Huntsman Chemical Company Australia Pty Limited v Narellan Pools Pty Limited

IN THE FEDERAL COURT OF AUSTRALIA NEW SOUTH WALES DISTRICT REGISTRY GENERAL DIVISION NSD 527 of 2010

BETWEEN: HUNTSMAN CHEMICAL COMPANY AUSTRALIA PTY LIMITED (ACN 004 146 338)

AND: NARELLAN POOLS PTY LIMITED (ACN 003 607 681)

JUDGES: MOORE, FLICK AND YATES JJ DATE: 1 February 2011 PLACE: SYDNEY

REASONS FOR JUDGMENT MOORE J: 1 I have had the considerable benefit of reading in draft the reasons of Flick J. I agree with the orders his Honour proposes and I generally agree with his reasons. Ultimately, the appeal fails (notwithstanding the primary judge's erroneous observation that the production operators were not called to give evidence) because the primary judge was correct in not accepting, as the appellant contends he should have, the production sheet provided compelling evidence completely answering the evidentiary case of the respondents. For reasons given by the primary judge, it was not evidence of that character. Whether the appeal could be disposed of in the way suggested by his Honour at [16] - [17] is a matter about which it is probably unnecessary to express a view. The appeal should be dismissed with costs. I certify that the preceding one (1) numbered paragraph is a true copy of the Reasons for Judgment herein of the Honourable Justice Moore.

Associate: Dated: 1 February 2011 IN THE FEDERAL COURT OF AUSTRALIA NEW SOUTH WALES DISTRICT REGISTRY GENERAL DIVISION NSD 527 of 2010

BETWEEN: HUNTSMAN CHEMICAL COMPANY AUSTRALIA PTY LIMITED (ACN 004 146 338)

AND: NARELLAN POOLS PTY LIMITED (ACN 003 607 681)

JUDGES: MOORE, FLICK AND YATES JJ DATE: 1 february 2011 PLACE: SYDNEY

REASONS FOR JUDGMENT FLICK J: 2 On 17 October 2008 there was filed in this Court an Amended Application and an Amended Statement of Claim. The Applicants claimed (inter alia) that the Respondent had engaged in misleading and deceptive conduct and relief was claimed pursuant to the Trade Practices Act 1974 (Cth) and also for alleged breaches of contract. 3 The Applicants, Narellan Pools Pty Limited and Narellan Franchise Pty Limited (together "Narellan"), were in the business of manufacturing and supplying fibreglass swimming pools. The Respondent, Huntsman Chemical Company Australia Pty Limited ("Huntsman"), was involved in the supply of chemical products, including "industrial resins". 4 In August 2007 an order was made pursuant to O 29 r 2 of the Federal Court Rules that the question of liability was to be determined separately and in advance of any determination as to damages. 5 The hearing took place over 16 days between October 2008 and June 2009. The primary Judge delivered his reasons for decision on liability in March 2010: Narellan Pools Pty Ltd v Huntsman Chemical Company Australia Pty Ltd [2010] FCA 267. Orders granting declaratory relief were made on 23 April 2010. Leave to appeal was granted on 26 July 2010. A Notice of Appeal was filed on 2 August 2010 appealing from "the whole of the judgment … handed down on 25 March 2010 at Sydney and the declarations and orders made … on 23 April 2010 at Sydney". 6 Narellan were successful before the primary Judge in establishing that a batch of chemical product supplied by Huntsman, a vinyl ester resin described as Hetron 942, was the cause of blistering in swimming pools it had manufactured and supplied. The Hetron 942 was used by Narellan in what was described as the "tie layer" as a barrier designed to prevent corrosion. It was common ground between the parties that the batch of Hetron 942 supplied was contaminated - but there was disagreement as to the source of the contaminant. Both parties maintained that they were not the source of the contaminant. 7 Huntsman now contends before this Court that a central conclusion reached by the primary Judge as to it being the source of the contaminant was a conclusion not open to him. It further contends that if it ultimately be concluded that no state of satisfaction can be reached that Huntsman was the source of the contaminant, then the case is one where, on the evidence before it, the Court should have found that Narellan had failed to make out its case rather than go on to conclude that Huntsman was, therefore, the source of the contaminant. Huntsman further contends that there has been a failure to provide reasons for that conclusion as to causation. Essentially, Huntsman contends that Narellan failed to discharge the onus of proving that the blistering was caused by the batch of Hetron 942 supplied. 8 The Grounds of Appeal focus attention in large part upon errors said to be exposed in paragraphs [174] to [176] of the primary Judge's reasons for decision. Those Grounds, however, also further contend that the manner in which the primary Judge proceeded constituted a denial of procedural fairness and that His Honour failed to provide reasons for his conclusions. 9 The appeal should be dismissed.

The Causes of Action Relied upon 10 Three representations were relied upon by Narellan to found their claims to relief. These representations were pleaded in the Amended Statement of Claim as follows: 9. The Respondent represented to the First Applicant in or around November 2000 ("the First Representations") that: a. Hetron 942/35 epoxy vinyl ester resin would be suitable for use by the First Applicant in use in the construction of fibreglass swimming pools; b. Adequate laboratory tests for Hetron 942/35 epoxy vinyl ester resin had been completed. … 11. The Respondent represented to the First Applicant from on or around 5 February 2001 until the use of the Resin 942 ("the Second Representations") that: a. the Hetron 942/35 epoxy vinyl ester resin which was to be (and from 26 February 2001 had been) manufactured and was proposed to be supplied ("Resin 942") was suitable for use by the First Applicant in the construction of fibreglass swimming pools. b. the Resin 942 proposed to be supplied to the First Applicant was the same as the Resin 942 supplied to the First Applicant in the Sample. … 12. The Respondent represented to the First Applicant from 19 September 2001 until the use of the Resin 942 ("the Third Representations") that: a. the balance of the Resin 942 which had not been supplied was suitable for use by the First Applicant in the construction of fibreglass swimming pools; b. the balance of the Resin 942 proposed to be supplied to the First Applicant was the same as the Resin 942 supplied to the First Applicant in the Sample; c. the balance of the Resin 942 which had not been supplied had been retested and was still "in spec". 11 The breaches of contract relied upon by Narellan included claims that there had been a breach of terms that the product supplied would be fit for purpose and of merchantable quality. 12 A sample batch of approximately 200 kg of the product had been provided by Huntsman to Narellan in or about late November or December 2000. Thereafter, in February 2001, Narellan placed an order for the supply of three tonnes of Hetron 942. The product in fact produced was thereafter supplied by Huntsman in two batches - one in March and the other in November 2001. As events unfolded, more product was in fact produced by Huntsman and accepted by Narellan than the three tonnes that had been originally ordered.

The Findings of the Primary Judge 13 Huntsman took issue with a number of findings made by the primary Judge. 14 But three relevant findings which assumed importance were: it was "beyond dispute that what was delivered to Narellan by the March Supply and the November Supply was not what was stated on the production sheet for Hetron 942": [2010] FCA 267 at [32]; it was "likely that the contaminant (which was uniformly present in a large proportion of the tie layer of the Resin 942 samples), was within the Resin 942 itself when it was delivered to Narellan in the two supplies": [2010] FCA 267 at [130]; and "the source of the contaminant in the quantity agreed by the experts is to be found in Huntsman's manufacturing process, rather than in some intervening act or default on the part of Narellan in the course of its fabrication process": [2010] FCA 267 at [176]. 15 At the very outset of the case before the primary Judge, and before this Court, Narellan identified 9 "key facts" which they maintained were determinative of all liability issues. These 9 matters, as extracted from the reasons for decision of the primary Judge, were expressed as follows: 1. Narellan has manufactured over 16,000 pools essentially without incident save for the pools manufactured with Resin 942. 2. Narellan's manufacturing processes were such "as to avoid incorrect resin application and/or contamination". 3. Huntsman has never supplied any of its Hetron 942/35 epoxy vinyl ester resin to any manufacturer before or since and only the Sample and the Resin 942 have ever been used in manufacturing swimming pools. 4. The Resin 942 (unlike the Sample) was all made in highly unusual and inadequately explained circumstances over three days. 5. The composition of the Resin 942 is uncertain because (a) "of an amount of 400 to 450kg that cannot be accounted for" and (b) because of a question as to the origin of components used. 6. The Sample (which was used successfully) "differed in composition from" the Resin 942 (which was used highly unsuccessfully). 7. A "signature" of a component consistent with terephthalate resin was present in the tie layers of affected pools; its origin is "unknown" and its presence in the tie layer is a contaminant and because of its location is unlikely to be present by reason of poor manufacturing technique. 8. The Resin 942 was supplied in two supplies and used (as resin in the tie layer): a. exclusively in 62 pools … b. together with another resin in 15 pools … 9. Of the pools that have failed to date all were pools in which Resin 942 was used in whole or in part. Huntsman did not agree that the evidence made out all of these facts. 16 Irrespective of the significance that may be attached to these 9 "key facts", there was no challenge on appeal by Huntsman to the findings that the product supplied was not what was stated on the production sheet or that the contaminant was within the Hetron 942 when delivered and that the source of the contaminant was not attributable to "some intervening act or default on the part of Narellan". The contaminant, it was accepted, was within the product supplied and was within the containers (known as "hazcons") when it was delivered by Huntsman to Narellan. 17 In such circumstances, it was difficult to see any reason why the appeal should not have been dismissed on that basis alone. It was difficult to see how any conclusion was open other than a conclusion that the product supplied to Narellan by Huntsman contained a contaminant and that the product was thus not in accordance with the representations that had been made or (alternatively) not fit for purpose and/or not of merchantable quality. That was the conclusion of the primary Judge and that was the substance of the declaratory relief granted by His Honour. 18 But substantial time was spent during submissions directed to the reliability of a "production sheet" recording the manner in which the Hetron 942 had been produced by Huntsman. 19 It is desirable that these submissions should also be addressed.

The Production Process and the Production Sheet 20 At the core of the case for Huntsman was the proposition that it could account for the entirety of the contents of the product supplied - other than for an amount of about 450 kg. If it could account for the entirety of the contents, whatever contaminant later found its way into the material used by Narellan, it could not be traced back to Huntsman. It must have come, so said Huntsman, from somewhere else. If that was the case, the source of the contaminant was unknown and - the case being evenly poised - liability could not be attributed to Huntsman. 21 Both Narellan and Huntsman relied upon expert evidence - the expert relied upon by Narellan was Professor Graeme George; the expert relied upon by Huntsman was Professor Robert Shanks. A conclave report of those experts concluded inter alia that the "estimated level of contaminant, if it is assumed to be a terephthalate resin, is around 30% which would be over 2000 kg". Terephthalate is, apparently, a thermoplastic polymer resin of the polyester family. Being a polyester resin it, apparently, had no role to play in the manufacture of Hetron 942, which is a vinyl ester. 22 The composition and ingredients of the Hetron 942 manufactured was thus the subject of extensive evidence, as was the composition of the product ultimately supplied to and used by Narellan. 23 The quantity of the product that was ordered by Narellan was 3,000 kg. To produce that amount of Hetron 942, it was common ground that 4,770 kg of raw materials had to be used. The formula or "recipe" was set out as follows in a document prepared by Huntsman and described as the "polyester manufacturing instructions": This document thus recorded the amounts of the individual ingredients to be used in manufacturing Hetron 942 and also the "process" whereby they were to be mixed. 24 The manner in which these instructions were said by Huntsman to have been followed for the batch supplied to Narellan was recorded in a "production sheet" that assumed prominence during the hearing at first instance and on appeal. Given that prominence and the handwritten notations and initials, it is reproduced in its entirety as follows: 25 Notwithstanding the fact that only 3,000 kg of Hetron 942 was ordered in February 2001, the production sheet discloses that 6,830 kg was produced - being "6 X 1000 kg" and "1 X 830 kg". 26 But the reason for the manufacture of this additional product matters not - according to Huntsman. What is of importance to Huntsman's case was that the production sheet can account for the entirety of all ingredients used, other than for about 450 kg. The explanation for the missing 450 kg or thereabouts advanced by one of Huntsman's witnesses that "some drums were added but not recorded on the sheet" was accepted to be "speculation" that could be "reject[ed] out of hand". 27 If the production sheet is accepted at face value, it discloses the content of the ingredients originally mixed - being 3,000 kg of H914, 1,260 kg of H922, etc. The total weight of those amounts is slightly in excess of 4,600 kg. The production sheet also relevantly discloses the specifications (or "specs") that the product had to satisfy. After the original ingredients were mixed, the product did not meet those specifications and further ingredients were sequentially added, namely: 100 litres of styrene; 100 litres of styrene; 50 litres of styrene; 50 litres of styrene; 1 drum of H914; 2 drums of H922; 1 drum of H922; and 3 drums of H914. The combined weight of these ingredients, according to Huntsman, accounted for the 6,830 kg produced - other than the amount of about 450 kg. His Honour, it would appear, was alert to the discrepancy in production quantity that may arise by reason of the need to add ingredients to bring the final product within the required specifications. During the evidence of Mr Knight there thus occurred the following exchange between His Honour and the witness: HIS HONOUR: Just before you ask that, I would just like to ask the witness a couple of questions. MR LEE: Yes, your Honour. HIS HONOUR: Mr Knight, when you are mixing these various products that your company makes, in your experience, following the first testing process, is it more likely to less likely that it will be at the first testing within spec?---That is generally unlikely to be the case. Right. So generally it is the case, is it, that at the time of first testing of a product - and I am not talking necessarily about the product which you are being cross-examined on closely this afternoon, I am talking about products across the range, and I am only asking you to tell me what your experience is?---Yes. Your experience is that at the time of first testing of whether it's within spec, it's more likely than not it will be out of spec in some - in respect of some indicia?---Yes. And when one looks at - have you got the bundle of documents there with you? If you look at 105, which is the particular batch we are concerned with in this case ---?---Yes. ---you will see there that there's a number of testing processes before the asterisks disappear off the items?---Yes. But if you go back to the previous page, to the particular product there that is being produced, do you see again that there's a number of testings that had to take place before it was brought within spec?---Yes. And that is more usual than not, is it?---Yes. Yes. Thank you. In this manner Huntsman accounted for all of the ingredients which were used and thereby excluded (in its submission) the prospect of Huntsman itself being the source of the contaminant that affected the Narellan pools. 28 Detailed evidence was also given on Huntsman's behalf as to the manufacturing process to exclude the prospect of human error and mistaken ingredients being used. Those involved were "trained to be vigilant". They tried to be "careful and conscientious" as "best as practical".

The "Patent Deficiencies" in the Production Sheet and Other Evidence 29 The difficulty for Hunstman was that the primary Judge found the "production sheet" to be infected with "patent deficiencies". His Honour relevantly concluded as follows: The Terephthalate Mystery [172] This matter has its provenance in the points of agreement of the experts at [2.1] of the second conclave report, in particular [2.1.1] to [2.1.6] … At [2.1.5] the experts agree: The estimated level of contaminant, if it is assumed to be a terephthalate resin, is around 30% which would be over 2000 kg. [173] Huntsman's description of it as a 'mystery' goes not so much to the identity of the contaminant, although the experts agreed it was 'uncertain' ([2.1.5]), but to how it got there in the quantity it did. Huntsman contended that it was not open for the Court to find that Huntsman tipped it in and in the absence of such a finding, or a finding as to how the contaminant got there in the quantity it did, then whatever caused the blistering, it was not Hetron 942; it was something else. According to Huntsman, as a matter of legal analysis, it may be seen as some form of novus actus interveniens which breaks the claim of causation. [174] It is clear from a number of 'the critical documents in the case', referred to by Narellan in opening, in particular those referred to at [55(1), (3), (4) and (5)] above (Ex 11: 25, 26, 34, 103, 105) that the Resin 942 was not manufactured in accordance with Huntsman's own specifications as prepared and approved by Dr Durrant. These specifications contemplated a yield of 4,770 kg of Hetron 942 per batch comprising 3,000 kg of Hetron 914 and 1,260 kg of Hetron 922 as well as other components in their quantities there listed. Only 3,000 kg of Hetron 942 was ordered by Narellan so that if the Resin 942 had been manufactured according to Huntsman's own specifications to yield 4,770 kg of Hetron 942, that would have been more than sufficient to fill Narellan's order. No explanation was forthcoming from Huntsman as to why 6,830 kg of Resin 942 was produced nor, apart from what was, or was not, recorded on the production sheet for Resin 942, the identity of the components of the additional quantity of 2,053 kg. The production sheet records that an additional four drums of Hetron 914, an additional three drums of Hetron 922 and an additional 300 litres of styrene were added but it was common ground between the experts, Professors George and Shanks, that despite these batch adjustments, there was another 400 - 450 kg of material, to yield 6,830 kg of Resin 942, which simply cannot be accounted for. Moreover, having regard to other patent deficiencies in the recordings made on the production sheet for the Resin 942, e.g., the absence of batch numbers for the Hetron 922 and Aerosil 202, and in the absence of Huntsman calling any of the operators responsible for the manufacture of the Resin 942, in particular GC and PK, I have significant reservations about the extent to which reliance can be placed on the production sheet for the Resin 942 as to the identity of the batch adjustments which were made and which resulted in an over-yield of some 2,053 kg. My reservations in this regard assume a greater importance, and indeed significance, in the face of the agreement of the experts, Professors George and Shanks, in the second conclave report that pools manufactured with the Resin 942 all have an infra-red signature that is consistent with a terephthalate impurity level of around 30% (see [95] above), namely, 2,049 kg (cf., 2,053 kg) of the total yield of 6,830 kg. [175] Accepting the agreed view of the experts that: 'Narellan has procedures in place to ensure that contamination cannot occur' ([2.2.2] of second conclave report); and that: 'Huntsman has procedures in place … reducing the risk of contamination' ([2.2.3] of that report), the fact remains that for the reasons referred to in [174] above, the production sheet for the Resin 942 is sufficiently doubtful in terms of its reliability and accuracy, that in the absence of evidence from those operators responsible for making the batch adjustments and writing up the production sheet, it is open for me to find, and I so find, that it does not properly record batch adjustments of over 2,000 kg. [176] For these reasons, and accepting, as I do, Narellan's submissions set out at [113] above, I am satisfied, on the balance of probabilities, that the source of the contaminant in the quantity agreed by the experts is to be found in Huntsman's manufacturing process, rather than in some intervening act or default on the part of Narellan in the course of its fabrication procedures. Particular emphasis was placed by Huntsman upon paragraph [174]. 30 At that paragraph, His Honour referred to there being: "No explanation … forthcoming from Huntsman as to why 6,830 kg of Resin 942 was produced…" His Honour there also referred to "other patent deficiencies in the recordings made on the production sheet" and instanced two, namely: "the absence of batch numbers"; and "the absence of Huntsman calling any of the operators responsible for the manufacture of the Resin 942". 31 On appeal, the explanation for the primary Judge stating that there had been a notable absence as to why additional Resin 942 had been produced was said by Huntsman to have been made clear. The implementation of the original recipe did not produce a product that was within specifications; additional material - being that identified - had to be added. 32 And the significance of "the absence of batch numbers" was a matter, not surprisingly, pursued in cross-examination at the hearing. Thus, Mr Durrant gave evidence on behalf of Huntsman. He was an organic chemist and was employed by Huntsman as a Senior Technical Specialist. He accepted that the entry of batch numbers was an "important part of the form" but resisted acceptance of a proposition that the omission of such numbers was "serious". The following exchange thus occurred in part with his cross-examiner as follows: Now, you would accept that the entry of batch numbers on the form is a very important part of the form?---Yes. And that has occurred in relation to all materials but two?---Yes. … It's an omission?---It's an omission but it's not serious. … You don't regard it as serious that one can't identify with precision the batch of some of the constituent elements that go into the production of this product?---Well, it's a minor fault, I would say. A minor fault?---Yes. Well, that minor fault means we don't know the HETRON 922 which was said to go into this product, we don't know the manufacturer of that HETRON 922?---We don't know the batch number. And we don't know the AEROSIL 202 batch number which went into this product?---No. And that's an undesirable state of affairs. You accept that?---Mildly undesirable. Now, isn't the entry of the batch numbers important because it's by reason of putting into the batch numbers that the plant operators are prompted, at that stage, to read the label on each material being used, to ensure that the correct one is added?---They don't go by batch numbers, they go by the actual name of the product. So it would be quite wrong for someone to say, would it, that by entering the batch numbers for the materials used the plant operators are prompted at that stage to read the label of each material being used to ensure the correct one is added?---I would think they would be better advised to read the chemical name rather than the batch number. Batch number doesn't tell them anything. … I see?---But it is handy when you - if there is a problem, it is handy for traceability, so you know which raw material, which batch of raw material was used. … It is imperative that plant operators are very vigilant in recording batch numbers?---It's not essential. 33 During the hearing before the primary Judge other "deficiencies" in the production sheet had also been identified. These were repeated on appeal as being: the inaccuracy in the date identified as "26/02/01". The inaccuracy was that the production of the Hetron 942 took place not on one day but over a course of three days; the "peak temp" not being recorded; and there being no indication as to where the additional drums of H914 and H922 came from. 34 Notwithstanding these deficiencies, Huntsman continued on appeal to defend the integrity of its manufacturing process. The deficiencies, it contended, did not expose it to the liability imposed by the primary Judge.

A Challenge to the Reasons Provided 35 Huntsman challenged the reasons of the primary Judge. It maintained that the reasons: included a finding which was manifestly wrong; and failed to refer to significant evidence adduced by Huntsman. Huntsman further contended that there was also: a failure to provide reasons for the conclusion that "on the balance of probabilities … a hydrophilic thixotrope was added to the Resin 942 which was responsible … for osmotic blistering". 36 The first challenge focussed attention upon His Honour's observation (at paragraph [174]) that Huntsman failed to call "any of the operators responsible for the manufacture of the Resin 942, in particular GC and PK". 37 It was common ground on the hearing of the appeal that the "production sheet" recorded 3 persons initialling and thereby assuming responsibility for different stages of the production process - namely, "PK", "GC" and "TS". These are the initials of Messrs Peter Knight, Gary Carlton and Terrence Sharland. Messrs Knight and Sharland swore affidavits which were read and both were cross-examined. Mr Carlton gave no evidence. Mr Knight explained in his affidavit Mr Carlton's involvement in the production process. The "production sheet" itself disclosed that it was Mr Carlton who added 1,260 kg of H922 at the outset and thereafter sequentially added styrene and drums of H922 to bring the product within specifications. The balance of the ingredients mixed at the outset were added by Mr Knight; Mr Sharland added a further 3 drums of H914 to bring the product within specifications. 38 His Honour was thus wrong to have observed that Huntsman failed to call "any of the operators responsible for the manufacture of the Resin 942, in particular GC and PK". He was correct to have observed that Huntsman failed to call Mr Carlton but incorrect in respect to Mr Knight. It was an error repeated by His Honour at paragraph [175] of his reasons for decision. The error assumes some significance as Huntsman took some effort to prove the integrity of its production process, including reliance upon the evidence of Messrs Knight and Sharland. 39 It is, however, a mistake to construe this sentence of His Honour's reasons divorced from the context in which it appears. The sentence - commencing with the word "[m]oreover" - was clearly supplementary to that which preceded it, namely the recounting of the additional ingredients that were later added and the unaccounted 450 or so kilograms of material. The inability to account for this material was "unprecedented". In the cross-examination of Mr Knight the following exchange thus occurred: I see. Now, has anyone ever told you that there's 450 kilograms of material in relation to this resin 942 that is unaccounted for?---Yes. Have you ever come across that ever in your experience?---No. Completely unprecedented in your experience?---Yes. It characterises the making of this particular batch, in your experience, the fact that there is something in that 6830 kilograms which you can't see by looking at what is being added on the manufacturing sheet - on the production sheet. Correct?---Correct. And the sentence upon which attention was focused by Huntsman was followed by His Honour's observations as to the excess yield of some 2,053 kg. His Honour recounted the agreement between the experts that pools manufactured with the product had "an infra-red signature that is consistent with a terephthalate impurity level of around 30% …, namely, 2,049 kg (cf., 2,053 kg) of the total yield of 6,830 kg". 40 The error in the reference to the failure to call operators, it is concluded, does not otherwise affect the integrity of the reasoning process of the primary Judge. A central finding made by His Honour was that the production sheet had "patent deficiencies". Irrespective of the failure to call Mr Carlton, those deficiencies remained. The failure to record the batch numbers for two of the ingredients originally mixed and the subsequent failure to record the batch numbers for the drums of H914 and H922 later mixed remained. The failure to record the "peak temp" following the additional additives remained. His Honour's reference to the failure to call "any of the operators" was but an instance of the "deficiencies" in the production sheet. That erroneous observation of the primary Judge, it is considered, was not prejudicial. And the "patent deficiencies" in the production sheet were but a part of a more generally expressed lack of satisfaction with the evidence of Huntsman. 41 Even where an error is exposed, an appeal may be dismissed in those circumstances where it may confidently be concluded that the primary Judge would have come, in any event, to the same conclusion: cf Bisley Investment Corporation v Australian Broadcasting Tribunal (1982) 40 ALR 233 at 246 per Lockhart J. Although Huntsman was correct to contend that the primary Judge had committed an error and that his statement as to the failure to call operators was indeed a part of his reasoning process (cf Vocisano v Vocisano (1974) 130 CLR 267 at 274 to 275 per Barwick CJ), the error was not such as to provide any sound basis for contending that His Honour would have reached any different conclusion had the error not been committed. 42 The attack founded upon His Honour's observation as to there being "no explanation" for the production of 6,830 kg of product, it is respectfully considered, fails for the same reason. There was, in any event, an explanation provided. 43 The second of the bases advanced by Huntsman challenges what it characterises as a failure on the part of the primary Judge to refer to evidence and his subsequent conclusion as to hydrophilic thixotrope being responsible for the osmotic blistering. 44 The duty of a Judge at first instance extends to a duty to refer to the evidence relevant to the submissions advanced and the findings relevant to the conclusions to be reached. But it is not a duty to refer to every submission and every piece of evidence. In Mifsud v Campbell (1991) 21 NSWLR 725 at 728 Samuels JA expressed the duty as follows: In Public Service Board of New South Wales v Osmond (1986) 159 CLR 656, the High Court (at 667) said that it was right to describe the giving of reasons as "an incident of the judicial process" although a normal but not a universal one. In Soulemezis v Dudley (Holdings) Pty Ltd (1987) 10 NSWLR 247, McHugh JA (at 278) makes some comments upon that holding, and goes on to say (at 281) that the failure to explain the basis of a crucial finding of fact involves a breach of the principle that justice must not only be done but must be seen to be done. Similarly, in my opinion, it is an incident of judicial duty for the judge to consider all the evidence in the case. It is plainly unnecessary for a judge to refer to all the evidence led in the proceedings or to indicate which of it is accepted or rejected. The extent of the duty to record the evidence given and the findings made depend, as the duty to give reasons does, upon the circumstances of the individual case. Accordingly, a failure to refer to some of the evidence does not necessarily, whenever it occurs, indicate that the judge has failed to discharge the duty which rests upon him or her. However, for a judge to ignore evidence critical to an issue in a case and contrary to an assertion of fact made by one party and accepted by the judge - as the defendant's denial of having consumed alcohol - may promote a sense of grievance in the adversary and create a litigant who is not only "disappointed" but "disturbed" - to use the words which appear in the New Zealand case of Connell v Auckland City Council [1977] 1 NZLR 630 at 634. It tends to deny both the fact and the appearance of justice having been done. If it does, as in my opinion is the case here, then it will have worked a miscarriage of justice and have produced a mis-trial and resulted in what I would take to be an error of law which is reviewable on appeal. Whether it is an error of law or an error of fact, it seems to me a failure by the judge to do what the nature of the office requires. Clarke JA and Hope A-JA agreed. The importance of providing reasons for a judicial decision cannot be denied: Beale v Government Insurance Office of NSW (1997) 48 NSWLR 430 at 441 to 442 per Meagher JA. 45 One of the issues pursued before the primary Judge was the cause of the osmotic blistering that did occur in the Narellan pools. One of the possible causes being advanced by Narellan was the prospect that a hydrophilic thixotrope, Aerosil 200, had been added to the H942 rather than Aerosil 202. The formula or "recipe" for the production of Hetron 942 provided that 50 kg of the Aerosil 202 was to be "blend[ed]" with the H914 and H922 in a "clean blend tank" in what was described as "the usual way". 46 In this context Huntsman now focuses attention, in particular, upon paragraph [127] of the reasons for decision of the primary Judge. That paragraph, together with paragraph [123], provides as follows: [123] The production sheet for the prior batch reveals that Wacker N20 was originally specified and replaced by Aerosil 200 (both hydrophilic thixotropes). Further, the evidence of the operator Mr Knight was that the person adding the thixotrope did not load this material into the hopper, it was added by a person on the previous shift. The batch of Resin 942 does not have a batch number attached for the thixotrope and there is evidence of the use of a hydrophilic thixotrope in the previous batch (which on the evidence is the same day as the batch of Resin 942 commenced). … [127] Given that a hydrophilic thixotrope was used in the previous batch in the same tank on the same day as the Resin 942 commenced production, there was real scope for such a thixotrope to have been mistakenly added to the Resin 942 and that this was something clearly considered by Huntsman and Professor Shanks. Although it is not necessary in order to conclude the case on liability, there is sufficient evidence to also conclude on the balance of probabilities that a hydrophilic thixotrope was added to the Resin 942 which was responsible (as it was in the Tranquility matter) for osmotic blistering. Huntsman submits that there was a failure to refer to evidence, including evidence as to the products having a distinctive packaging and the washing of the tanks between batches of product being manufactured. Huntsman contends that it "was an error for the trial judge to move, without reasons, from stating that there was 'real scope' that such a thixotrope could have been 'mistakenly added' to concluding 'on the balance of probabilities that a hydrophilic thixotrope was responsible … for osmotic blistering". 47 But Huntsman's submission is misplaced. Paragraph [127] does not record any finding or conclusion of the primary Judge. That paragraph is simply the end of that part of His Honour's reasons where he sets forth possible reasons for the failure of the batch of Hetron 942. Paragraph [120] of the primary Judge's reasons for decision is preceded by a heading: "Other possible reasons for the failure of the Resin 942". Thereafter His Honour deals with two possible reasons, namely: "1. Unaccounted for 400 kg to 450 kg of material" (paragraph [120]); "2. Possibility of a hydrophilic thixotrope added to the Resin 942 instead of Aerosil 202" (paragraph [121] to [127]). Paragraph [127], and the preceding paragraphs, it was submitted on appeal by Narellan - and not challenged by Huntsman - was a verbatim repetition of submissions previously made. 48 So much is apparent from a reading of His Honour's reasons for decision from paragraphs [120] through to [127]. It is also apparent from the immediately succeeding paragraphs where His Honour does in fact set forth his conclusions in respect to these particular submissions as follows: Conclusion on likelihood of the cause of failure due to a defect contained within the Resin 942 [128] Reference was made to the responses given by Dr Durrant in cross-examination extracted in [80] above. [129] There are a multitude of possible theories as to what could have happened during the manufacture of the Resin 942 at the Huntsman factory. Each one of the above theories cannot be excluded and is plausible - with some more likely than others. [130] As stressed above, it is simply impossible to understand with precision what went wrong: all that can be known is that it is far more likely than not that the problem arose from the Resin 942 supplied. It is not necessary to put it higher than it is likely that the contaminant (which was uniformly present in a large proportion of the tie layer of the Resin 942 samples), was within the Resin 942 itself when it was delivered to Narellan in the two supplies. It follows that the evidence is overwhelming that the cause of the blistering and failure of these pools was the Resin 942 itself. The reference to paragraph [80] of the reasons for decision was a reference to an extract of the evidence of Mr Durrant. The witness was there asked not to comment upon "what caused this problem" but rather whether it was "more likely than not that something in what was delivered as a result of this manufacturing process caused the problem". The answer to that question was "probably, yes". The subsequent conclusion in paragraph [130] "that the cause of the blistering and failure of these pools was the Resin 942 itself" was unexceptional. 49 To the extent that there was a more wide-ranging challenge to the inadequacy of His Honour's reasons, the submission is rejected. The primary Judge was not satisfied with the explanation provided by Huntsman. His Honour was not satisfied that the production sheet accurately recorded the ingredients mixed to produce the batch of the product that was supplied to Narellan and His Honour was not "impressed" with the expert relied upon by Huntsman: [2010] FCA 267 at [101]. The summary conclusion on the "Terephthalate Mystery" given at paragraph [176], records the basis upon which His Honour proceeded to find that Huntsman was liable for the contamination. 50 Huntsman was unsuccessful in satisfying the primary Judge as to the integrity of its manufacturing process. The reasons provided by the primary Judge, it is respectfully concluded, adequately set forth an account of the submissions advanced for resolution and His Honour's reasons for not accepting the account advanced on behalf of Huntsman.

An Evenly Balanced Case? 51 A final submission advanced on behalf of Huntsman thus need not be resolved. But it can, in any event, be summarily rejected. 52 Huntsman's further submission was that if the Court could not be satisfied, on balance, that the contamination was introduced at its plant by Huntsman it ought not impose liability on Huntsman for the blistering. This further submission contended that "[a]s a matter of law, a plaintiff does not establish its case by demonstrating that its own improbable explanation is fractionally less unlikely than the defendant's improbable explanation". 53 The analogy was sought to be drawn by Huntsman to the decision in Rhesa Shipping Co SA v Edmunds [1985] 1 WLR 948. The plaintiffs' vessel, the Popi M, had sunk. The plaintiffs claimed on insurance which covered loss caused by "perils of the seas". Various explanations for the sinking were propounded prior to the trial. The only explanation provided by the plaintiffs at trial was that the vessel had sunk after colliding with a submerged submarine. But the submarine was never detected and never seen. The judge at first instance and the Court of Appeal upheld the claim. The House of Lords allowed an appeal holding that the plaintiffs had failed to establish their claim. In so concluding, Lord Brandon of Oakbrook (with whom the other members of the Court agreed) said at 951: The second matter is that it is always open to a court, even after the kind of prolonged inquiry with a mass of expert evidence which took place in this case, to conclude, at the end of the day, that the proximate cause of the ship's loss, even on a balance of probabilities, remains in doubt, with the consequence that the shipowners have failed to discharge the burden of proof which lay on them. Thereafter, Lord Brandon identified as follows a number of reasons for concluding that the plaintiffs had not made out their case: The first reason is one which I have already sought to emphasise as being of great importance, namely, that the judge is not bound always to make a finding one way or the other with regard to the facts averred by the parties. He has open to him the third alternative of saying that the party on whom the burden of proof lies in relation to any averment made by him has failed to discharge that burden. No judge likes to decide cases on burden of proof if he can legitimately avoid having to do so. There are cases, however, in which, owing to the unsatisfactory state of the evidence or otherwise, deciding on the burden of proof is the only just course for him to take. The second reason is that the dictum can only apply when all relevant facts are known, so that all possible explanations, except a single extremely improbable one, can properly be eliminated. That state of affairs does not exist in the present case: to take but one example, the ship sank in such deep water that a diver's examination of the nature of the aperture, which might well have thrown light on its cause, could not be carried out. The third reason is that the legal concept of proof of a case on a balance of probabilities must be applied with common sense. It requires a judge of first instance, before he finds that a particular event occurred, to be satisfied on the evidence that it is more likely to have occurred than not. If such a judge concludes, on a whole series of cogent grounds, that the occurrence of an event is extremely improbable, a finding by him that it is nevertheless more likely to have occurred than not, does not accord with common sense. This is especially so when it is open to the judge to say simply that the evidence leaves him in doubt whether the event occurred or not, and that the party on whom the burden of proving that the event occurred lies has therefore failed to discharge such burden. After referring to this decision, in Kuligowski v Metrobus [2004] HCA 34, 220 CLR 363 at 386, Gleeson CJ, McHugh, Gummow, Kirby, Hayne, Callinan and Heydon JJ more succinctly observed: [60] … A failure to find a matter alleged does not establish the truth of the contrary of that which is alleged … See also: Paton v National Mutual Life Association of Australasia Ltd [2000] FCA 684 at [66] per Cooper J. 54 Huntsman's argument seeking to invoke these principles in the present appeal is rejected. The simple fact is that the primary Judge found that the evidence was not evenly balanced. He found that the account provided by Huntsman as to the integrity of its manufacturing process was wanting. His Honour concluded that, on the balance of probabilities, Narellan had made out their entitlement to relief: Evidence Act 1995 (Cth), s 140. See also: Chen v Zhang [2009] NSWCA 202 at [47] per Sackville AJA (Campbell JA and Handley AJA agreeing). His Honour was not choosing "between guesses, where the possibilities are not unlimited, on the ground that one guess seems more likely than another or the others": Jones v Dunkel (1959) 101 CLR 298 at 305. Dixon CJ there went on to state that "[t]he facts proved must form a reasonable basis for a definite conclusion affirmatively drawn of the truth of which the tribunal of fact may reasonably be satisfied". In the present appeal there was a "reasonable basis" for the conclusion reached by the primary Judge and for the reasons he expressed.

Conclusions 55 None of the Grounds of Appeal, including the asserted denial of procedural fairness, have been made out. Each of the issues canvassed during the course of the proceeding before the primary Judge has been addressed in His Honour's reasons for decision and the Appellant was afforded an opportunity to make submissions in respect to each issue. No error has been exposed which would warrant the appeal being allowed. 56 The appeal should be dismissed. 57 There is no reason why the normal rule as to costs should not prevail. The Appellant should pay the costs of the Respondents. I certify that the preceding fifty-six (56) numbered paragraphs are a true copy of the Reasons for Judgment herein of the Honourable Justice Flick.

Associate: Dated: 1 February 2011 IN THE FEDERAL COURT OF AUSTRALIA NEW SOUTH WALES DISTRICT REGISTRY GENERAL DIVISION NSD 527 of 2010

BETWEEN: HUNTSMAN CHEMICAL COMPANY AUSTRALIA PTY LIMITED (ACN 004 146 338)

AND: NARELLAN POOLS PTY LIMITED (ACN 003 607 681)

JUDGES: MOORE, FLICK AND YATES JJ DATE: 1 february 2011 PLACE: SYDNEY

REASONS FOR JUDGMENT YATES J: 58 I have had the advantage of reading in draft the reasons for judgment prepared by Flick J. I gratefully adopt his Honour's summary of the issues at trial and of the evidence before, and findings made by, the primary judge. I agree with his Honour's conclusion that the appeal should be dismissed with costs. 59 The appellant (Huntsman) did not challenge the finding that Resin 942, a bespoke product that had been applied to the swimming pools fabricated by the respondents (Narellan), was contaminated. Indeed, in this appeal, Huntsman submitted that the only satisfactory explanation for the cause of the osmotic blistering in the swimming pools fabricated by Narellan was that the Resin 942 was contaminated by around 2000 kilograms of terephthalate based compound. The primary issue that divided the parties at trial was the source of the contamination. 60 After reviewing the evidence in detail (including the considerable body of expert evidence) and recording, in detail, the competing submissions advanced by the parties, the primary judge rejected Huntsman's submission that the blistering was caused by Narellan's fabrication procedures: see, in this regard, [167]-[171] of the primary judge's reasons. Huntsman does not appeal from that finding. However it does seek to contest the further finding by the primary judge ([176] of the primary judge's reasons) that, on the balance of probabilities, the source of the contaminant in the quantity agreed by the experts was to be found in Huntsman's manufacturing process. 61 In this connection Huntsman submitted that it was not incumbent on it to come up with a "theory on contamination". It submitted that, in the face of two equally improbable alternatives, the primary judge erred by reasoning, in effect, that, if Narellan was not the source of the contamination, then Huntsman must be the source, even though no other rational explanation as to the cause or source of the contamination was advanced. 62 In support of its submission that it was not the source of the contamination, Huntsman contended that it could adequately account for how Resin 942 was actually manufactured, including, by way of explanation, how a significant overproduction of the resin had occurred. 63 Huntsman advanced two contentions based on what it submitted were errors relating to findings made by the primary judge. First, it submitted that the primary judge erred by finding that no explanation had been given for the overproduction of the resin. Secondly, it submitted that the primary judge erred in finding that Huntsman had failed to call any of the operators responsible for the manufacture of the resin, "in particular GC and PK". Each of these findings related to the primary judge's assessment of the reliability and accuracy of Huntsman's production sheet for Resin 942, which played a central role in Huntsman's defence that it was not the source of the contamination. 64 In my view the primary judge did err by finding that the operators responsible for the manufacture of Resin 942 had not been called. There were three operators recorded on the production sheet: "PK" (Peter Knight), "TS" (Terrence Sharland) and "GC" (Gary Carlton). Both Mr Knight and Mr Sharland were called as witnesses and were cross-examined. Mr Carlton was not called. 65 However, I agree with Flick J that this error does not provide a sound basis for contending that the primary judge would have reached a different conclusion had the error not been committed. It is plain that the production sheet contained a number of deficiencies to which the primary judge referred. Even with the benefit of the evidence given by Mr Knight and Mr Sharland (which the primary judge, in any event, noted elsewhere in the reasons - including the fact that each had given evidence in relation to the production sheet for Resin 942: see [85] and [86]), those deficiencies remained and were unexplained. 66 I also agree with Flick J that the attack based on the primary judge's finding that no explanation had been given for the overproduction of Resin 942, fails. 67 The primary judge's finding, in this regard, must be seen in the context of, and is explained by, the immediately succeeding sentence in his Honour's reasons (at [174]), in which his Honour correctly recorded that the production sheet showed that four additional drums of Hetron 914, three additional drums of Hetron 922 and an additional 300 litres of styrene had been added as batch adjustments in the production of the resin. However, importantly for present purposes, the primary judge focussed on the critical fact that it was common ground between the experts that, quite apart from these batch adjustments, another 400 to 450 kilograms of material (that is, to yield the 6830 kilograms of resin ultimately produced) "simply cannot be accounted for". 68 The notable failure of the production sheet to account for 400 to 450 kilograms of otherwise unknown material was the focus of the primary judge's finding in this regard. That matter, together with the deficiencies in the production sheet, both as to the constituents used (such as the absence of batch numbers for Hetron 922 and Aerosil 202) and process steps taken (such as the failure to record peak temperature when certain constituents were added), as noted by Flick J, plainly pointed to the fact that the production sheet was not an accurate or complete statement of how Resin 942 had been produced. And yet it was critical to Huntsman's case that it be able to give a cogent account of the production of Resin 942 in order for it to establish, as it sought to do, that it was improbable that it was the source of the contamination. The cogency of that account rested squarely on an acceptance of the production sheet as a contemporaneous record giving an accurate and reliable account of the process steps taken, and constituents used, in the production of Resin 942. The production sheet could not fulfil that purpose. 69 Huntsman also submitted that the primary judge erred in finding (at [127] of the primary judge's reasons) that a "hydrophilic thixotrope" had been added to Resin 942 and was responsible for the blistering. However, as Flick J points out, Huntsman's submission is misplaced because no such finding was made by the primary judge. Narellan's submissions on appeal made clear that this "finding" was, in fact, no more than a recounting by the primary judge of the submissions which it had advanced at trial, which canvassed a number of possible causes for the contamination consistent with Huntsman being the source. The primary judge's analysis of the parties' competing submissions commences from [157] of his Honour's reasons. The conclusion expressed in [177] makes clear his Honour's position. His Honour said: For all the foregoing reasons, and notwithstanding that the evidence does not permit me to find the precise constitutional deficiency or deficiencies in the Resin 942 that caused the osmotic blistering, I am satisfied, on the balance of probabilities, that the cause of that blistering in pools manufactured by Narellan was the use of Resin 942, either alone or mixed with another vinyl ester resin, in the 'tie layer' of those pools. It follows that the primary judge did not err as Huntsman has contended. 70 I agree with Flick J that Huntsman's contention that the primary judge was faced with two improbable alternatives for the source of the contamination founders on the assumption on which it is based. In light of the deficiencies in the production sheet and the revelation that 400 to 450 kilograms of unknown material had been added to the disclosed constituents and could not be accounted for satisfactorily, the integrity of Huntsman's intended manufacturing process for Resin 942 was seriously called into question, with the result that the case was not evenly balanced, as Huntsman contended. 71 In my respectful view it was not only open to the primary judge to find that the source of the contamination was to be found in Huntsman's manufacturing process, but his Honour was correct in so concluding on the balance of probabilities. I certify that the preceding fourteen (14) numbered paragraphs are a true copy of the Reasons for Judgment herein of the Honourable Justice Yates.