Dermot Woolgar successfully establishes reasonableness of sums paid in settlement of product liability claims for defective construction adhesive
A. Summary
Following a 5-week trial, Dermot Woolgar has successfully obtained judgment in DIPT Ltd & Ors v Sanglier Ltd; Sanglier Ltd v Apollo Chemicals Ltd [2023] EWHC 426 (TCC) for his clients’ insurers in a series of test cases concerning the supply of a defective sprayable pressure-sensitive adhesive.
The adhesive had been used widely throughout the United Kingdom and continental Europe during 2013 and 2014 in joinery fabrication, shopfitting and general construction works. It failed prematurely, becoming powdery and embrittled, causing extensive delamination.
The contractors who had purchased the defective adhesive claimed against the suppliers for the cost of the remedial works which they had carried out, and for related losses. On recommendations made by loss adjusters appointed by the suppliers’ insurers, the contractors’ claims were all settled. The suppliers then sued to recover the sums that had been paid in settlement of the contractors’ claims.
The defendants denied that the adhesive was defective and made many criticisms at trial of the loss adjusters, vigorously contesting the basis upon which the settlements had been reached and their reasonableness. The adhesive was found to be defective and none of the criticisms succeeded. Judgment was given for the full amount of the settlements.
The judgment provides a helpful summary of the principles to be applied when a claimant seeks to recover sums which it has paid in settlement of claims made by third parties. The case also usefully illustrates the application of those principles in practice.
B. Background to the proceedings
The claimants are a group of associated companies which supply construction tools and materials to the joinery, shopfitting and construction trades in the UK. In 2012 they asked a UK distributor of adhesives to develop a new pressure-sensitive adhesive with a high solids content. The distributor approached Sanglier Ltd, a leading supplier of construction adhesives. In turn Sanglier approached Apollo Chemicals Ltd to develop and manufacture the adhesive. Apollo duly developed the adhesive and sold it to Sanglier, which put it into pressurised canisters and, under the terms of an exclusivity agreement with the distributor and the claimants, then sold it on to the distributor for supply throughout the UK and Ireland. Sanglier also sold the adhesive under a similar exclusivity agreement to a Dutch distributor for supply throughout continental Europe.
The UK distributor sold the adhesive to the claimants, who in turn sold it to a variety of joinery, shopfitting and other construction contractors.
Pressure-sensitive adhesives (“PSAs”) are used globally. Typically a bond made with a PSA can be expected to last 10 years. Many of the bonds made with this adhesive failed prematurely. After only a matter of months the adhesive became friable, embrittled and powdery. The claimants received an increasing number of complaints from contractors about the unsatisfactory performance of the adhesive. Convinced that the adhesive had some defect in its chemistry, the claimants decided to withdraw it from the market.
The contractors then carried out remedial works using alternative PSAs and sought to recover their costs and other losses from the claimants, on the basis that the adhesive had not been of satisfactory quality nor fit for purpose, in breach of the terms implied by the Sale of Goods Act 1979.
The claimants notified their insurers, who appointed loss adjusters and solicitors to investigate the contractors’ claims.
The claimants made initial investigations into the chemistry of the adhesive using the services of a specialist laboratory, while their insurers appointed Burgoynes to carry out further analyses. Sanglier undertook its own investigations into the bond failures using their own employees, aided by some spectroscopic examinations which they commissioned externally.
The upshot of these investigations was that the claimants and their insurers were persuaded that the failures were attributable to a defect in the chemistry of the adhesive, whereas Sanglier considered that there was no defect and that the failures were all the result of “misapplication” of the adhesive – in other words, poor workmanship by the contractors.
Sanglier suggested that, had there been a defect in the chemistry of the adhesive, it would have failed whenever and wherever it had been used, which Sanglier claimed had not occurred. But the claimants and their insurers considered that the multiplicity of the failures militated against poor workmanship. They pointed out that its method of application was no different from other PSAs, and that the contractors were all experienced in the application of PSAs. The expert evidence that had been received from Burgoynes also suggested that poor workmanship could not explain why the bonds which had failed had worked initially, nor could poor workmanship at the point of application account for the subsequent significant change in the physical properties of the adhesive.
In these circumstances the claimants’ insurers instructed the loss adjusters to proceed on the footing that the adhesive had been defective, and to focus their efforts instead on analysing the quantum of the contractors’ claims.
The claims were duly investigated and settled. Although invited to participate in this process, Sanglier declined. In total, the claimants’ insurers paid in excess of £2.1m to settle the contractors’ claims. The claimants alleged that they had also suffered loss of business and loss of profits. The claims, including the legal costs of investigating and settling the contractors’ claims, amounted in total to £2.8m.
C. The proceedings
The claimants took an assignment of the distributor’s cause of action and commenced proceedings against Sanglier in 2018.
Shortly afterwards, in early 2019, Sanglier commenced proceedings against Apollo. In its action Sanglier sought not only to pass on to Apollo any liability which it might be found to have to the claimants in the first action, but also to recover the sums which it had paid to settle various claims which it had faced from its Dutch distributor, following failures of the adhesive in continental Europe.
The two actions were then conjoined. This had the effect of making Apollo a defendant in all but name to the claimants’ action.
Vis-à-vis the claimants, both Sanglier and Apollo continued to deny that the adhesive was defective and they continued to allege poor workmanship by the contractors. They were also critical of the settlements which had been reached with the contractors. But as between themselves they accepted that the adhesive had in fact been defective, for which they blamed each other. Sanglier alleged that Apollo had manufactured the adhesive with insufficient antioxidant and an unusual tackifier. Apollo alleged that some of the canisters into which Sanglier had put the adhesive had been contaminated with manufacturing residues. Sanglier disputed this, pointing out that the canisters had been mass produced by a reputable Portuguese manufacturer and that the canisters had been routinely used by itself and other PSA manufacturers for years.
Apollo also relied on a limitation of liability clause in its terms and conditions, by which it confined its liability to Sanglier for breach of contract to the price of the goods. This would have capped its liability to Sanglier at less than £50,000. Sanglier challenged the reasonableness of this clause under sections 6(1A) and 11 of UCTA.
Directions were given for the trial of a selection of the UK contractors’ claims, and a selection of claims from Germany and Austria which had been raised with the Dutch distributor.
Permission was also given for expert evidence in the fields of materials failure and organic chemistry.
D. The trial
The parties called 23 witnesses at the trial. Among them were representatives from the contractors who had used the adhesive in the UK, in Austria, and in Germany, three loss adjusters from Sedgwick who had been responsible for considering the claims that had been submitted by the UK contractors, and a representative from the Portuguese manufacturer of the canisters.
Additionally the parties called 4 experts in the fields of materials failure and organic chemistry. Judges in the TCC customarily have to evaluate complicated expert evidence, but even by TCC standards the expert evidence in this case was unusual and intricate. The judge, HHJ Mark Pelling KC, rightly noted, in a carefully understated observation during the course of the trial, that the expert evidence was “not of this court’s usual currency”.
The defendants mounted a vigorous attack at trial on the settlements which had been negotiated by the loss adjusters. They complained that the loss adjusters could not have been satisfied that the contractors had actually used the adhesive, nor that the delaminations about which complaints had been made were associated with the use of the adhesive as opposed to other adhesives. They complained about the failure of the claimants to keep records of the batch numbers of the adhesives which they had supplied to the contractors, and they were critical of the contractors for not keeping records of the batch numbers of the adhesives which they had used in the course of their businesses. They criticised the insurers and their solicitors for instructing the loss adjusters to concede liability and causation and to focus instead on quantum. They criticised the loss adjusters for failing to take into account the probability that there had been workmanship errors. They accused the loss adjusters of failing to subject the quantum of the contractors’ claims to sufficient scrutiny, of failing to exploit a limitation of liability clause in the claimants’ terms and conditions, of using rates which were excessively generous or inconsistent, and of failing to exclude losses which they said were too remote. They also alleged that the loss adjusters had generally failed to give adequate explanations in their written and oral evidence for the settlement figures, with the result, they said, that they did not know how the loss adjusters had actually arrived at the settlement figures.
A notable aspect of the trial was the oral evidence given by Sanglier’s expert in materials failure and organic chemistry, Professor Ivan Parkin. According to his written reports the failure of the adhesive was attributable to the use by Apollo of insufficient antioxidant in the adhesive and an excessive proportion of what he contended was an unusual tackifier. If there had been any workmanship errors, they would have only accelerated the failure of the adhesive, which in his view because of its chemistry was “doomed to fail” anyway. But the difficulty with these views was that the adhesive had not always failed – judging, that is, from the level of complaints that had been received. Apollo contended that this feature supported its case that the failures were attributable to the fact that some of the canisters had been contaminated, and that the time to failure had varied in accordance with the degree of contamination. Apollo’s counsel suggested that only 5-10% of the adhesive which had been sold by Sanglier had resulted in failure. Having heard Apollo’s counsel make this point repeatedly during the course of the trial, Professor Parkin changed his mind, and when towards the end of the trial he came to give his oral evidence, he then blamed the failures predominantly on the method of application that had been used by the contractors. He suggested that bonds made with the adhesive would in fact have been capable of lasting 10 years.
E. Discontinuance post-trial
After the end of the trial, but before judgment, Sanglier filed a notice of discontinuance in its action against Apollo, and agreed to pay a substantial proportion of Apollo’s costs. On any view this was an unusual development. Why Sanglier chose to discontinue at this point in the proceedings remains a matter of speculation, but the significant change in its expert’s views at trial cannot have helped its prospects. It also faced difficulties in disputing the reasonableness of Apollo’s limitation of liability clause. It had an identical limitation in its own terms and conditions (which it was unable to rely upon in defending the claimants’ claims, because it had failed to incorporate those terms and conditions into the sales which it had made with the UK distributor), and it could not allege that it had been unable to obtain insurance except at prohibitive cost – for, as it frankly acknowledged, it had been insured. Its insurers had however declined cover, on grounds which it contested.
The task of resolving the conflicting, and abstruse, expert evidence concerning the defective chemistry of the adhesive had seemed unenviably difficult, on the face of the experts’ reports. It had probably been made a good deal easier by the unattractively late, and drastic, change in Sanglier’s expert’s views at trial. But the discontinuance made it unnecessary to resolve the expert evidence. One suspects that the trial judge must have been, privately, at least a little relieved.
F. The judgment: core findings
The judge, HHJ Mark Pelling KC sitting as a Judge of the High Court, found in summary that:
(1) the adhesive had failed as a result of some defect in its chemistry – whether that was because of an error in its formulation or contamination in the canisters he expressly declined to say; the adhesive was accordingly neither of satisfactory quality nor fit for purpose;
(2) there had been no meaningful failure by any of the test case contractors to apply the adhesive in a proper and workmanlike manner; if there had been any such failures, they were statistically insignificant; and
(3) the settlements were reasonable, both in principle and in amount.
G. The judgment: reasons and observations concerning liability
Some of the reasons given by the judge for his conclusions on liability are of wider interest – to product liability lawyers, construction lawyers, insurance lawyers, and other commercial litigators.
(1) The judge was satisfied that the delaminations were associated with the use by the contractors of the adhesive, and not some other PSA. The pattern of invoices established how much adhesive had been bought, how much had been used, and over what period of time. There was no evidence that the use of any other PSA had been associated with widespread delaminations. The defendants’ complaint about the failure to record batch numbers was “entirely unreal. Had this been a claim concerning components for use in the repair and maintenance of public transport certified aircraft, for example, this approach would have been justified. However, the product was an industrial adhesive used in low tech, low cost applications…This is an entirely unregulated activity…There is no legal or other obligation to obtain or retain such records…”.
(2) It was inherently improbable that the delaminations had been caused by poor workmanship. The failures were too widespread; the symptoms upon failure were too similar; the instructions for the use of the adhesive did not require any techniques which were materially different from those used when applying other PSAs; the contractors were not new entrants into the market nor were any of their employees novices in the use of PSAs; and there was no evidence that the use of other PSAs was associated with a similar rate of failure.
(3) On this last point the judge was strikingly critical of the defendants: “If the magnitude of failure experienced was similar in respect of all PSAs, that would point circumstantially towards the defects being the result of workmanship rather than product defects. It is a signal feature of this case that there is no such evidence. This is a startling omission given the time and expense that has been expended in defending these claims…I infer that such evidence has not been adduced simply because it does not exist.”
(4) If there was something in the nature of the adhesive which made it particularly susceptible to failure unless used in a particular way – on one view of the evidence, this was a possibility – that should have been made clear. Neither defendant had done so.
(5) Apollo’s assertion that only 5-10% of the adhesive which had been sold had been implicated in failure was unsound and “probably undershoots the likely failure rate significantly”. The claimants had correctly submitted that “the only thing one can say with certainty is that we simply do not know what the rate of failure was as a proportion of the total number of canisters that were sold”.
(6) The allegations which the defendants were making against each other concerning defects in the chemistry of the adhesive were not consistent with the notion that the failures were caused by poor workmanship.
(7) Professor Parkin’s contention in his written reports concerning defective formulation of the adhesive “was absolute in its terms and offered no explanation for why there had not been a statistically universal failure of bonds…”. His change of mind in the witness box was “a last-minute attempt to explain away what was a real difficulty…that should have been apparent from a very early stage…”. His change of mind had been “profoundly unconvincing”, and his new theory had been put forward “as little more than a debating point to meet an inconvenient truth”.
H. Relevant legal principles to be applied in actions to recover settlements paid to third parties
The judgment contains a useful summary of the principles which should be applied when a claimant seeks to recover sums which it has paid in settlement of claims made by third parties, drawing upon Siemens Building Technologies FE Ltd v Supershield Ltd [2009] EWHC 927 (TCC) per Ramsey J at [80] and [2010] EWCA Civ 7 per Toulson LJ at [29]; 125 OBS (Nominees1) & Or v Lend Lease Construction (Europe) Ltd & Or [2017] EWHC 25 (TCC) per Stuart-Smith J at [186-7], and BP plc v AON Ltd [2006] EWHC 424 (Comm), per Colman J:
(1) It is not necessary for the claimant to prove on the balance of probabilities that it was or would have been liable to the third parties, nor that it was or would have been liable to them in the amount of the settlements.
(2) Rather, the claimant must show that the defendant’s breach of contract caused the losses which the claimant incurred in satisfying the settlements which it made with the third parties, and that those losses are not too remote.
(3) Unless the third parties’ claims were of sufficient strength reasonably to justify their settlement and unless the amounts paid in settlement were reasonable having regard to the strength of the claims, the claimant will be unable to show that its losses were caused by the defendant’s breach of contract but, unless the claim is so weak that no reasonable party would take it sufficiently seriously to negotiate any settlement involving payment, it cannot be said that the loss attributable to a reasonable settlement was not caused by the eventuality or the breach.
(4) In general if, when a party is in breach of contract, a claim by a third party is in the reasonable contemplation of the parties as a probable result of the breach, then it will generally also be in the reasonable contemplation of the parties that there might be a reasonable settlement of any such claim by the third party.
(5) The test of whether the amounts paid in settlement were reasonable is whether the settlements were, in all the circumstances, within the range of settlements which reasonable people in the position of the claimant might have made. Those circumstances will generally include (a) the strength of the claims; (b) whether the settlements were as a result of legal advice; (c) the uncertainties and expenses of litigation; and (d) the benefits of settling the claims rather than disputing them.
(6) Whether the amount paid in settlement was reasonable is to be assessed at the date of the settlements, when necessarily the issues between the claimant and the defendant were still unresolved.
(7) Reasonable settlements are encouraged by the courts particularly where strict proof would be very expensive.
(8) The test of reasonableness is generous, reflecting the fact that the defendant has put the claimant in a difficult position by its breach.
(9) A claim will generally have to be so weak as to be obviously hopeless before it can be said that settling it is unreasonable.
(10) Where the settlement is shown to be prima facie reasonable, the evidential burden of proving the unreasonableness of a settlement falls upon the defendant.
I. Application of these principles on the facts
Applying these principles, the judge rejected all the criticisms which had been directed at the loss adjusters:
(1) It was within the reasonable contemplation of Sanglier that, in the event that it committed a breach of the implied terms as to satisfactory quality and fitness for purpose, contractors might make claims against the claimants which might then reasonably be settled. “That is particularly so in a product liability claim of this sort where…the margin made by the onward sale of the product would be modest…and any resulting claims were likely to be made by many different individual [contractors] for sums that were likely to be modest and certainly below the level at which it would be economic to dispute them”.
(2) If, contrary to his conclusions, a proportion of the failures had been caused by poor workmanship, the proportion would have been small “and difficult to detect without expensive and time-consuming investigation.” It had been, accordingly, reasonable when settlements were being negotiated to leave this matter out of account. “The multiplicity of small claims that the claimants were faced with was classically the sort of ‘difficult situation’ in which a claimant is placed by a defendant’s breach that justifies a generous approach to settlements…the expense of litigation at the suit of multiple different [contractors] based on multiple different customer claims each of which was individually of modest value was almost bound to be disproportionate and thus the benefits of settling the claims rather than disputing them were both obvious and compelling”.
(3) The instructions to the loss adjusters had been given not by the claimants but by the insurers and their solicitors, who had an interest in achieving settlements that were as low as possible (an observation which echoes Coulson LJ’s comments in Brit Inns Ltd & Ors v BDW Trading Ltd [2012] EWHC 2143 (TCC) at para 56).
(4) The decision to instruct the loss adjusters to concentrate on quantum, and the ensuing settlements, were based on legal advice which had considered the possibility of poor workmanship but which had discounted that possibility on the basis of apparently competent technical expert advice. The claimants had correctly submitted that the claimants, their insurers and the insurers’ solicitors “were taking a reasonable view on what were reasonable materials, following a reasonable examination of the product”.
(5) “The correct test involves asking whether, in all the circumstances, the settlements were within the range of settlements which reasonable people in the position of the claimants might have made, tested at the date when the settlements were entered into. I am unable to accept that a decision to negotiate settlements quickly by reference to quantum issues alone in the circumstances as they were [at the relevant date] took the settlements outside that range, in particular having regard to the costs involved in requiring [the contractors] to prove liability. Indeed, the point is not one that could credibly be taken given the expert technical evidence then available.”
(6) The loss adjusters had achieved significant reductions in the claims that had been made by the contractors. Indeed one contractor claimed to have been driven into liquidation by its settlement, “though I suspect that the level of settlement was not the sole cause of its demise”.
(7) Lastly, Sanglier had submitted that the court could not assess the reasonableness of the settlement figures “without market comparators”, and that it should ignore the claimants’ invitation to rely on the loss adjusters’ opinions as to what was reasonable when they were giving evidence only as witnesses of fact, without the benefit of such comparators. That submission got short shrift: “If Sanglier wanted to assert that the quantum [of the] settlements fell outside the range…which reasonable people in the position of the claimants might have made, then it was for Sanglier to adduce evidence to that effect from an adjuster or other person qualified to give such evidence…No such evidence was adduced.”
J. Conclusions
Full-scale attacks on the reasonableness of settlements, of the kind that the defendants mounted in this case, are rare. Even more rarely do they succeed. Given that the evidential burden will often fall upon the defendant to prove that a settlement was unreasonable, the prospects of discharging that burden are in practice especially negligible where settlements have been negotiated by loss adjusters. Further, it should be borne in mind that the reasonableness of a settlement will never arise for consideration unless the claimant first establishes the defendant’s breach. In cases where a product has been sold and resold repeatedly creating a potential chain of liabilities, once the defendant’s breach has been established an instruction of the kind that was given to the loss adjusters in this case in effect to treat the claimants’ breach as having been established will retrospectively be shown to have been correct. In practice, therefore, a defendant’s protestations in these cases that breach should have been contested in order to achieve a more favourable settlement will almost invariably sound rather hollow.
Dermot Woolgar
31 March 2023
The DIPT claimants were represented by Dermot Woolgar, instructed by Olu Dansu and Jenni Pateman of DAC Beachcroft LLP
Apollo was represented by Carlo Taczalski and Michael Harper, instructed by Greg Dawson and Nick Hibbert of Keoghs LLP. Their commentaries on this case can be found here and here.
