Anti-Corrosion Pte Ltd v Berger Paints Singapore Pte Ltd and another appeal
[2011] SGCA 57

Case Number	:	Civil Appeal Nos 224 and 240 of 2010 (Suit No 989 of 2009)
Decision Date	:	02 November 2011
Tribunal/Court	:	Court of Appeal
Coram	:	Chao Hick Tin JA; Andrew Phang Boon Leong JA; V K Rajah JA
Counsel Name(s)	:	Philip Fong Yeng Fatt, Jonathan Yuen Djia Chiang and Joana Teo Swee Ling (Harry Elias Partnership LLP) for the appellant in CA No 224 of 2010/respondent in CA No 240 of 2010; Ang Cheng Hock SC, Sathiaseelan s/o Jagateesan, Kenneth Lim Tao Chung and Ramesh Kumar Ramasamy (Allen & Gledhill LLP) for the respondent in CA No 224 of 2010/appellant in CA No 240 of 2010.
Parties	:	Anti-Corrosion Pte Ltd — Berger Paints Singapore Pte Ltd

[LawNet Editorial Note: This was an appeal from the decision of the High Court in [2010] SGHC 351.]

2 November 2011

Judgment Reserved

V K Rajah JA (delivering the judgment of the court):

Introduction

1       This appeal and cross-appeal concern a dispute between a paint manufacturer and a painting subcontractor on whether latent defects in the supplied paint led to the serious discolouration of the internal surfaces at various parts of a building project. The subcontractor sued for losses, ie, expenses which it had incurred in repainting the affected surfaces. The manufacturer in turn made a counter-claim for the balance sum due on the paint sold to the subcontractor.

2       A High Court judge (“the Judge”) dismissed the subcontractor’s claim and allowed the manufacturer’s counterclaim for the sum of $72,676.62; see Anti-Corrosion Pte Ltd v Berger Paints Singapore Pte Ltd [2010] SGHC 351 (“the judgment”). In dismissing the claim by the subcontractor, the Judge found at [19] of the judgment that “the [appellant] has not proven on balance of probability [sic] that the defects in [the paint] had caused the discolouration”. Pertinently, at [21] of the judgment, the Judge also determined:

… that if I had to decide on the issue of the warranty, I would have found that [the Respondent]] had warranted all its products used in any of the plaintiff’s projects for a period of 5 years provided that the use was based on a proposed paint system by the defendant.

Both parties now appeal against different aspects of the judgment below. The subcontractor contends that the Judge had erred on the issue of causation while the manufacturer is appealing against the above-mentioned dictum of the Judge on a warranty purportedly given by it to the subcontractor. We now give our decision and the reasons for it.

Background

Parties to the dispute

3       The subcontractor is Anti-Corrosion Pte Ltd (“the Appellant”). It provides construction and renovation services, which include paint application works. Its managing director is Lim Choon Lin Vincent (“Vincent Lim”). Among the Appellant’s personnel, he was the one who liaised most often with the Respondent’s representatives.

4       Berger Paints Singapore Pte Ltd (“the Respondent”) is the paint manufacturer and supplier. Its sales representative, Joseph Yong, conducted the sale negotiations with Vincent Lim. Rajeev Goel is the Respondent’s regional technology manager who was in charge of the Respondent’s investigations after the paint discolouration occurred. Its superior was Jaideep Nandi, the chief executive officer of the Respondent.

Facts

5       The Respondent supplied the Appellant with paint on three separate occasions prior to the project in question. The first time it did so was in 2005 for the painting of the external surfaces of a Toh Guan dormitory. The project was completed without any complaints.

6       Subsequently, in January 2006, the Respondent proposed to the Appellant and supplied a paint system for the Toh Guan Road East Capital One project (“the Toh Guan project”). Unlike the earlier project, the painting works were for external and internal surfaces. The Respondent’s Decora Emulsion paint (“the paint”) was used for the internal surfaces. The Respondent’s product data sheet stated, inter alia, that the Appellant was to “[a]pply a suitable sealer coat such as Berger Plastaseal or Berger Water-Based Sealer” prior to the application of the paint.^{[note: 1]} Despite this, the Respondent’s proposed paint system for internal surfaces did not include a sealer coat. It merely stated:^{[note: 2]}

2.    Internal Concrete Surfaces Including Ceiling

Primer Coat    : 1 coat of Berger Decora Emulsion

Finishing Coat    : 1 coat of Berger Decora Emulsion

On receipt of the Respondent’s proposal, Vincent Lim expressed concerns over the necessity of applying a sealer coat. His concerns were quickly brushed aside by Joseph Yong, vide a letter dated 12 January 2006:^{[note: 3]}

Subject    :Proposed Paint System for Toh Guan Road East Capital One

With reference to the above subject and our discussion regarding the coating for the internal concrete surfaces including ceiling.

We are pleased to forward out letter for your submission to your client that it is not a [sic] necessary to apply a sealer coat before applying Berger Decora Emulsion even though it was stated in our product datasheet.

[emphasis added]

This was not all that Joseph Yong did to prevail upon the Appellant to accept the Respondent’s proposal. He again directly addressed the Appellant’s concerns in a subsequent letter (“13 January letter”):^{[note: 4]}

Subject: Warranty

With reference to the above subject, we will be providing you with five (5) years warranty on our products used in any of your up coming project as long as is based on our proposed paint system [sic].

[emphasis added]

7       The Toh Guan project was completed uneventfully. After this, the Respondent supplied paint to the Appellant for a third project at No 2 Toh Tuck Link (“the Toh Tuck Link project”). For this, it did not submit a proposal for a paint system.

8       The Respondent then supplied the Appellant with the same Decora Emulsion paint for a project at Bukit Batok Street 23 (“the Bukit Batok project”) - the subject matter of the present appeals. We set out now the circumstances that led to this supply of paint for the Bukit Batok project. The Appellant had been awarded the contract to paint the internal and external surfaces of the buildings by the main contractor, Eng Siang Lee Construction Co Pte Ltd (“ESL”), on 15 August 2007. As before, the Respondent’s proposal for the paint system made on its official notepaper and captioned “Proposed Paint System for Bukit Batok St 23” did not include a sealer coat.^{[note: 5]} Again, the Respondent’s Decora Emulsion paint was the only paint identified as appropriate for the painting the internal surfaces. There was absolutely no suggestion that it was not fit for the intended usage or that a sealer coat was necessary. However, unlike the Toh Guan project, the Appellant did not receive additional written confirmation from the Respondent that a sealer coat was unnecessary.

9       The Appellant’s account was that, having learnt about the Appellant’s intended bid for the Bukit Batok project, Joseph Yong approached Vincent Lim in July 2007 with a proposal to use the Respondent’s paint. The Appellant asserts that it had relied on the Respondent’s assurances on the suitability of the paint in submitting its winning bid for the project. Joseph Yong, it maintains, gave similar verbal assurances about the suitability of using the paint without the prior application of a sealer coat to Vincent Lim. Vincent Lim testified that this was the reason why he did not seek written confirmation from the Respondent that a sealer coat was unnecessary. He was also led to believe the warranty, as stated in the 13 January letter, would continue to apply to this project.

10     The Respondent disputed the Appellant’s claim that Joseph Yong actively sought the latter’s business support or that he advised the Appellant on the paint system. The Respondent claimed that it supplied paint to the Appellant on a “supply-only” basis^{[note: 6]} and that the Appellant purchased the paint not because of Joseph Yong’s advice but because it was cheap. Regrettably, we note that the Respondent did not take steps to ensure that Joseph Yong testified in Court. All the Respondent could baldly state was that its advice on the appropriate paint system for the Bukit Batok Project was not given. We therefore accept Vincent Lim’s evidence on these issues.

11     The requisite painting works were carried out between September 2007 and April 2008. During this period, the Respondent delivered the paint to the Appellant as and when the latter required it. According to the Respondent, the paint came from several different production batches. The Judge at [6] of the judgment noted that the paint came from 31 batches but the Respondent now asserts that the correct number of batches is 30. In the final analysis, it does not matter which is the true number. However, what is clear is that the paint supplied came from no less than 30 different batches. These constituted about 44,000 litres of paint. The Appellant was required to sign a delivery order and would be given a tax invoice every time the paint was delivered. The parties, we ought to mention, disagree over whether the terms of these documents formed part of the contract, in particular whether exemption clauses contained in them could be used to limit the Respondent’s liability and/or exclude terms implied by the Sale of Goods Act (Cap 393, 1999 Rev Ed) (“SGA”).

12     Sometime in April 2008, it was observed that there was serious pinkish discolouration of the paint on the internal surfaces of the Bukit Batok Project. The Judge found, in the judgment at [12] that the discolouration was widespread and occurred in most, if not all, of the surfaces. There is no dispute over this finding of fact. On 16 April 2008, ESL (see above, at [8]) complained to the Appellant that the paint on the internal surfaces had discoloured. The Appellant promptly relayed this complaint to the Respondent in a letter dated 18 April 2008.^{[note: 7]} In this letter, it alleged the Respondent’s paint was defective and that it was “very worry [sic] about [the Respondent’s] products” as another complaint had surfaced in relation to the paint it had used for the Toh Tuck Link project. It reiterated this allegation in a further letter dated 21 May 2008.^{[note: 8]} The Respondent responded on the same date denying its paint was defective. It emphatically asserted that Decora Emulsion was a low cost interior emulsion. It added that “[o]ne of the key criterion [sic] for the paint to be applied is that a sealer coat is used prior to the application of the paint” and while “[i]n certain cases, the paint may perform without the usage of sealer … that is not a recommended practice...”.^{[note: 9]} The Respondent also stated that it had commenced an internal investigation and offered to share the results of this with the Appellant to show that the paint was not defective. Significantly, the letter also stated that:

Our finding is in the presence of moisture if the surface is alkaline, the pink colour comes out clearly if a sealer coat is not applied to seal the substrate. This is hardly a problem with the paint but more an issue of substrate preparation”.

The Appellant replied on the following day stating that its own investigations ruled out the presence of moisture content. It bitingly added:^{[note: 10]}

Although in your data sheet it is mentioned that a sealer coat is required prior application of Berger Decora Emulsion [sic], you had proposed to us a paint system which does not include a sealer coat, i.e. 2 coats of Berger Decora Emulsion. If you were aware that direct application without sealer coat would result in a change of colour, your representative should have known better than to propose this very same system to us for this project. [emphasis added]

13     The parties tried to resolve their differences through a series of meetings, on-site inspections and correspondence. However, the dispute between the parties eventually became intractable. The Appellant then carried out the repainting works on the Bukit Batok project, without reference to the Respondent, using paint from a different manufacturer, Haruna (S) Pte Ltd and incurred additional costs of $52,242.75.

14     In the meantime, the Respondent continued investigating the cause of the paint’s discolouration. Rajeev Goel led an in-house team to inspect the Bukit Batok project site and conduct several tests to pin-point the cause(s) of the discolouration. It also approached the Construction Technology Division of Setsco Service Pte Ltd (“Setsco”), one of Singapore’s largest industrial testing and inspection companies, to assist in verifying its internal findings. Unknown to it, the Appellant had already commissioned Lee Sweet Choong, Elizabeth (“Elizabeth Lee”), who is presently the Divisional Director of Setsco’s Biological and Chemical Technology Division, to investigate the cause of the discolouration. At trial, she was its expert witness. As Setsco declined to represent the Respondent, it then engaged Wong Chung Wan, a director of MAEK Consultant Pte Ltd and part-time lecturer at the School of Design and Environment of the National University of Singapore, to be its expert witness.

Summary of pleadings and the decision below

15     On 23 November 2009, the Appellant filed its claim for reimbursement of costs of supplies and labour incurred in repainting the affected surfaces. It pleaded the following:

It had incurred costs of $1,258,222.22 in repainting the affected surfaces. After setting off the Respondent’s invoice for paint materials amounting to $72,676.62, the Appellant claimed $1,185,545.60 from the Respondent.

16     The Respondent’s pleadings in response were:

17     The Respondent also counter-claimed for $72,676.62 owed by the Appellant for the paint and other related items sold to it.

18     The Judge dismissed the Appellant’s claim and allowed the Respondent’s counterclaim. While he accepted that the representations Joseph Yong made in relation to the Toh Guan project – that a sealer coat of paint was unnecessary – also applied to the Bukit Batok project he nevertheless found that the Appellant had not proved that the defects in the paint caused the discolouration. His key findings were:

(a)     “[A]t the time of painting, the internal surfaces of the Project probably did not have excessive moisture levels” (the judgment at [15]).

(b)     “It was the point on the alkalinity of the skim coat that [the Judge] found much harder to rule out” (the judgment at [16]). The Judge’s reasons were that, during her tests, Elizabeth Lee used skim coats that were unrepresentative of the ones used at the project. He was also unsure whether the paint samples she tested on were representative of the paint used. He also expressed concern that storage conditions and time lag may have affected the composition or quality of the paint samples.

(c)     It was unlikely that there was a manufacturing problem for several reasons. First, the Appellant had used the paint previously without encountering problems, and no defects were noticed when the paint was mixed and applied. Second, he accepted the Respondent’s evidence that the paint complied with the manufacturing specifications. He also found it difficult to understand why the Appellant took eight months to complain about the discolouration. Lastly, the Judge felt that the discolouration occurring in patterns was “more consistent with the explanation that [the paint] had reacted with the skim coat” (the judgment at [19]).

Issues before this Court

19     The Appellant appeals against the Judge’s decision that the defects in the Respondent’s paint had not been established to be the cause of the discolouration. From the Appellant’s and Respondent’s predominantly factual submissions, it is clear the crux of both parties’ cases is that the Court should prefer the testimony of their respective expert witnesses. As mentioned above at [2], the Respondent is, in addition, specifically appealing against the Judge’s alternative finding of fact that it had warranted all of its products used in any of the Appellant’s projects for a five-year period.

20     The following issues arise in this appeal:

(a)     Has there been a breach of the contract? Answering this question requires answering the prior question as to whether Joseph Yong gave any assurances to Vincent Lim about the suitability of its paint without a sealer coat, which formed part of the parties’ contract.

(b)     What was the cause of the discolouration of the paint?

(c)     Is the Respondent’s liability limited by exemption clauses?

The first issue: Has there been a breach of the contract?

Did Joseph Yong give any assurances to Vincent Lim that a sealer coat was unnecessary for the Bukit Batok project?

21     As stated above at [9], the Appellant claimed Joseph Yong had approached Vincent Lim with a proposed paint system for the Bukit Batok project, an account which the Respondent hotly disputed. If Joseph Yong had indeed made statements to Vincent Lim, these could form part of the contract between the parties. It is thus necessary to determine whether such statements were ever made.

22     Contrary to the Respondent’s submissions, the paint proposal was not a mere formality which the Appellant needed to submit to ESL. There was plausible evidence that Joseph Yong knew that the Appellant was relying on the Respondent’s advice on the paint system. Vincent’s Lim’s unequivocal testimony that Joseph Yong had made representations about the suitability of the paint without a sealer coat and the provision of a warranty remained unshaken. We also note that in a letter dated 22 May 2008, the Appellant stated that Joseph Yong had made paint proposals to Vincent Lim for the Toh Guan dormitory and Toh Guan project. The Respondent’s reply letter dated 23 May 2008 did not rebut the Appellant’s assertions that it was Joseph Yong who actively sought the Appellant’s patronage. That being so, and in light of the prevailing evidence, we take these assertions as proved and infer that if Joseph Yong gave assurances to Vincent Lim for the earlier projects, he probably did the same for the Bukit Batok project as well. Further, Joseph Yong did not testify on behalf of the Respondent during the trial. Taking into account this unsatisfactory evidentiary position, we find that the Respondent was unable to effectively rebut the Appellant’s contentions on this issue. We therefore decide that it was likely that Joseph Yong had advised the Appellant on a paint system for the Bukit Batok project, that he had verbally assured Vincent Lim a sealer coat was unnecessary and that a warranty would be provided.

23     The Respondent submitted that it was unreasonable for the Appellant to rely on the former’s paint system without making cross checks of its own, in particular against the paint’s data sheet, which stated a sealer coat ought to be used. However, during the trial, the Respondent’s Jaideep Nandi had no alternative but to agree that the Appellant was also entitled to rely on the paint proposal made in relation to the Bukit Batok Project (see also above at [9]). The relevant portion of the transcript is as follows: ^{[note: 11]}

Q:… I put it to you that since [the Respondent] did not face any discolouration problems in the Toh Guan project, without using sealer, [it] was similarly entitled to rely on your identical paint proposal for Bukit Batok; do you agree?

A:Yes.

The parol evidence rule does not preclude incorporation of Joseph Yong’s representations to Vincent Lim.

24     It was undisputed that the tax invoices and delivery orders formed part of the contract. However, an issue that arises is whether Joseph Yong’s oral statements to Vincent Lim that a sealer coat was not required and that a warranty was provided were incorporated into the contract. This in turn raises two further issues. The first is whether consideration of the evidence of these statements falls foul of the parol evidence rule. In our opinion, it does not. In Zurich Insurance (Singapore) Pte Ltd v B-Gold Interior Design & Construction Pte Ltd [2008] 3 SLR(R) 1029 at [112], this court noted:

[T]he parol evidence rule only operates where the contract was intended by the parties to contain all the terms of their agreement. Where the contractual terms are ambiguous on their face, it is likely that the contract does not contain all the terms intended by the parties. Furthermore, in order to ascertain whether the parties intended to embody their entire agreement in the contract, the court may take cognisance of extrinsic evidence or the surrounding circumstances of the contract (see [40] above). [emphasis added]

25     It is trite that during the course of negotiations leading to the conclusion of a valid and binding contract, a number of statements may be made. Some may be intended to have contractual force; others may not. On the facts of the present case (outlined at [5]–[8] above), we find that the tax invoices and delivery orders did not comprise the entirety of the contractual relationship between the parties, and that the parol evidence rule does not preclude the incorporation of Joseph Yong’s statements to Vincent Lim into the contract.

Joseph Yong’s statements formed part of the contract

26     The second issue, arising from our ascertainment of the contractual terms, is whether Joseph Yong’s statements were contractual terms or non-binding representations merely intended to induce the other party to enter into the contract without imposing any responsibility for breach of contract. As Lord Moulton observed in Heilbut, Symons & Co v Buckleton [1913] AC 30 at 51, the true test of a statement having contractual force is whether there is:

... evidence of an intention on the part of either or both of the parties that there should be contractual liability in respect of the accuracy of the statement.

27     We agree that the proper inquiry should focus on an objective assessment of the parties’ intentions and take note of two particular factors. First, Joseph Yong’s statements to Vincent Lim were clearly important to the latter, who was purchasing the paint for the purpose of particular usage. Second, we are inclined to think that if the assurances as to the warranty and sealer coat had not been given, Vincent Lim would not have accepted the proposed paint system and entered into the contract with the Respondent. In Dick Bentley Productions Ltd and Another v Harold Smith (Motors) Ltd [1965] 1 WLR 623 at 627–628, Lord Denning MR with his customary acuity pointed out:

... if a representation is made in the course of dealings for a contract for the very purpose of inducing the other party to act upon it, and actually inducing him to act upon it, by entering into the contract, that is prima facie ground for inferring that it was intended as a warranty. It is not necessary to speak of it as being collateral. Suffice it that it was intended to be acted upon and was in fact acted on. But the maker of the representation can rebut this inference if he can show that it really was an innocent misrepresentation, in that he was in fact innocent of fault in making it, and that it would not be reasonable in the circumstances for him to be bound by it.

Since the Respondent undeniably had greater expertise and knowledge of its own products than the Appellant, the inference that Joseph Yong’s statements were intended to be a warranty could not be rebutted. Joseph Yong made his statements to Vincent Lim intending the latter to rely on them. We therefore find the following assurances given by Joseph Yong were express terms of the contract:

Has there been a breach of an express term of the contract?

28     We now consider whether there has been a breach of the contractual terms based on Joseph Yong’s statements about the Bukit Batok project. It was common ground that there were only three possible causes of the paint discolouration, that is to say: defects in the paint’s formulation, the condition of the internal surfaces being painted and/or poor workmanship in applying the paint. However, we immediately rule out the third possibility as a cause of the paint discolouration for two uncontroversial reasons. Poor application of paint often manifests in blisters, bubbles, craters, colour mismatch, gloss variations and/or peeling.^{[note: 12]} None of these were present on the project’s internal surfaces.^{[note: 13]} Furthermore, it was the Appellant’s undisputed evidence that bonding between the different paint coatings, as well as the bonding between coating and the surface, was good. This indicated that the paint had been properly and uniformly applied.

29     A skim coat was used to smooth certain parts of the internal surfaces. It was the Respondent’s case that it was the interaction between the paint and the alkaline skim coat which caused an alkali attack that resulted in the discolouration. This was also what the Judge appeared to regard as the most likely cause of the discolouration (see above at [18(c)]). This in turn required consideration of whether there was excessive moisture on the internal surfaces. As water is needed as a carrier for alkali movement,^{[note: 14]} its absence would rule out such likelihood. Rajeev Goel testified that when he placed his hand on the internal surfaces, he felt moistness. The Judge, quite rightly, rejected this test because it was unscientific and, importantly, not conducted contemporaneously. Instead, he preferred the testimony of Sundararajan s/o Nadarajan (“Sundararajan”), the onsite supervisor from ESL, the main contractor, testified that as a matter of invariable practice, 30 days had to lapse before he would give permission for paint to be applied on the surface.^{[note: 15]} This was to ensure that the surface was absolutely dry. Sundararajan also affirmed:^{[note: 16]}

After I test on the surfaces to be painted to check that there is no excess moisture, I then tell [the Appellant] to start painting works. If there is excess moisture on the concrete then I will not allow [the Appellant] to start painting.

At trial, he further elaborated on ESL’s procedure for testing for moisture:^{[note: 17]}

A:We have the moisture gauge. We use it on the wall, If the moisture gauge shows, the reading shows -- there are three levels, one is green, one is yellow and one is red. So if the green level is somewhere -- if the green level is somewhere 1 to 2, it’s dry.

Even if it is green sometimes, it is 3, I will not allow it.

30     From this, it is clear that ESL had proper procedures in place to ensure there was minimal moisture on the internal surfaces before any paint application took place. On this issue, the Judge correctly concluded that there was no excess moisture on the surfaces when the painting works commenced.

31     On the basis of this finding, the Judge could have then ruled out surface conditions as a possible cause of the paint discolouration. If there was inadequate moisture for there to be an alkali attack, then the surface conditions could not have caused the paint discolouration. However, surprisingly, he did not adopt this approach. One reason he gave for not doing so was that Elizabeth Lee could not explain why the discolouration was varied on most, but not all, of the surfaces. In his judgment at [19], he noted:

... the fact that the discolouration occurred in patterns rather than random patches suggested interaction with an underlying substance, rather than defects in [the paint] alone.

We have difficulty with the Judge’s thinking on this point. In the first place, there was no satisfactory evidence that clearly proved that the parts of the interior walls on which discolouration occurred were the same areas as those on which the skim coat was applied. Furthermore, uncontroverted evidence was adduced to establish that formwork was used to cast the plaster and concrete into shape. This presence of formwork appears to be a plausible reason for the rectangular patterns of discolouration on several of the affected surfaces. Indeed, this is what the Respondent’s own expert, Wong Chung Wan, opined as a possible cause of the patterned discolouration in his report:^{[note: 18]}

[T]he regular (rectangular) patches of discolouration divided by narrow “non-discoloured” strips on ceilings which infers that the characteristics of the substrates had contributed to the phenomenon. The well formed and consistent configurations of the discoloured patches are reminiscent of the formworks arrangement for concreting. Without first hand visual inspection, it is difficult to ascertain the cause of such pattern. [emphasis added]

32     Given this evidence adduced on behalf of the Respondent, we are puzzled how the Judge ruled otherwise, viz, that the pattern of discolouration indicated reaction with the substrate rather than an underlying defect with the paint: see [31] above. It seems to us that the good adhesion between the different paint coats indicated that there was no alkali movement between the various layers of paint. As explained in both Setsco reports by Elizabeth Lee, an alkali attack is typically characterised by chalking, blistering or efflorences, none of which were present.^{[note: 19]} In the light of the absence of any credible evidence of excess levels of moisture before the paint was applied, we are satisfied that an alkali attack had not occurred.

33     Another reason the Judge proffered for not accepting the Appellant’s case theory was that it had used the same paint before without any complaints. We disagree with the Judge’s reasoning on this point on two grounds. The Judge failed to appreciate that the Appellant had earlier complained that the paint which it had used for the Toh Tuck Link project was problematic: see [12] above. In addition, given that the Respondent did not attempt to show that the paint used in the different projects came from the same batches, the Judge could not and should not have assumed that the paint used in the Toh Guan project, against which there were no complaints, came from the very same batches of paint supplied for the Bukit Batok project. As the paint used were not from the same batches, this reason for not accepting the Appellant’s case theory is more than questionable.

34     We also do not accept the Respondent’s submission that the compliance of the paint with its manufacturing specifications indicated that the paint was free from defects. The Respondent’s own evidence is that the paint came from 30 different batches. It therefore had to show that the paint from all these batches were free from defects. Here, the Respondent was unable to even begin to discharge this burden as it did not show that all the paint samples it had tested were representative of the paint from the 30 batches used for the Bukit Batok project.

35     The final reason which the Judge gave for not accepting the Appellant’s evidence was that he did not understand why the Appellant took eight months to complain about the discolouration. The Appellant explained that once a floor in the project building had been painted, it would be handed over to ESL. It then immediately informed the Respondent of the discolouration once it had learnt about it from ESL. This is not disputed. We see no reason why this entirely logical explanation was not accepted. In the result, we find that the surface conditions were plainly not the cause of the paint discolouration.

36     Having eliminated two of the three possible causes, it seems to us that the paint discolouration was more likely than not caused by defects in the paint’s formulation. This view is reinforced by the settling effect or coagulation of the paint samples that Elizabeth Lee tested. We preferred her evidence over the Respondent’s experts for several reasons. With a sharp focus on determining the cause of the discolouration, her inquiry was more comprehensive than the reports adduced on behalf of the Respondent. The in-house report only had a narrow remit of determining the alkaline resistance of the coated cement panels. This was not decisive of the issue. Further, Wong Chung Wan’s reports were restricted to evaluating Elizabeth Lee’s report. He had not conducted any independent testing or analysis of his own and had not been to the building site. The Judge did not accept Elizabeth Lee’s conclusions because she had not “explain[ed] to [his] satisfaction how the disintegration of the paint necessarily evidenced defects in [the paint]” and because “the discolouration [occurring] in patterns rather than random patches suggested interaction with an underlying substances [sic], rather than defects in [the paint] alone” (at [18] and [19] of the judgment). However, as we noted above at [31], the patterns could have been related to the formwork used during the painting process. In addition, Elizabeth Lee’s findings have to be considered in the light of the parties’ acceptance that the discolouration could only have been caused by one or a combination of three things: defects in the paint, poor workmanship and/or surface conditions. While Elizabeth Lee did not irrefutably prove scientifically that there were defects in the paint, she convincingly established that the discolouration was not caused by poor workmanship or the surface conditions and therefore logically proved by the process of elimination that defects in the paint or its unfitness were the root causes of the discolouration. The Respondent tepidly suggested that she had not conducted her tests on representative samples. This conveniently glosses over the fact she had tested two types of samples. The first type was paint provided by the Appellant. The second type was dry samples which her team had cut from the internal surfaces of the project. The Respondent’s criticism could apply to the former, but not the latter.

37     The legal burden is of course on the Appellant to prove that defects in the paint or its unfitness led to its discolouration. Its evidence eliminated surface conditions and poor workmanship as causes of the discolouration, leaving defects in the paint or its unfitness as the probable cause. On this issue we note that Jaideep Nandi acknowledged that when other possible causes for discolouration were eliminated then paint failure could be attributed to the paint itself:^{[note: 20]}

When I refer a paint system one of my basic criteria that I want to ensure is that there is low moisture on the wall, low alkalinity on the wall.

If after that, the paint fails, I am clear that the paint is failing surely because of the paint. I mean there is no doubt in my mind. And if you ask me whether the paint does fail ever, yes obviously we have ...

Once the evidence logically demonstrated that defective paint was prima facie the likely cause of the discolouration, the tactical burden shifted to the Respondent to show that this was not correct on a balance of probabilities. We should explain that the tactical burden is one that is borne by the opponent of the issue after the proponent has discharged his evidential burden: see Colin Tapper, Cross & Tapper on Evidence (Oxford University Press, 12th Ed, 2010) at p 126. The opponent must, in the words of Denning J (as he then was), “call evidence or take the consequences, which may not necessarily be adverse”: see Presumptions and Burdens (1945) 61 LQR 379 at 380. What is needed in any particular case to assess whether this burden has been discharged is a fact-centric exercise circumscribed by commonsense. The Respondent could have, inter alia, adduced evidence that it had not received any complaints from other customers about the paint sold from the same 30 odd batches. Oddly, it failed to do so. Indeed, there is no evidence on record of the paint from these batches being sold to other parties and or that it been used. What happened to all this paint? Instead, the Respondent adduced evidence that attempted to merely show that the manufacture of the paint used for the Bukit Batok project would have, in the ordinary course, complied with the paint’s technical specifications. During cross-examination, Elizabeth Lee rightly accepted that if the paint formulation conformed to standard procedure, there was unlikely to be a problem. This evidence was relevant but not decisive of this issue, especially since the Respondent did not adduce evidence to show the paint it conducted tests on were from the very same batches of paint supplied to the Appellant. The Respondent ought to have also shown that the paint was free from defects from stowage at the point when it was handed over to the Appellant. Regrettably, it failed to do so. Tellingly, we also note that the Respondent did not produce any documentation to substantiate the findings of its internal investigations that the substrate preparation was the root cause of the discolouration despite its insistence that it maintained “detailed logs”.^{[note: 21]}

38     Accordingly, we find that the paint discolouration was more likely than not caused by defects in the Decora Emulsion paint supplied by the Respondent. There has therefore been a breach of the express term of the contract that the Respondent’s paint would be free from defects and fit for application on the internal surfaces of the Bukit Batok project, without the need for a sealer coat (see [27(a)] above). This finding obviates the need to consider whether there is a breach of the implied terms of the SGA.

The second issue: What was the cause of the discolouration of the paint?

Test of causation

39     The but for test of causation in tort also applies in ascertaining whether there has been factual causation for the purposes of contract law. In Sunny Metal & Engineering Pte Ltd v Ng Khim Ming Eric [2007] 3 SLR(R) 782, this court stated (at [63]):

In our view, there is no reason why the “but for” test in tort cannot also be used in contract cases to determine the issue of causation in fact... Accordingly, we adopted the tortious test for causation in fact in considering the issue of causation in SME’s claims in both tort and contract. [emphasis in original]

40     We note that in his judgment at [19], the Judge held that “[the Appellant] has not proven on balance of probability [sic] that the defects in [the paint] had caused the discolouration”. Seen against the entire tenor of his judgment and how he was singularly focussed in pin-pointing a single cause for the discolouration of the paint, it appears the test of factual causation the Judge applied was whether the defective paints were proven on a balance of probabilities to be the only possible cause, rather than the but for cause. This may seem to be a semantic splitting of hairs, but the evidentiary burden on a party and the substantive threshold of a cause of action are two different matters. The balance of probabilities lies in the province of the law of evidence while the test for factual causation belongs in the substantive law of contract.

Applying the test of factual causation

41     As the Judge seemed to have applied the wrong test of factual causation, we have to consider afresh whether the Appellant successfully showed on a balance of probabilities that defects in the paint caused its discolouration.

42     The causal inquiry overlaps with the inquiry as to whether there were defects in the paint. Having eliminated poor workmanship and surface conditions as possible causes of the paint discolouration, we find that defects in the paint were the but for cause.

The third issue: Is the Respondent’s liability contractually limited?

There is an express warranty

43     Whether there is a warranty in respect of the Bukit Batok project is the subject of the Respondent’s cross-appeal. The parties have adopted different interpretations of the 13 January 2006 letter to the Appellant (see [6] above). The Appellant contends that it is an open warranty that applies to all future projects. The Respondent has argued that it only applied to the Toh Guan project. The Judge observed obiter that “[the Respondent] had warranted all its products used in any of [the Appellant]’s projects for a period of 5 years provided that the use was based on a proposed paint system by [the Respondent]” (the judgment at [21]).

44     We agree with the Judge’s finding that the 13 January 2006 letter was a warranty. The phrase “in any of your up coming project [sic]” clearly means what it plainly states - the warranty was not limited to the Toh Guan project. The Respondent submitted that a warranty cannot be so open-ended but did not cite any authorities for this proposition. Even accepting the Respondent’s argument on its face, the warranty was not carte blanche since the warranty would only apply if the Appellant not only purchased the Respondent’s paints, but also did so in compliance with the latter’s proposed paint system.

The express agreement overrides the exemption clauses

45     Our finding that there is a warranty raises the question whether the Respondent is still entitled to rely upon the exemption clauses in the tax invoices and delivery orders to limit its liability, if any, to the Appellant.

46     In our view, the exemption clauses were not effective in the light of the express warranty. It is well established that an exemption clause contained in a written contract can be overridden by an express warranty given at or before the time the contract was concluded: Couchman v Hill [1947] KB 554. For a concise summary of the law in this area, one may turn to Edwin Peel, Treitel: The Law of Contract (Sweet & Maxwell, 13th Ed, 2011) at para 7-041 which reads as follows:

Overriding undertaking. An exemption clause in a document with reference to which the parties contract can be overridden by an express inconsistent undertaking given at or before the time of contracting. Thus a buyer of goods by auction can recover damages for breach of an oral undertaking given at the time of sale although the printed conditions of sale exempt the seller from all liability for defects. To bring this rule into operation, there must be an “express specific oral promise” which is inconsistent with the exemption clause …

We note that Chitty on Contracts (Sweet & Maxwell, 30th Ed, 2008) vol 1 at para 14-134 also emphatically states:

Collateral warranties and guarantees. A party who would otherwise be entitled to rely on an exempting provision will not be permitted to do so if he gives an express oral warranty which runs counter to the tenor of the written exemption. A warranty given before the agreement is entered into may also be enforced as a collateral contract the consideration of which is the entering into of the written agreement. Thus in Webster v Higgin an oral warranty as to the present condition of a car was enforced as a collateral contract in return for which a contract of hire-purchase, which contained exempting provisions, was signed. [emphasis added]

47     In addition to finding that the 13 January 2006 letter was indeed a warranty intended to have contractual force in relation to future projects such as the Bukit Batok project, we also consider it to have been expressly incorporated into the parties’ contractual relationship since it was the Appellant’s undisputed evidence that it relied on Joseph Yong’s statements and the 13 January letter before entering into the contract with the Respondent. This warranty takes precedence over any exemption clauses in relation to subject matter it embraces. As such, the Respondent cannot rely on the latter to limit its contractual liability.

48     To summarise, we depart from the Judge’s ruling on the issue of causation. He had made an understandable but, nevertheless, clearly demonstrable error. The parties had agreed that there were three possible causes of the discolouration: defects in the paint’s formulation, the condition of the internal surfaces being painted and/or poor workmanship in applying the paint. For the reasons stated at [28] and [29] to [35], we eliminated poor workmanship and the internal surfaces respectively as the likely causes of the discolouration. With the Appellant prima facie proving that defects in the paint had caused the discolouration, the tactical burden shifted to the Respondent to show that the paint it supplied was not defective. It failed to do so. Consequently, we find that the Appellant has proven on a balance of probabilities that defects in the paint were the cause of the discolouration.

Conclusion

49     We therefore allow the appeal and dismiss the cross-appeal, with taxed costs here and below to the Appellant. The assessment of the damages due from the Respondent to the Appellant is remitted to the Judge. The usual consequential orders are to follow.

---

^{[note: 1]}Record of Appeal Vol III(D) at p1190

^{[note: 2]}Appellant’s Core Bundle (Vol II) at p 16.

^{[note: 3]}Appellant’s Core Bundle (Vol II) at p 17

^{[note: 4]}Appellant’s Core Bundle (Vol II) at p 19

^{[note: 5]}Appellant’s Core Bundle (Vol II) at p 112.

^{[note: 6]}Respondent’s Case at p 29, para 69.

^{[note: 7]}Appellant’s Core Bundle (Vol II) at p 21.

^{[note: 8]}Appellant’s Core Bundle (Vol II) at p 38.

^{[note: 9]}Appellent’s Core Bundle (Vol II) at p 39.

^{[note: 10]}Appellant’s Core Bundle (Vol II) at p 41.

^{[note: 11]}Official Transcript: Day 5, 2 July 2010, p 79; Record of Appeal Vol III(L) at p 3491

^{[note: 12]}See second report of Setsco, at Appellant’s Core Bundle at p111

^{[note: 13]}See first report of Setsco at Appellant’s Core Bundle at p 67

^{[note: 14]}This was admitted by the Respondent’s own witnesses: a) Wong Chung Wan’s report dated 18 May 2010 at RA Vol III(F) 1681 b) Rajeev Goel’s evidence at trial at RA Vol III(L) 3600 and Sharon Lee at RA Vol III(M) 3741.

^{[note: 15]}Offical Transcript, Day 1, 28 June 2010; Record of Appeal Vol III(J) at pp 2796 and 2797

^{[note: 16]}AEIC of Sundararajan s/o Nadarajan dated 11 June 2010 at para 5.

^{[note: 17]}Offical Transcript, Day 1, 28 June 2010; Record of Appeal Vol III(J) at p 2796

^{[note: 18]}AEIC of Wong Chung Wan, filed on 18 May 2010, at p 31; Record of Appeal Vol III(F) p 1682

^{[note: 19]}Appellant’s Core Bundle Vol II at p 101; Record of Appeal Vol V (B) at p 4717 and Appellant’s Core Bundle Vol II at p 67

^{[note: 20]}Official Transcript, Day 5, 2 July 2010 at lines 15 to 22; Record of Appeal Vol III(L) at pp 3458.

^{[note: 21]}Appellant’s Skeletal Submissions at para 12.