbraith also argued that the non-disclosure of the GMA could not fairly be described as material because the margins that were disclosed in relation toLindauer were, on average, higher than those required by the GMA. It was Lion'smisuse of the information in its model, and the assumptions that it made in relation to other brands when validating its model that had led to Lion gaining a false impression of the margins that were likely to be sought by Progressive. He relied on the evidence of Mr Hagen, to which we have referred above at [68].[95] We agree with Mr Galbraith that the fact that the GMA was an arrangement that suited Pernod and Progressive but was not necessarily a model for future trading between Lion and Progressive is significant. If the GMA was an arrangement that attached to the brands and would have bound Lion after the sale, then obviously its non-disclosure would have been highly material. That would have been the case, for instance, if Lion had purchased Pernod in its entirety, and the trading terms between Pernod and Progressive continued unchanged after the sale. But the present context is quite different from a standard sale and purchase of a business, because of the factthat only some of Pernod's business was being sold, and that the sale of that businesswas on terms that allowed the buyer to set its own agenda with its customers in the future. Another important aspect of this commercial context was the fact that Pernod was remaining in the market as a significant player and was likely to face significant competition from Lion. Lion was already a player (albeit a relatively small one) in the wine distribution market and would, as a result of the acquisition, become a much more significant competitor of Pernod.[96] Mr Galbraith pointed out that the evidence of Mr Sexton, Progressive'sMerchandising Manager, was that the agreed margin in the GMA was what Progressive would expect to receive, whether a GMA was in place or not. Therefore it could be said that it did not make a significant difference to the basis on which Pernod traded with Progressive. Against that, however, Mr Sexton accepted in evidence that the GMA was a significant piece of information for a buyer of the brands that were sold by Pernod.[97] Mr Galbraith placed reliance on the evidence of Pernod's expert, Mr Hagen. As noted earlier, Mr Hagen pointed out that the weighted average margin of the sales of Lindauer as disclosed in the TTF was [ ] per cent, and [ ] per cent for Special Reserve, that is, higher than the [ ] per cent agreed margin under the GMA. Mr Hagen's evidence in this regard was disputed by Lion's expert witness, MrWilson, on the basis we have already described at [75].[98] The question of a breach of warranty 3 needs to be closely considered in the context of the wording of the warranty itself and the commercial negotiations leading up to the sale and purchase agreement. While Allan J found that the GMAwas significant to Lion and that it must have had some impact on Progressive'sapproach to negotiations over pricings and margins, the question that needed to be answered was whether its non-disclosure was, to use the words of warranty 3, "[not] materially accurate and ... materially misleading in its context." The question is notwhether the GMA was significant to Lion, but whether its non-disclosure meant the TTF was materially inaccurate or materially misleading, in the sense described inColeman v Myers.20 Would the information about the GMA have assumed actualsignificance in the deliberations of a reasonable purchaser in Lion's position?[99] In answering that question, regard must be had to the fact that the lengthy discussions between Pernod and Lion about disclosure of trading terms wascharacterised by Pernod's steadfast refusal to disclose trading terms between it andthe supermarkets. Whether the reasons for that refusal were valid is beside the point. The fact is that Lion was fully aware throughout the negotiation process that its modelling was being done on the basis of sales data received from AC Neilson and assumptions about margins at various pricing points that were based, at least to someextent, on Lion's own experience and assumptions about the market, but did not take into account information about Progressive's margin requirements that was availableto Lion.[100] When arrangements were made for disclosure of the TTF, the arrangement was highly circumscribed, being limited to only two of the brands being sold (albeit the most popular) and only in relation to a limited time period, being the period between the date of the contract and Christmas 2010. As mentioned earlier at [51]–[52] the information related only to two price points (one in respect of Lindauer Special Reserve) and said nothing at all about the margins on the other brands. Only limited historic information about the volumes of sales at different price points was given.[101] We accept that Pernod knew that Lion had a model and wished to have somesales data to "validate" its model. But Pernod did not make any representation thatthe information it gave was representative, that is, that it would provide a comprehensive picture of the margins obtained in relation to all the brands being20 See [40] above.sold at various price points or the likely demands of Progressive for margins on suppliers after the date of the agreement.[102] Although the Judge found that the GMA was of actual significance to Lion,he concluded at [417] that "Lion, acting reasonably, would have accorded [the GMA] very little weight". That conclusion, in the context of the causation analysisand the analysis of warranty 4, is hard to reconcile with the Judge's earlierconclusion that a breach of warranty 3 had occurred.[103] In our view, the information in the TTF about Lindauer and Lindauer Special Reserve was not materially inaccurate or materially misleading in the sense described at [98] above. The context did not change that. It did not purport to be comprehensive or representative information. Indeed, Lion did not treat it as such when it used the disclosed data, and made assumptions based on it, in the preparation of the integrated model. It was given on the basis that Pernod was refusing to give any more comprehensive information and Lion was left to make its own judgments about future margins. It did not involve any representation about the overall relationship between Pernod and Progressive.[104] The GMA was a short term arrangement between Pernod and Progressive. Its continuation was a matter of choice between supplier and purchaser, and there was nothing to require Lion to enter into a similar arrangement as, indeed, it subsequently refused to do. If the information provided in the TTF had purported to be representative information about all of the brands then the non-disclosure of the GMA would have been materially misleading. But we conclude that the information in the TTF was limited in scope and, within that limitation, was accurate and notmisleading. We therefore uphold Pernod's argument on this point.Breach of warranty 4?[105] Warranty 4 is much more generic in its wording.21 Lion's case is that if theGMA had been disclosed before the sale and purchase agreement was entered into, it would have led Lion to reduce materially its assessment of the value of the brands.21 The text of warranty 4 is set out at [28] above.It argued that its position in relation to the GMA was the position of a reasonable purchaser.[106] Allan J found against Lion in relation to warranty 4. The Judge saw the issue of whether non-disclosure of the GMA might reasonably have been expected to lead a reasonable purchaser to reduce its assessment of the value of the brands to a material extent as tied up with the issue of causation in relation to the breach of warranty 3. He therefore undertook his causation analysis, to assess the extent to which the breach of warranty 3 that he had found to have occurred (the information in the TTF being materially misleading) had caused loss to Lion, before dealing with the issue as to whether there had been a breach of warranty 4.[107] Mr Farmer was critical of this process of reasoning, because he said it conflated the issues of breach and causation. We accept there is some validity in that criticism, because the actual analysis of warranty 4 tended to be something of a footnote to the causation analysis in relation to the finding of a breach of warranty 3.[108] The core of the Judge's finding on causation in relation to warranty 3, whichled directly to his finding of no breach of warranty 4, is contained in paragraphs [417]–[421] of the judgment under appeal:[417] I held earlier that Pernod was in breach of the para 3 warranty because, as I note above at [361], it was materially misleading not to disclose the GMA. Nevertheless, had information about the GMA been available to Lion before the agreements were concluded, I consider that Lion, acting reasonably, would have accorded it very little weight. The GMA applied only for limited periods of time, never more than a few months. It was notbinding on Lion. Moreover, it applied across the whole of Pernod's winerange, the greater part of which was remaining with Pernod. The disclosure of the GMA would no doubt have been of great interest to Lion, but having regard to the other information available (and in particular the notified requirements of Progressive which did indeed foreshadow very closely theultimate result of Lion's margin negotiations with Progressive) it is impossible to say as a matter of practical common sense that the failure todisclose the GMA has led to Lion's loss. Had Lion paid sufficient attentionto other market information already available to it, it would in my opinion have modelled likely future revenue at a figure that has in fact provedachievable in the light of Lion's later negotiations with Progressive.[418] A further matter to be taken into account is Pernod's position. Indisclosing the Trading Terms File and choosing not to disclose detail of the GMA, Pernod was entitled to expect that Lion would have regard to all of the information already within its possession. Pernod was disclosing itsmargin arrangements for one client and two brands only. In my view, Pernod was entitled to expect Lion not to rely on that information alone, but to call upon its own extensive history and experience in the wine distribution industry.[419] Ultimately, Lion appears to have achieved margins in respect of the Lindauer brands that were consistent with those which the Trading TermsFile might have suggested would be appropriate. Progressive's negotiatedmargins for other brands were higher than Lion expected, but in my opinion, there was no proper basis upon which Lion was entitled to assume that Progressive would have been content with lower margins for those other brands.[420] Standing back and adopting a practical common sense approach tothe question of causation, I am not satisfied that Pernod's failure to disclosethe GMA to Lion caused Lion's claimed losses.[421] It follows that I am also satisfied that there has been a breach of para 4.1. In my view, armed with the knowledge that Lion had, disclosure of the GMA would not have reasonably caused Lion to reduce its assessment of the value of the sale assets as a whole. If Lion had been acting as a reasonable purchaser, it would have attached far more weight to the information it already possessed and the GMA would have had little or no impact.[109] The evidence from Lion's witnesses was that disclosure of the GMA wouldhave led it to value the brands differently and would have led it to either not complete the purchase or to offer a lower price. The question as to whether a breach of warranty 4 occurred therefore turns on whether Lion was acting reasonably, that is, whether a reasonable purchaser would also have assessed the value of the business materially differently (and at a materially lower level) if it had known about the GMA prior to the purchase of the brands.[110] The High Court Judge concluded that if it had been acting reasonably, Lion would have accorded little weight to the GMA, had the GMA been disclosed to it, for the reasons he set out at [417] of his judgment. We now consider each of those reasons in turn.GMA applied for limited periods of time[111] It is correct that the GMA applied for limited periods of time, as highlighted at [18] above. But from 19 April 2010 until 1 January 2011, GMAs were effectively rolled over as soon as they expired. The agreed margin, however, differed for eachperiod. We do not see this as a particularly significant factor in itself, but it does emphasise that there was no long term commitment by Pernod to the GMA and the rollovers presumably reflected the fact that both Pernod and Progressive saw it as being advantageous to them. Mr Ramounet, Finance Director of Pernod, and Mr Sexton of Progressive confirmed that they saw the GMA as beneficial to their respective companies. The evidence of Mr Sexton was that, whether there had been a GMA or not, Progressive would have been looking for overall margins of the same magnitude across the Pernod range.The GMA was not binding on Lion[112] We see this as a particularly significant factor. If Lion had purchased a company that owned the brands and had been bound by the terms of the GMA, particularly if the GMA was a long term arrangement, then its claim in relation to the non-disclosure of the GMA would have been unassailable. But, in fact, it was buying the brands on the basis that it would then enter into new terms with Progressive that were suitable to it and Progressive. What Pernod had chosen in the past to do in its relationship with Progressive for the whole of its wine range would not necessarily determine what Lion and Progressive would agree for Lion's range,being the brands Lion purchased from Pernod and the brands it already had.[113] Lion's claim is essentially premised on the basis that the agreed margin underthe GMA at the time of the sale ([ ] per cent) and the fact that the agreed margins have progressively increased and were probably going to increase again left Lion inthe position where it was faced with Progressive's expectations on margins framed by the GMA. But Progressive's expectations on margins would have also reflected the margins it was obtaining from other wine suppliers, including Lion, that is, the market power it could exert over its suppliers. Mr Sexton confirmed thatProgressive's margin expectations were the same, whether a GMA was in place or not. As the Judge found, Lion had other sources of information which indicatedwhat Progressive's expectations were. Lion's model was assessing the futurerevenue after the implementation of the sale and purchase agreement, not the historical margins achieved by Pernod in its sales to Progressive in the previous year.The fact that the agreed margins had fluctuated significantly in that year showed how the past could not be said to be a proper indicator of the future.The GMA applied across Pernod's range[114] We agree that this is significant, because Pernod was selling only a proportion, about 30 per cent, of its brands and was continuing to trade with Progressive under the GMA for the remainder. The GMA involved a global margin across that range, rather than applying to individual brands.Lion had other information available to it[115] We agree with the Judge that a reasonable purchaser in the position of Lion would have taken into account the information Lion had about trading with Progressive in preparing its model. We have dealt with this already at [78] [84]above. We agree with the Judge's assessment of that information and his conclusionthat a reasonable purchaser would have taken it into account in its modelling. Given that this information would have, if taken into account, signalled the likely margins Progressive would require after the settlement of the sale and purchase agreement, it was not possible to say as a matter of practical common sense that the disclosure of the GMA would have led a reasonable purchaser to reduce materially its assessment of the value of the brands. As the Judge said, if Lion had paid sufficient attention to that information, it would have modelled its likely future revenue at a figure similar to what was later proven to be achievable in its negotiations with Progressive. Thisis a fatal blow to Lion's case in relation to warranty 4.Pernod was entitled to expect Lion to consider other information[116] We agree with the Judge's conclusion that, since Pernod was disclosing onlymargin arrangements for two of the Lindauer brands as sold to Progressive, it was entitled to assume that Lion would not treat this as representative of all sales to all retailers of all brands. As we have demonstrated at [54]–[63], Lion's application ofthe disclosed information in the preparation of its integrated modelled involved numerous assumptions which were not, and could not be, fairly drawn from thedisclosed information. It was a recurring theme in Lion's written submissions thatthe disclosed information was representative of all brands, but that was simply not the case and not represented to be the case. Indeed, it was obvious from the context that Pernod was refusing to disclose margin information about all the brands. Pernod maintained the position throughout the negotiations (at least up until the compromise regarding the confidentiality undertaking) that no pricing information would be released. How Pernod managed its margins as between it and Progressive was obviously commercially sensitive.[117] The information that Pernod did disclose was limited to Lindauer and to the trading window in the period up to Christmas, where Lion had a real interest because it was going to have to honour the position that had been agreed between Pernod and Progressive.22 It did not purport to represent the ongoing position in relation to Lindauer, and said nothing at all about the other brands.Conclusion: warranty 4[118] We conclude, as the Judge did,23 that, given the knowledge Lion had aboutProgressive's margin for wine trading and its future expectations, the disclosure ofthe GMA by Pernod to Lion could not have reasonably caused Lion to reduce its assessment of the value of the assets as a whole. We see this as arising from a combination of two factors. The first is the information Lion had available to it thatought to have alerted it to Progressive's expectations as borne out in the subsequent negotiations between Lion and Progressive. The second is that Lion simply misapplied the information it was given by Pernod. It drew conclusions from it that were not fairly open to it and it treated it as representative of all of the brands when it demonstrably was not.[119] We conclude, therefore, that the Judge was correct to find that there was no breach of warranty 4.22 The evidence was unclear as to whether this was a legal obligation or simply a practical reality, but Lion did in fact honour it.23 High Court judgment at [421].Causation[120] Our conclusion that there was no breach of either warranty 3 or warranty 4 makes it unnecessary to deal with causation. There is some artificiality about doing so when no breach has been found. However, in case the matter goes further, it may be helpful to summarise our views briefly.[121] The High Court Judge found that there was a breach of warranty 3 but it had not caused loss, essentially for the same reasons that he (and we) found there was no breach of warranty 4. Mr Farmer said the judge had wrongly found there was anovus actus interveniens because Lion failed to take into account information it had available to it and thus broke the chain of causation. He said this misrepresented the nature of the claim, namely that the value of the brands with a GMA was lower than the value of the brands without one, as Lion thought was the case.[122] In fact, Allan J simply made a common sense analysis of the facts and concluded that the cause of the loss was not the failure to disclose the GMA (whichAllan J had found was a breach of warranty 3) but Lion's own actions in thepreparation of its model, that led it to make assumptions about margins on supplies of non-Lindauer brands to Progressive that could not fairly be attributed to the very limited information about Lindauer that Pernod agreed to disclose. In effect, the Judge found that Lion should have and, if acting reasonably would have, assessed the value of the brands without a GMA as essentially the same as the value of the brands with a GMA. The fact that it did not was not fairly attributable to the non- disclosure of the GMA.[123] We agree with Allan J's analysis. If we had found a breach of warranty had occurred, we would have found that the breach did not cause any loss.Loss[124] We take the same approach to this issue and set out our views in summary form. Again, we acknowledge that there is a degree of artificiality in doing so.[125] In broad terms, we accept Pernod's argument that Lion has not proven thatthere is any difference between the value of the brands assuming disclosure of the GMA and the value of the brands assuming non-disclosure of the GMA. That follows from our conclusion that warranty 4 was not breached.[126] Lion's loss calculation involved what Mr Hagen called a "meaningless exercise" of calculating the amount that Pernod would have had to pay toProgressive as top up payments under the GMA if there had been a GMA throughout the 12 month period before the signing of the sale and purchase agreement (there was not, in fact)24 and if the agreed margin under that GMA was [ ]per cent throughout the period (in fact, the agreed margins varied)25.[127] We do not see how that could provide a basis for the calculation of expectation loss. It was common ground that the value of the business was calculated by attributing a present value to its future cashflows. In those circumstances, the focus of any loss calculation has to be on predicted future cashflows, not past ones.[128] Another problem with this approach is that it assumes the top up paymentsmade by Pernod were a cost to it. But Pernod's evidence was that it managed themargins to ensure that top ups were payable by it to Progressive, not vice versa.[129] In our view the proper measure of loss required a valuation of the brands applying the information available to Lion, and a valuation applying that information plus the fact that a GMA with an agreed margin of [ ] per cent was in place.26 For the reasons we have given in our analysis of warranty 4, that exercise would reveal that the two were essentially the same and that no loss was incurred by Lion.24 See [18] above.25 See [18] above.26 That view accords with that expressed by Lion in its submissions. We do not consider thatLion's methodology is consistent with that submission.Result[130] We conclude that there was no breach of warranty 3 and uphold the argumentto that effect by Pernod. We also uphold the High Court Judge's finding that therewas no breach of warranty 4 and dismiss the appeal.Costs[131] Costs should follow the event. We order Lion to pay Pernod costs for a standard appeal on a band A basis and usual disbursements. We certify for two (not three) counsel.[132] Much of the pricing information in this judgment is confidential. We therefore make an order limiting distribution of the unredacted version of the judgment to the parties and their legal representatives. A redacted version will be made publicly available. To protect the confidential commercial information in the Court file, we make an order that no search of the file shall be allowed unless permitted by a Judge of this Court after hearing from the parties.Solicitors:Russell McVeagh, Auckland for AppellantBell Gully, Auckland for Respondent