BTI 2014 LLC v Pricewaterhousecoopers LLP & Anor

This claim for damages for professional negligence was issued by the claimant (“BTI”) on 28 October 2014. The second defendant is only a nominal defendant. The claim against the first defendant (“PwC”) is concerned with its audit of the 2007 and 2008 annual accounts of the second defendant (then known as Arjo Wiggins Appleton Ltd, or “AWA”) in October 2008 and May 2009 respectively.

In circumstances that I will explain, the claim was stayed for several years pending the outcome of related proceedings by BTI against AWA’s former parent company, Sequana S.A. (“Sequana”) and the directors of AWA (“the Directors”). This claim was revived following the decision of the Court of Appeal in those proceedings, by which time BTI had served Amended Particulars of Claim. On 26 March 2019, PwC applied to strike out the claim and alternatively for summary judgment to be entered in its favour. I heard argument on that application over three days from Mr Salzedo QC on behalf of PwC and Ms Day QC and Mr Thomson QC on behalf of BTI.

BTI, which is a wholly-owned subsidiary of BAT Industries plc (“BAT”), sued Sequana and the Directors as assignee of AWA, claiming recovery of very large dividends paid by AWA to Sequana in December 2008 (€443 million, “the December dividend”) and May 2009 (about €135 million, “the May dividend”). The dividends were paid against the background of PwC’s audit of AWA’s annual accounts in October 2008 and May 2009. That claim failed. Rose J held that the accounts relied on by the Directors for payment of the dividends were proper accounts for the purposes of Part 23 of the Companies Act 2006 (“the 2006 Act”) and that accordingly the dividends could not be recovered from the defendants. BTI did not appeal that decision.

BTI now alleges that PwC negligently audited AWA’s 2007 and 2008 annual accounts thereby causing AWA loss, in that the Directors would not have resolved to pay the very large dividends had PwC acted non-negligently. They would not have done so, BTI alleges, because the accounts of AWA would then have shown that it had considerably greater liabilities and fewer distributable reserves. BTI claims against PwC loss in the full amount of the December and May dividends paid to Sequana.

The grounds on which PwC applies to strike out the claim, alternatively for summary judgment, are the following:

The relevant facts giving rise to the claims against Sequana and PwC can be summarised briefly as follows.

Another wholly-owned subsidiary of BAT, Appleton Papers Inc (“API”), purchased two paper coating businesses from National Cash Register Company (“NCR”) in 1978. API operated in the Lower Fox River area of Wisconsin. Under the terms of the sale and purchase agreement, API took over NCR’s liabilities, including any environmental liabilities, and BAT agreed to indemnify NCR against API’s failure to discharge those liabilities. At a later time, API’s immediate parent company was separated from the BAT group and changed its name to AWA, but API’s and BAT’s liabilities remained. The paper businesses purchased by API had previously been responsible for polluting the Lower Fox River. In the 1990s, environmental liability claims were notified against NCR and API in this regard, comprising clean-up costs

An agreement was made between BAT, NCR and API in 1998 to share out these environmental liabilities. BAT and API agreed to assume liability for 55% up to a total of $75 million. It was later determined that liability in excess of that amount would be allocated as to 60% to BAT and API. There was also agreement in relation to possible liability for further identified decontamination sites (“Future Sites”) where NCR or API might have “arranger” liability (that is to say, liability for facilitating or contributing indirectly to contamination). One such site was the Kalamazoo River in Michigan, in relation to which the first intimation of liability was issued in 1998 and a request for information from the Environmental Protection Agency was received by NCR and API in 2003.

By 2000, it was clear that API would have a substantial liability in relation to the Lower Fox River, though its amount was uncertain, and there was a risk of a future claim in relation to the Kalamazoo River and other Future Sites.

In that same year, AWA was acquired by Sequana. It sold off API in 2001 on terms that AWA would indemnify API against certain environmental liabilities. In that way, both BAT and AWA had contingent liabilities in respect of API’s direct and indirect liability for remediation costs and NRDs. AWA purchased an insurance policy (“the Maris policy”) to pay for these future liabilities. By November 2008 the policy was worth about $250 million.

After the sale of API, AWA ceased to trade. The proceeds of sale of AWA’s businesses were lent to Sequana, with the result that, in time, the only assets of AWA were the inter-company receivable from Sequana, the Maris policy and certain other historic insurance policies. By the end of 2006, the Sequana receivable was valued at £464.6 million in AWA’s accounts, which showed a fully paid-up share capital of in

In 2008, the Directors decided to explore ways of releasing to Sequana tied up capital in AWA. To achieve this, they proposed to reduce the share capital of AWA from €318.6 million to €1 million and then pay one or more large dividends to Sequana, which could be set off against the inter-company receivable. Before these steps were taken, PwC audited the 2007 accounts. This work was completed on 28 October 2008

On 15 December 2008, the Directors each signed a solvency statement and Sequana, as sole shareholder, passed a special resolution to reduce the share capital. On the following day, AWA prepared a new set of interim accounts that reflected the reduced share capital (“the December interim accounts”). This time, the Lower Fox River provision was €58.4 million, which derived from new estimates of the aggregate remediation liability provided to AWA in November 2008. On 17 December 2008, the board of AWA approved the December interim accounts and resolved to pay the December dividend by way of set-off against the receivable. This left an outstanding balance of the Sequana debt of €142.5 million.

In the first part of 2009, the Directors undertook work on the necessary Lower Fox River provision for the 2008 annual accounts (to 31 December 2018). The conclusion was eventually reached that the Maris policy was sufficient to cover the best estimate of liability and that no further provision was therefore needed. On 18 May 2009, PwC gave an unqualified certificate that the accounts gave a true and fair view of the state of AWA’s affairs. These accounts (“the 2008 accounts”) showed distributable reserves of €137 million. The Directors approved them. On the same day, the board of AWA resolved to pay the May dividend by way of set-off against the Sequana debt.

Still on the same day, Sequana sold AWA to its former general counsel, Mr Gower, then acting as a consultant to AWA, and another connected person who had advised AWA in relation to the liability issues. From that time, Sequana was no longer exposed to any risk that its debt to AWA would have to be used to fund AWA’s environmental indemnity liabilities, and AWA was left with assets of only about €3 million in excess of the Maris policy to meet any such liabilities.

In none of AWA’s relevant accounts was provision or disclosure made in relation to potential liability at the Kalamazoo River or other Future Sites.

When, within less than a year after these events, it became clear that NCR and API’s liability and therefore AWA’s and BAT’s exposure was significantly greater than the value of the Maris policy, and claims were notified in respect of the Kalamazoo River, the question of the lawfulness of the December and May dividends was considered. AWA in due course brought the claim against Sequana and its former directors. The benefit of that claim was assigned to BTI in September 2014 as part of a funding agreement with BAT.

BTI issued the claim against PwC within a month of the assignment of the claim against Sequana and the directors. BAT had previously brought its own claim as creditor against Sequana, under section 423 of the Insolvency Act 1986, seeking an order for repayment of the dividends. This claim was heard by Rose J at the same time as BTI’s claim against Sequana and the Directors. BAT succeeded in relation to the May dividend but not the December dividend. The Court of Appeal upheld her decision, but 9 days after that, on 15 February 2019, Sequana entered insolvency protection in France and on 15 May 2019 went into compulsory liquidation. No part of Sequana’s liability to BAT has been paid.

The issues that Rose J had to decide in the first claim were, in broad terms, the following:

Section 836 of the 2006 Act, headed Justification of distribution by reference to relevant accounts, provides as follows, so far as material:

“(1)

Whether a distribution may be made by a company without contravening this Part is determined by reference to the following items as stated in the relevant accounts –

(a)

profits, losses, assets and liabilities;

(b)

provisions of the following kinds –

(i)

where the relevant accounts are Companies Act accounts, provisions of a kind specified for the purposes of this subsection by regulations under section 396;

(ii)

where the relevant accounts are IAS accounts, provisions of any kind;

(c)

share capital and reserves (including on distributable reserves).

(2)

The relevant accounts are the company’s last annual accounts, except that –

(a)

where the distribution would be found to contravene this

Part by reference to the company’s last annual accounts, it may be justified by reference to interim accounts……

…..

(3)

The requirements of –

section 837 (as regards the company’s last annual accounts), section 838 (as regards interim accounts), and…

must be complied with, as and where applicable.

(4)

If any applicable requirement of those sections is not complied with, the accounts may not be relied on for the purposes of this Part and the distribution is accordingly treated as contravening this Part.”

The directly relevant parts of sections 837 and 838 of the 2006 Act are the following:

“837 Requirements where last annual accounts used

(1)

The company’s last annual accounts means the company’s individual accounts –

(a)

that were last circulated to members in accordance with section 423……

(2)

The accounts must have been properly prepared in accordance with this Act, or have been so prepared subject only to matters that are not material for determining (by reference to the items mentioned in section 836(1)) whether the distribution would contravene this Part.

(3)

Unless the company is exempt from audit and the directors take advantage of that exemption, the auditor must have made his report on the accounts.

(4)

If that report was qualified –

(a)

the auditor must have stated in writing (either at the time of his report or subsequently) whether in his opinion the matters in respect of which his report is qualified are material for determining whether a distribution would contravene this Part, and

(b)

a copy of that statement must –

(i)

in the case of a private company, have been circulated to members in accordance with section 423….

The relevant accounts for the purposes of the December dividend were the December interim accounts, not the 2007 accounts, and so section 838 applied in relation to that dividend. The interim accounts therefore had to enable a reasonable judgment to be made (by the Directors) as to the amounts of AWA’s profits, losses, assets and liabilities. It was common ground before Rose J that the statutory requirements in this regard were equivalent to the requirement that individual accounts be properly prepared in accordance with section 396, so as to give a true and fair view, though of course there was no requirement for the interim accounts to be audited.

The relevant accounts for the purposes of the May dividend were the 2008 accounts, and so section 837 applied as regards that dividend.

In both cases, the requirement (in effect) that the accounts be properly prepared in accordance with the 2006 Act is qualified, for the purposes of relevant accounts justifying a distribution, in that deficiencies in the accounts that do not prevent the directors from determining whether a distribution may be made in accordance with section 836 are immaterial. It was on that basis that Rose J held that any deficiencies in disclosure were irrelevant, since they were immaterial for determining the matters stated in section 836. She held that the December interim accounts did enable a reasonable judgment to be made as to the amounts of the items in section 836(1) and that the 2008 accounts were properly prepared in accordance with the Act, thereby giving the Directors the power to declare the December and May dividends respectively. Other issues that Rose J decided (in particular whether the directors breached their duties in exercising the power and whether either dividend was a transaction made for the purpose of putting assets beyond the reach of a possible creditor) are not directly material to the issues with which this judgment is concerned and I need explore them no further.

In reaching her decision, Rose J heard detailed expert accountancy and US environmental law (“CERCLA”) evidence called by each side.

In relation to main issue (ii) above (appropriate provision for Lower Fox River liability), the central issue that the Judge had to decide was whether the provision made in the relevant accounts was the “best estimate”, in accordance with FRS 12, of the value of the Lower Fox River liability. She held that the provision had to be equal to the best estimate (at the time when the accounts were approved) of the amount that would be needed to buy off that liability. The making of the best estimate was, of course, ultimately a matter for the Directors, not the auditors.

An important issue arose in argument in that regard as to what question exactly the Judge decided. Did she decide (having heard evidence from each of the Directors as well as the expert accountancy and CERCLA witnesses) that the provision made in each set of accounts was the best estimate that the Directors could have made on the basis of the information that they in fact had about the liability, or did she decide that it was objectively the best estimate of the value of the liability on the basis of information that would have been reasonably available to the Directors at the time? That is a question of importance because BTI argues in this claim that if the auditors had audited the 2007 and 2008 accounts non-negligently, the Directors would in fact have had more and better information available to them than they did in fact have about the extent of the liability and risks. With the benefit of that further information, BTI contends, the Directors would have made different provision or been required to make disclosure in relation to such liabilities and in consequence they would not have declared the dividends that they did (or any dividend).

Having considered the judgment of Rose J, it is evident that she assessed the appropriateness of the estimate made by the Directors in the light of the information that they had but also the expert evidence that was given about the extent of the potential liability and the legal basis for it. Rose J identified the test under FRS 12 as being “the amount that an entity would rationally pay to settle the obligation at the balance sheet date” (para 376(a)). She refers in numerous places to the expert evidence of the CERCLA lawyers. Although she did not find it necessary to resolve points of significant disagreement between the legal experts, it is clear that she took their evidence into account in deciding whether the provision in the accounts was a best estimate of the amount that would have to be paid to rid AWA of the liability, together with hearsay evidence of the opinions of Mr Bates and Mr Gower and reports produced by Aon and CCL in November and December 2008 on which Mr Gower relied. Since the Judge did not at any stage decide which of the expert witnesses she preferred on any issue (and said that she found them both very helpful to her understanding), it is reasonable to conclude that she took their opinions as representative of a range of expert opinions that could be held at the relevant times (December 2008 and May 2009) about the extent of liability.

In the event, the only element of the computation of the provision in the December interim accounts that Rose J needed to decide was the reasonableness of the computation of NCR and API’s NRDs liability, in the figure of $35 million. That is because challenge to the likely aggregate amount of remediation costs and the appropriateness of taking 60% of such costs as representing the NCR/API share of liability had been abandoned by BTI during the trial. The two challenges to the NRDs provision (para 383 of the Judgment) were to the appropriateness of an assumed settlement with the government as the basis of the best estimate and the justification for a settlement figure of $35 million.

In respect of the settlement question, Rose J relied on the experts’ evidence that such claims generally tended to settle and the fact that a settlement offer had previously been made. Although her conclusion is that “there was plenty of evidence on which the directors could conclude that the best estimate of likely liability for NRDs should be based on the likelihood of settlement with the Government rather than on the pessimistic view that the case would fight”, that is clearly not a conclusion reached only on the basis of the Directors’ actual knowledge. In respect of the quantum, the Judge specifically reviews the experience of another company, Georgia Pacific, in settling its NRDs liability with the Government, as well as discussions between the Directors and Mr Gower and others that in fact took place about a figure of $35 million. The Georgia Pacific evidence was provided by the expert witnesses.

Rose J then held that the Directors were entitled to have regard to Mr Gower’s opinions and concluded:

“… there is no reason to doubt that the NRD figure of $35 million was the best estimate of all those involved in calculating the provision on the basis that this was the likely settlement figure that would ultimately be negotiated with the Government to settle NCR/API’s liability.”

Although, in isolation, this conclusion appears ambiguous as to the subjectivity or objectivity of the best estimate, it is clear, in context, that the Judge was not simply asking a narrow question, namely whether on the basis of the limited information that the Directors had their conclusion was the best estimate that they could make. Had that been the approach, she would not have had to consider (and reject) the argument that the Directors should have sought independent specialist legal advice from someone other than Mr Gower. There would also have been no reason for her to refer to the expert opinion evidence in reaching her conclusion.

In relation to the 2008 accounts, the two challenges to the adequacy of the provision that the Judge had to determine were: whether a reduction in NCR/API’s share of liability from 60% to 38% was justified, and whether a reduced figure for NCR/API’s NRDs liability of $18.8 million was reasonable as a best estimate.

In addressing the first challenge, Rose J reviewed various events that had taken place between December 2008 and May 2009. One of the Directors, Mr Martinet, explained in his evidence that he took these factors into account when assessing the best estimate. In addressing those factors, the Judge took into account not just what the Directors actually knew about them but what the expert witnesses said about their significance. The question ultimately, on the basis of all the evidence, was whether the Directors were justified in reducing the anticipated share of liability, i.e. whether it was reasonable for them to do so. That, by its nature, is an objective question and the Judge approached it in that way. She addressed, by reference to the expert evidence, factors that BTI alleged should have caused the Directors to increase rather than reduce the amount of the provision. She also referred to what the Directors said about what they knew at the time and the advice that Mr Gower and others had given.

In paragraph 428 of the judgment, Rose J concluded:

“I have not seen anything in all the evidence before me which suggests either that that reduction was unreasonable or that any of the many other people involved in this exercise expressed a view that the figure should be something different”

and in para 428 she said:

“I find therefore that there was plenty of evidence which justified Mr Martinet’s and Mr Courteault’s conclusion that 38% was the best estimate as at 18 May 2009 of NCR/API’s likely ultimate share of the costs of the Fox River remediation.”

Again, although that final finding may be ambiguous on the subjective/objective issue, in context it is clear that the Judge was considering objectively whether the reduction in the provision was reasonable, based on the evidence actually available to the Directors and on the expert evidence that she had heard. In no respect did the Judge say that the Directors made the best estimate that they could, though other information – if brought to their attention – might or would have required a different estimate to be made.

In relation to the second point of challenge, Rose J identified the basis on which the provision for NRDs liability had changed from the provision in the December interim accounts and referred to the views expressed by Mr Gower and Mr Bates to the Directors at the time as well as to the opinion of Mr Tenpas, one of the expert CERCLA witnesses. The conclusion expressed is that the estimate of $18.8 million has plenty of support in the evidence and was reasonable as a best estimate. That is clearly an objective assessment of the adequacy of the estimate.

The position, on these matters, is that BTI called expert evidence of US CERCLA law at the first trial, to support its case on what provision was appropriate for the likely liability of NCR/API and why the provision made was insufficient. Rose J took that evidence into account in deciding whether or not the estimate of AWA’s exposure was a reasonable best estimate made by the Directors but did not decide which of the experts’ opinions on disputed matters she preferred. Rose J did not decide whether NCR/API’s liability was reasonably assessed at 60% of the total remediation liability because BTI conceded that issue after the evidence had been given.

Judgment on the first trial was given on 11 July 2016. BTI appealed but ultimately not against the findings relating to the adequacy of the provision in the accounts. BTI only appealed on the ground that the Directors had acted in breach of their fiduciary duties by not taking into account the interests of the creditors of AWA when they declared the dividends. The starting point of the Court of Appeal’s analysis was therefore that BTI and BAT accepted that the May dividend was paid in compliance with Part 26 of the 2006 Act.

The claim against PwC had in the meantime been stayed, to await the outcome of the first claim, and later the appeal. As I have said, this claim was issued promptly by BTI and was served on PwC in February 2015. The original Particulars of Claim were served on 29 May 2015. At that stage, there were separate communications between solicitors acting for BTI and the Sequana defendants on the one hand and solicitors acting for BTI and PwC on the other hand relating to the conduct of each set of proceedings.

BTI had raised with PwC by letter dated 10 April 2015 various issues about compliance with the pre-action protocol. On 14 April 2015, PwC agreed to respond in due course, but had not done so when BTI wrote again on 5 May 2015, alleging that it appeared that PwC had no intention to engage meaningfully. BTI then referred to the Sequana proceedings and stated:

“It has become increasingly clear to us that there is an overwhelming level of overlap between the claims against your client and the claims against Sequana and the Former Directors, such that they clearly ought to be tried together. The legal, factual and accounting issues are the same or very similar in both sets of claims. A court trying each set of claims will have to consider the same events, same documents and nearly all of the same issues. The expert and the factual evidence required in each trial would have to deal with the same issues. There would generally be a vast duplication of time, effort and cost involved in having the two sets of claims tried separately. There would also be a very serious risk, to put it at its lowest, of inconsistent findings of fact and law. In light of the above, our clients intend to make an application to have the [Sequana] and BAT Dividend Claims tried with this Claim.”

BTI proposed a trial in October or November 2016, noting that this would necessitate the adjournment of the trial in the other claims, which by that time had been fixed for February 2016. BTI stated that they had written separately to the Sequana defendants in that regard and invited PwC’s comments.

On 6 May 2015, PwC sent a response to the earlier letter about the pre-action protocol, informed BTI that they had received the letter dated 5 May 2015 and said they would await BTI’s response on the pre-action protocol issues before they would comment on the 5 May letter. BTI replied by letter dated 15 May 2015 and said they would return to the joint trial issue once the other issues raised in their reply had been dealt with. After receipt of the Particulars of Claim in the PwC claim, BTI wrote again pointing out that the overlap with the other claims was “obvious and overwhelming” and stating that they had already made applications in both sets of proceedings for a joint trial. The hearing date was fixed for 2/3 July 2015. PwC were told that Sequana and the Directors had indicated their opposition to the application.

In a witness statement in support of the application, Mr Lloyd of BTI’s solicitors gave examples of the high level of overlap in the issues arising in the respective claims. He pointed to the background facts, the relevant transactions (being the declaration of the two dividends) and further stated:

“A central question in both the Dividend Claims and the PwC Claims is whether either the 2008 or the 2009 Dividend could lawfully be paid by [AWA] to Sequana under the Companies Act 2006, and in particular whether [AWA]’s 2007 and 2008 Accounts, and various interim accounts which were based on those annual accounts, had been properly prepared and, if not, whether they would if they had been properly prepared shown [AWA] as having distributable profits. This gives rise to the identical issues of fact and law across all 3 cases.”

Mr Lloyd also pointed out that, if the claims were not heard together, it was likely that third party disclosure would be sought against PwC, and further that there would be a risk of inconsistent findings and ultimately inconsistent judgments of the court. He accepted, in the application, that a joint trial would mean that the current hearing date for the other claims would have to be vacated.

In evidence in reply, PwC asserted that the claim against it was misconceived and that it wished to apply to strike out, or for summary judgment, on various grounds. PwC understandably wished to avoid the substantial cost of a trial where the claim was “completely misconceived”. Its position was therefore that it would be premature to make an order for a joint trial that would deprive it of its procedural rights to seek a summary adjudication of the claim against it. PwC acknowledged the substantial overlap of the claims and that, if the separate claims had been brought at the same time, there would have been much to be said for managing them together towards a single trial. PwC said that if the court nevertheless gave directions for a single trial, the earliest possible trial date would be January or February 2017.

The applications were listed before Mann J on 3 July 2015, following a day’s prereading of the papers. To be heard at the same time were applications by BTI in the Sequana claim to amend the Particulars of Claim and for permission to adduce expert US law evidence. The parties exchanged lengthy skeleton arguments. PwC maintained its position as set out in correspondence: its stance was therefore to oppose a joint trial. Sequana and the Directors strongly opposed any adjournment of the fixed trial date, on the basis that it was “set in stone” by previous agreement; that any later trial date would be uncertain, in view of interim applications to be brought by PwC; that the allegations against the individual directors were very serious and had been hanging over them for some time; and that there was no prejudice to BTI in having separate trials. There was therefore opposition from all defendants, on cogent grounds, to the joint trial proposal.

On the day before the hearing of the applications, BTI reached agreement with the Sequana defendants that their trial would proceed in the existing trial window and that permission to amend and for expert evidence would be granted. PwC was notified of this late in the evening and was told that BTI would therefore not persist in its application against PwC.

At the hearing, Mann J was informed that the application against PwC fell away, as a consequence of the agreement reached with the Sequana defendants. The Judge reviewed the terms of the consent order with which he was presented and made some suggestions of changes to it. The order eventually made in the PwC claim was a consent order, which dismissed the application for a joint trial, gave directions for

Subsequently, the strike out application was issued on 27 July 2015 and Mann J then gave further directions for filing evidence and a three-day hearing. By a further consent order dated 27 October 2015, the parties agreed to stay the PwC claim until the other claims had been concluded. They later agreed to continue the stay during BTI’s and BAT’s appeal against Rose J’s order.

Each party seeks to extract matters to its advantage from the procedural steps that were taken and the agreement reached in July 2015. BTI argues that it properly sought to have a joint trial of all the claims, in order to save time and costs and avoid the risk of inconsistent decisions, but that it realistically recognised (and the court endorsed its view) that this was not possible. Indeed, the court – albeit by consent – dismissed its application for a joint trial. In those circumstances, BTI contends, it cannot on any view be an abuse of the court’s process for it now to seek to have a trial of the PwC claim. The agreement to stay the PwC claim pending the outcome of the other claims was a pragmatic and sensible course, in the hope that – whatever the outcome of the other claims –it might be unnecessary to have a full trial of the PwC claim.

PwC argues that BTI could and should have pursued its application for a joint trial and that it compromised its application for tactical reasons, namely obtaining the Sequana defendants’ agreement to the relief sought in its other applications and to ensure the absence of PwC (who might be expected to argue for the adequacy of the audited accounts in all respects) from the trial against Sequana and the Directors. PwC contends that it was implicit in the first consent order that, if a claim that BTI had characterised in its application as raising substantially the same or identical central issues as the PwC claim was unsuccessful at trial, the claim against PwC would not be pursued.

In my judgment, both sides are reading too much into the disposal of the joint trial applications. The application was properly made by BTI, who acknowledged that there were good reasons for having a single trial and wished to have one. There were indeed good case management reasons for having a single trial, all other things being equal. But all other things were not equal at that time. It would have been unfair to make the Sequana defendants wait up to a year longer for their trial, and it would have been unfair to deprive PwC of the opportunity to strike out the claim against them at an early stage. The merits of the applications were finely balanced but there was every prospect that the applications, if pursued, would have failed. The dismissal of the application in the PwC claim by consent cannot be treated as exactly the same, for present purposes, as if it had been dismissed because it failed. The dismissal of the application was a convenient means of disposing of it, by agreement. If the proposed course had been obviously inappropriate, there is no doubt that Mann J would have resisted such a course being taken. It can therefore properly be inferred that there was nothing inherently wrong with the prospect of two successive trials on substantially the same subject matter taking place, in the event that PwC’s strike out application failed.

I reject the suggestion by PwC that it was implicit in the consent order that if the Sequana claim failed BTI should not be entitled to proceed against PwC. If that had been implicit, the court itself (which had read in detail over the previous day) would

BTI sets out a summary of the basis of its claim against PwC in para 5 of the APOC: “PwC audited [AWA]’s financial statements for the periods ending 31 December 2007 (“the 2007 Accounts”) and 31 December 2008 (“ the 2008 Accounts”). The Claimant’s claims are (in summary) for breach of duty and negligence in connection with PwC’s involvement in and auditing of the 2007 Accounts and/or the 2008 Accounts as a result of which [AWA] paid dividends to its parent and sole shareholder, Sequana, which were unlawful and/or would not otherwise have been paid of (i) €443 million on or about 17 December 2008… and/or (ii) €135,181,358.55 on or about 18 May 2009…”

BTI’s case on causation is that but for the 2007 accounts, the December dividend could not and would not have been paid. The relevant accounts for this purpose were the December interim accounts, which were in material respects the result of the deficient 2007 accounts that PwC audited. Its case in relation to the May dividend is similar, save that in this case the relevant accounts were the 2008 accounts. The dividends, BTI alleges, were paid by reason of PwC’s breaches of duty. It alleges that, had PwC carried out its audits non-negligently, AWA’s directors would have increased the provision in the December interim accounts and revisited and increased the provision in the 2008 accounts, or alternatively would not have prepared these accounts on a going concern basis, and that in the circumstances no dividends could or would have been paid to Sequana. Alternatively, it is pleaded that these consequences would have followed if PwC had declined to give an unqualified audit opinion, either at all or on a going concern basis, or alternatively required disclosure in relation to the going concern basis or further contingent liability in relation to the Lower Fox River and the Kalamazoo River.

The loss claimed by BTI is pleaded as follows:

“By reason of PwC’s breaches of duty as aforesaid, [AWA] suffered loss and damage. The Claimant claims damages in the full amounts of the 2008 Dividend and/or the 2009 Dividend, alternatively the amounts of the said dividends which, but for

At para 207 of the APOC, BTI pleads that the objective purpose of PwC’s two relevant audit reports included protecting AWA from losses caused by the payment of a dividend in reliance on a negligently prepared audit report, and alternatively, at para 208, that PwC assumed responsibility to exercise a duty of care as regards such loss.

In between the summary of the claim and these allegations of causation and loss, BTI pleads at great length the factual background to the issues of liability of NCR/API, the plan to reduce AWA’s capital and pay dividends to Sequana in order to reduce the inter-company receivable, and PwC’s role and knowledge as auditor of AWA over many years. In particular, it is alleged that PwC knew about AWA’s indemnity liabilities, that these related to the Lower Fox River and the Future Sites, that previous audited accounts of AWA contained provision for these liabilities and disclosure of contingent liabilities, that there was a significant risk of very high liability, and that the Directors intended to reduce AWA’s capital and declare dividends to Sequana to be set off against the inter-company receivable, so that Sequana could then sell AWA and remove any risk of liability in future.

The core allegations relating to PwC’s audit of the 2007 accounts are:

BTI alleges that PwC failed to perform appropriate audit procedures and to obtain sufficient and appropriate audit evidence to identify and properly assess the risks of material misstatement of the provision in the 2007 accounts, and that it ought not to have expressed an unqualified opinion where liability was based on an assumed NCR/API share of 60% of the remediation costs. It is alleged that PwC failed to obtain advice from an appropriately qualified and independent expert in various respects relating to the basis of liability and the extent to which other polluters might be required to contribute. BTI pleads that such advice would have led to the conclusion that there was a substantial likelihood that NCR/API would have to bear 100%, or at least well in excess of 60%, of the remediation costs.

BTI further pleads that PwC should have known that reliance on the views of Mr Gower and others connected to management of AWA was inappropriate, and that they should not have expressed an unqualified opinion in the absence of disclosure in relation to the Lower Fox River and the Kalamazoo River. PwC should have taken account of the case management order (including the order for a trial in December 2009 of knowledge and fault issues) in the Whiting litigation and, in general, PwC should have gone much further in testing and enquiring into the basis of management’s assumptions about, and evidence of, the extent of liability. Had it done so, there could and would have been no unqualified opinion that entitled the Directors of AWA two months later to approve the December interim accounts and declare the December dividend.

The core allegations relating to the 2008 accounts are that –

BTI alleges that PwC failed to perform appropriate audit procedures and to procure sufficient and appropriate evidence to identify and properly assess the risks of material misstatement in the draft 2008 accounts, which made no provision for Lower Fox River liability in excess of the value of the Maris policy, and that such provision should have assumed NCR/API’s liability for 100% of the remediation costs, or at least substantially in excess of 38% of them. It also alleges that PwC should not have accepted an assumption that AWA would have no liability in respect of the OU1 zone or that the best estimate of its liability for NRDs was $11.3 million.

BTI alleges that PwC failed to obtain advice from an appropriately qualified, independent expert in various respects relating to the basis of liability, in particular the impact of the Burlington Northern decision and the extent to which other polluters might be required to contribute. BTI pleads that such advice would have led to the conclusion that there was a substantial likelihood that NCR/API would have to bear

BTI further pleads that PwC should have known that reliance on the views of Mr Gower or others connected with management was inappropriate; and should not have accepted without proper investigation a reduction in the proportion of NRD/API’s liability from 60% in the 2006 and 2007 accounts to 38%. Further, PwC should have concluded that the going concern assumption was inappropriate, or alternatively should not have accepted that the accounts gave a true and fair view on that assumption without disclosure of the material uncertainty about AWA’s ability to continue as a going concern. PwC should have required disclosure in relation to the Lower Fox River of contingent liability in excess of the Directors’ estimate and should not have expressed an unqualified opinion in the absence of disclosure in relation to a Kalamazoo contingent liability.

From this review of the APOC, it is apparent that there is some overlap, indeed a significant overlap, between the issues that were decided by Rose J (and issues that she would have decided but for a concession by BTI after the evidence had been heard) and the issues raised against PwC. The main issues that were decided by Rose J, or were conceded, and that would be directly or indirectly in issue again are:

However, in this claim, those issues would fall to be considered against a backdrop of the 2007 accounts that BTI contends should and would have been materially different had PwC complied with its duty to AWA as auditor. Rose J was not concerned with the propriety of the 2007 accounts and did not consider whether they gave a true and fair view. She was only concerned with: the November 2008 interim accounts, on the basis of which the declarations of solvency for the capital reduction were made; the December interim accounts, on the basis of which the December dividend was declared; and the 2008 accounts, on the basis of which the May dividend was declared.

Although there is factual overlap between the issues now raised in relation to the 2007 accounts and the issues decided by Rose J, the former fall to be considered at an earlier time, October 2008, when PwC audited the 2007 accounts. If PwC should have investigated further at that time, rather than in May 2009, the question is what further information, cautions or views they should have provided to the Directors at that time, or what information or evidence they should have required the Directors to obtain.

What the Directors would have done in October 2008 in consequence may (and, BTI submits, would) impact on the content of the December interim accounts and the 2008 accounts. The issues here are therefore different: not just whether the provision made in the December interim accounts and 2008 accounts was the best estimate or whether those accounts give a true and fair view, but what provision or disclosure would in fact have been included in the 2007 accounts, the December interim accounts and the 2008 accounts if PwC had carried out its audit non-negligently in October 2008 and May 2009.

New issues raised in this claim in relation to the 2008 accounts are the modelling error, the appropriate treatment of OU1 liability, the appropriateness of the going concern assumption or any disclosure relating to that, and whether a contingent liability should have been disclosed in relation to the Lower Fox River.

In order to succeed on this claim, BTI needs to prove a number of matters, not merely that PwC failed to do what reasonable auditors would have done in October 2008 and May 2009. Mr Salzedo is therefore right in principle to submit that it is not sufficient for BTI to say that it is entitled to see the allegations of negligence investigated in a further claim. That is merely the starting point, though the issues raised by the allegations of breach are relevant to the ultimate assessment of whether the claim is abusive and will colour the issues of causation and loss.

In relation to the allegedly negligent audit of the draft 2007 accounts, BTI has to establish that if PwC had acted non-negligently in October 2008 the Directors would have become aware of further information or advice, or would better have understood the risks, which would have led them to make different provision for the Lower Fox River liabilities and/or disclose contingent liabilities for the Lower Fox River and the Kalamazoo River, or alternatively that PwC would not have provided a clear audit report on the basis of the draft 2007 accounts; and that as a result the Directors would have carried forward those matters into the December interim accounts and would not have declared the December dividend, either at all or in the same amount.

In relation to the allegedly negligent audit of the draft 2008 accounts, BTI has to establish that if PwC had acted non-negligently in October 2008 and May 2009 the Directors would have become aware of further information or advice, or would have better understood the risks, which would have led them to make different provision for the Lower Fox River liabilities and/or make disclosure of contingent liabilities for the Lower Fox River and the Kalamazoo River, or decide that the accounts could not properly have been prepared on a going concern basis or without disclosing a material uncertainty in that regard, or alternatively that PwC would not have provided a clear audit report on the basis of the draft 2008 accounts; and that as a result the Directors would not have declared the May dividend, either at all or in the same amount.

It follows that for BTI to succeed on breach of duty and causation, it will have to show that the December interim accounts and the 2008 accounts would have been different (in various respects) from the accounts that Rose J held showed a true and fair view of AWA’s affairs.

In relation to these allegations, PwC’s case is that BTI’s claim is or involves a collateral attack on the judgment of Rose J because she decided core issues (on which BTI must succeed in its claim against PwC) against BTI, in particular the appropriateness of the December interim accounts and the 2008 accounts. PwC submits that that is an abuse of process because BTI is seeking to re-litigate issues that were decided in a long trial involving the most relevant parties, in which BTI was able to call the relevant evidence and was not unable to call any material evidence. PwC submits that a claim against it is “downstream” from a decision on the issues that Rose J decided against BTI, in the sense that BTI first has to succeed on those issues before it has a valid claim against PwC.

Alternatively, PwC submits that there can realistically be no new material evidence that bears on the core issues, and accordingly there is no realistic prospect of BTI obtaining a different outcome, or of establishing a breach of duty. PwC submits that it has been decided by Rose J that the Lower Fox River provision in the December interim accounts and the 2008 accounts were the best estimates for the purposes of FRS 12 and that those accounts present a true and fair view of AWA’s affairs. If that is so then, by whatever process PwC reached the decision to provide an unqualified opinion on the audited accounts, that opinion cannot have been negligent.

The law on abuse of process where no estoppel per rem judicatam or issue estoppel arises was most recently summarised by the Court of Appeal in Michael Wilson & Partners v Sinclair [2017] EWCA Civ 3; [2017] 1 WLR 2646. In that case, the claimants issued an arbitration claim against Emmott, their former employee, alleging that he received shares as a bribe. The defendant declined an invitation by the claimants to become a party to the arbitration. The arbitrator rejected the claimants’ claim. The claimant then issued proceedings against the defendant, alleging that he had knowingly assisted Emmott to receive the shares as a bribe. The judge struck out the claim as an abuse of process, on the basis that the allegation was entirely inconsistent with and contrary to the findings made in the award, to which the claimants were a party; but the Court of Appeal reinstated the claim. It rejected an argument that there could be no abuse of process where the prior determination was an arbitration award, but stated that it would be a rare case where proceedings against a non-party to an arbitration could be said to be an abuse.

The Court pointed out that the defendant was seeking to take the benefit of an arbitration award by which he would not have been bound had it been decided in favour of the claimants. Simon LJ stated that this lack of mutuality was a highly material if not dispositive factor. He reviewed a series of well-known decisions on abuse of process and stated, at [48]:

“The following themes emerge from these cases that are relevant to the present appeal.

(1)

In cases where there is no res judicata or issue estoppel, the power to strike out a claim for abuse of process is founded on two interests: the private interest of a party not to be vexed twice for the same reason and the public interest of the state in not having issues repeatedly litigated; see Lord Diplock in Hunter’s case [1982] AC 529, Lord Hoffmann in the Arthur J S Hall case [2002] 1 AC 615 and Lord Bingham in Johnson v Gore Wood & Co [2002] 2 AC 1. These interests reflect unfairness to a party on the one hand, and the risk of the administration of public justice being brought into disrepute on the other, see again Lord Diplock in Hunter’s case. Both or either interest may be engaged.

(2)

An abuse may occur where it is sought to bring new proceedings in relation to issues that have been decided in prior proceedings. However, there is no prima facie assumption that such proceedings amount to an abuse: see Bragg v Oceanus [1982] 2 Lloyd’s Rep 132; and the court’s power is only used where justice and public policy demand it, see Lord Hoffmann in the Arthur JS Hall case.

(3)

To determine whether proceedings are abusive the court must engage in a close ‘merits based’analysis of the facts this will take into account the private and public interests involved, and will focus on the crucial question: whether in all the circumstances a party is abusing or misusing the court’s process, see Lord Bingham in Johnson v Gore Wood & Co and Buxton LJ in Laing v Taylor Walton [2008] PNLR 11.

(4)

In carrying out this analysis, it will be necessary to have in mind that: (a) the fact that the parties may not have been the same in the 2 proceedings is not dispositive, since the circumstances may be such as to bring the case within ‘the spirit of the rules’, see Lord Hoffmann in the Arthur JS Hall case; thus (b) it may be an abuse of process, where the parties in the later civil proceedings were neither parties nor their privies in the earlier proceedings, if it would be manifestly unfair to a party and the later proceedings that the same issues should be re-litigated, see Sir Andrew Morritt V-C in the Bairstow case [2004] Ch 1; or, as Lord Hobhouse put it in the Arthur JS Hall case, if there is an element of vexation in the use of litigation for an improper purpose.

(5)

It will be a rare case where the litigation of an issue which has not previously been decided between the same parties or their privies will amount to an abuse of process, see Lord Hobhouse in In re Norris ”

As was the case in Michael Wilson & Partners v Sinclair, these principles were substantially undisputed before me. It is on their application to the facts that the parties differ. It is to be noted that, where a defendant has not previously been sued,

Examples of such rare cases are the decisions in Laing v Taylor Walton, a decision of the Court of Appeal to which Simon LJ referred, and Arts & Antiques Ltd v Richards [2013] EWHC 3361 (Comm); [2014] PNLR 10, a decision of Hamblen J. PwC relies on those cases and submits that BTI’s claim is indistinguishable in principle from them and that, as a result, the BTI claim is abusive and should be struck out.

Mr Laing brought a claim against a former business partner, Mr Watson, to establish the true effect of agreements that had been made between them. There was a factual dispute about what exactly had been agreed orally that led to a later written agreement being made. The judge at the first trial rejected Mr Laing’s account of the disputed oral exchanges and accepted Mr Watson’s account, so Mr Laing lost. Mr Laing then sued the solicitors who had acted for him in preparing the written agreement, contending that they had negligently failed to translate correctly what had been agreed orally into writing. But what Mr Laing alleged had been agreed orally, which the solicitors had negligently failed to translate, was his account of the agreement with Mr Watson that the judge had rejected at the first trial. The allegation of negligence depended on first establishing that Mr Laing’s version of the oral agreement was true and was what the solicitors should have drafted. As Mr Salzedo put it, the claim against the solicitors was “downstream” from the terms of the oral agreement.

Buxton LJ, giving the leading judgment of the Court, said at [22] that:

“… Everything said to us… in criticism of H.H. Judge Thornton’s judgment could have been said to H.H. Judge Thornton (and mainly was so said); and could have been deployed in the appeal from H.H. Judge Thornton that was never brought. What is sought to be achieved in the second claim is, therefore, not the addition of a matter that, negligently or for whatever reason, was omitted from the first case, but rather a relitigation of the first case on the basis of exactly the same material as was or could have been before H.H. Judge Thornton.”

The Lord Justice was able to conclude that any material passing between Mr Laing and the solicitor had – so far as relevant – been deployed in the first trial, or could have been so deployed but Mr Laing had chosen not to do so. This led to the conclusion, at [25], that:

“… It would bring the administration of justice into disrepute if Mr Laing were to be permitted in the second claim to advance exactly the same case as was tried and rejected by H.H. Judge Thornton. If H.H. Judge Thornton’s judgment was to be disturbed, the proper course was to appeal, rather than seek to have it in effect reversed by a court not of superior but of concurrent jurisdiction hearing the second claim. That the second claim is in substance an attempt to reverse H.H. Judge Thornton is important in the context of wider principles of finality of judgments. In Hunter, at 545D, Lord Diplock said that the proper course to upset the decision of a court of first instance was by way of appeal. Where, wholly exceptionally, a collateral, first instance, action can be brought it has to be based on new evidence, that must be such as entirely changes the aspect of the case: see per Earl Cairns L.C. in Phosphate Sewage v Mollison (1879) 4 App Cas 801 at 814. ”

Buxton LJ rejected an argument that the Phosphate Sewage principle had no application where the second claim was between different parties. He stated at [27]:

“… In order to succeed in the new claim Mr Laing has to demonstrate not only that the decision of H.H. Judge Thornton was wrong, but also that it was wrong because it wrongly assessed the very matters that are relied on in support of the new claim. That is an abusive relitigation of H.H. Judge Thornton’s decision not by appeal but in collateral proceedings, and in substance if not strictly in form falls foul of the Phosphate Sewage rule.”

The Arts & Antiques case concerned a claim under an insurance policy. In an arbitration, the arbitrator found that the claimants had failed to satisfy the requirements of a condition precedent in the policy. An attempt to appeal or set aside the award was dismissed by the court. The claimant then brought proceedings against the brokers who had arranged the policy, arguing among other things that the condition precedent had not validly been incorporated in the policy and that the brokers had fabricated a copy of the policy incorporating it. Hamblen J held that the second claim fell to be struck out as an abuse of process in so far as it rested on an argument that the condition precedent had not formed part of the policy. The Laing case and the first instance decision in Michael Wilson (later reversed by the Court of Appeal) were cited to the judge. He said:

“I agree that the Taylor Walton v Laing decision supports a finding of abuse of process in this case. Aside from the fact that the earlier decision was in arbitration, the two cases are analogous. There is in this case no new evidence which casts doubt on the Arbitrator’s decision. Indeed, for reasons set out below, such further evidence as there is confirms the correctness of his decision. That decision has sought to be challenged by appeal but the application has been dismissed on the basis that the decision is ‘not open to serious doubt’. For the issue to be relitigated in this court involves a collateral attack on the Arbitrator’s final and binding decision. Further, that decision relates to the terms of the contract as between A & A and Zurich, which have been determined in accordance with the agreed contractual machinery, namely by arbitration. In all the circumstances, I conclude that it would bring the administration of justice into disrepute, and would be oppressive and unfair on

What was it about the facts of the Laing and Arts & Antiques cases that made them rare cases where litigation between different parties would bring the administration of justice into disrepute among right-thinking people and so amounted to an abuse of process? In both cases, the judges said that the second claim amounted to a collateral attack on the correctness of the first decision. A collateral attack exists not only where it is the purpose of a claimant by bringing a new claim to overturn a prior decision against him, but where a reversal of findings in that decision on the basis of the same evidence is a precondition of success in the second claim, even if it otherwise appears to raise a legitimate complaint. That is because such a claim necessarily involves alleging that the first decision was wrongly decided on the basis of the material that was before the first court.

It is clear, therefore, that highly material considerations are whether the very same issue is to be re-litigated and whether further cogent evidence is likely to be available to the claimant in the second claim, which the claimant did not reasonably have available to deploy in the first claim, or whether the same evidence will be deployed. Where, as in the Laing case, evidence was available at the first trial from the only directly relevant witnesses and there was no other material evidence on which the claimant could rely in the second claim that might alter the outcome, the second claim is necessarily a collateral attack on the correctness of the first decision. The proper means of challenging the correctness of the first decision is to appeal the findings.

Thus, although it is not prima facie an abuse of process to bring a second claim against a different party, who would not have been bound by any material findings in the first claim, and usually will not be so, it may be abusive in a case where success on the second claim will involve re-litigating the very same issues as in the first claim and the court reaching different conclusions on those issues on the basis of the same evidence. That is why a close “merits based” analysis of the facts of any particular case is required, and the ultimate question is a general one of whether, in all the circumstances, a party is abusing or misusing the court’s process.

PwC’s submissions on the first ground of its application are essentially these:

BTI’s case is that although there is overlap with the issues decided by Rose J, there are other important issues to be decided and, where there is overlap, there will be “a wealth of further material evidence” available to the court, including new documents from AWA, documents and witnesses from PwC, and additional expert evidence. BTI in fact places considerable emphasis on the expert evidence that will be adduced. This will be directed, in the first place, to the adequacy of provision/disclosure in the 2007 accounts, which were not questions that Rose J had to decide. BTI cites as an example of the material expert evidence the fact that both expert witnesses at the first trial agreed that disclosure was necessary in the 2007 accounts and that FRS 12 had not been satisfactorily complied with (although this evidence did not go to any issue that Rose J had to decide). I did not have that evidence before me, but in reply PwC did not take issue with this assertion.

BTI relies on the fact that there will be new issues that fall to be determined in this claim, albeit ones closely related to the issues considered by Rose J. Even if some of the same issues about the 2008 accounts will be live, the real question at this trial will be not whether the accounts as prepared showed a true and fair view but what different accounts would in fact have been approved by the Directors had PwC acted non-negligently in October 2008 and May 2009. In addition, there are further arguments about the 2008 accounts that were not directly raised before Rose J (for various reasons) and so were not decided by her, such as the OU1 liability, the going concern assumption and the modelling error.

Persuasively and attractively though Mr Salzedo developed his argument, I consider that this case is distinguishable in principle from the Laing case, and does not amount to a collateral challenge to Rose J’s decision. My reasons are the following.

To succeed in his claim against the solicitors, Mr Laing had to prove that they had failed to give proper effect to agreement “A” that he had made with Mr Watson. In proceedings between Laing and Watson, the judge had decided that the agreement reached was “B” not “A” and rejected Mr Laing’s evidence. The solicitors’ draft was in terms of “B”. There was no material evidence, other than the evidence that the judge had heard, on the question of what agreement Laing and Watson had reached. If “B” was the right answer, the solicitors had not been negligent. Mr Laing was therefore effectively gaming the justice system by re-running the same argument and

In this case, however, BTI contends that the Directors would in fact have made different provision and disclosure in the accounts if PwC had done what it should have done. It is not simply a claim based on what the relevant accounts should have stated: it is a claim based on what the Directors would have done had they had had the benefit of a non-negligent audit report.

It is not an abuse of process to bring a claim against a professional adviser contending that a previous claim that the claimant lost would not have been lost if the adviser had advised correctly: Arthur JS Hall & Co v Simons [2002] 1 AC 615. In that case, principally concerned with advocates’ immunity from suit, Lord Hoffmann explained at p.705H that although the lost claim is res judicata between the parties, there is no objection on public interest grounds to the claimant alleging that the outcome would have been different if the advocate had not been negligent. So here, BTI alleges that if PwC had been non-negligent, different information and advice would have led the Directors to make different provision and disclosure in the 2007 accounts, which would have been carried into the December interim accounts and the 2008 accounts.

To succeed, BTI does not (at least arguably) have to prove that those accounts did not show a true and fair view: it has to prove that, with correct advice from PwC, the Directors would in fact have made different provision and disclosure and in consequence of that would not then have paid the dividends. If a conclusion that with the benefit of proper advice and fuller information they would have done so is to any extent inconsistent with Rose J’s conclusion that the accounts show a true and fair view, that inconsistency is a logical consequence of BTI’s success in establishing (based on new evidence) that the Directors would have acted differently, not a necessary foundation for the success of the claim. There is nothing inherently abusive about proceedings that might reach a result that is inconsistent with a previous decision of the court, as Lord Hobhouse explained in the Arthur JS Hall case:

“To challenge in later litigation an earlier non-binding decision between different parties is not itself abusive, provided there are good reasons for doing so. So far as questions of law are concerned, the doctrine of precedent contemplates this. So far as questions of fact are concerned, each court has to try and decide questions of fact on the evidence adduced before it. Judicial comity and common sense take care of most situations in practice but the law does tolerate the possibility of apparently inconsistent decisions. The element of vexation is an aspect of abuse, the use of litigation for an improper purpose, trying to have repeated bites at the same cherry. The objectionable element is not the risk of inconsistency.”

Although there are issues relating to the 2008 audited annual accounts that will be reexamined, such as the appropriateness of the 38% provision, the question of what the

A further but related basis on which PwC alleges abuse is that, even if the claim is not a collateral attack on Rose J’s judgment, nevertheless there will be no different evidence on which BTI can rely from the evidence at the Sequana trial, and that accordingly the claim is either abusive or doomed to fail. The basis of PwC’s argument is that BTI adduced before Rose J the expert evidence that it said properly reflected the true risk and extent of liability at the Lower Fox River and Future Sites and the evidence of AWA’s directors explaining what they did; that all this evidence was taken into account by Rose J in determining whether the provision actually made was the best estimate, and that accordingly there can be no further or better evidence on which BTI can rely to argue that some different provision or disclosure should have been made.

If the issues for trial were limited to whether the December interim accounts and the 2008 audited annual accounts showed a true and fair view of AWA’s affairs, that argument would have had considerable force. BTI has not said that the expert evidence that it called at the Sequana trial was deficient, or wrong; however, it does intend to call different expert evidence to address the 2007 accounts and the new issues.

Although BTI argues that further documents from AWA and disclosure from PwC may provide support for the opinion of its expert witness, thereby making that evidence more compelling, it has not identified or described any such document or class of documents.

In relation to new AWA documents, BTI has had a contractual right since 2014 to enforce production by AWA of its documents. Mr Lloyd in his witness statement dated 28 May 2019 said that BTI would shortly have access to those further documents, but despite that, at the hearing in October 2019, no helpful document had been identified. The argument based on new AWA documents is therefore at best speculative and at worst unfounded. Disclosure from PwC may or may not assist BTI’s case on negligence, but there is no inference that can be drawn that there will be found a “smoking gun” indicating that PwC knew that the risk of liability was greater than that provided for in the 2008 accounts. BTI puts forward no argument that a particular document or type of document that will help its case is likely to exist; if it had thought so when pursuing the Sequana claim it is inherently likely that it would have made a non-party disclosure application, or issued a witness summons, but no such application was made. There may, of course, be documents on disclosure that support BTI’s case, but equally there may not be. BTI is waiting to see.

However, the argument of PwC that the evidence will be the same has to be considered in the context of BTI’s case that PwC was negligent in auditing the 2007 accounts and that these accounts should have made different provision or disclosure, and of the new issues with the 2008 accounts. There was evidence at the first trial that the 2007 accounts were in some respects deficient. The 2007 accounts were, however, not the material accounts in the first trial – Rose J did not have to evaluate that part of the evidence and it cannot be assumed that BTI adduced full evidence on all the shortcomings of the 2007 accounts. BTI will be entitled to call further expert evidence in relation to those accounts. I remind myself that on a strike out application it is not for BTI to prove what the further material evidence will be but for PwC to prove that there can be none.

For the reasons given above, if BTI succeeds in proving that the 2007 accounts should have been in materially different terms, that becomes part of the factual material on which the question of what provision the Directors would have made in the December interim accounts has to be considered; and then in turn the question of what provision or disclosure would have been made in the 2008 accounts. There will be evidence that was not before Rose J about the OU1 liability, the going concern basis of the accounts and the modelling error, all of which could justify a conclusion that the Directors would have made different provision or disclosure in May 2009 had they considered these matters.

I therefore conclude that PwC cannot establish that the evidence at trial will necessarily be the same (and no better for BTI) than the evidence at the first trial.

Standing back from the argument and considering the claim against PwC more broadly, it is not in the nature of an abusive claim that will bring the administration of justice into disrepute. The claim was live for 2 years before Rose J’s judgment and nothing in the terms of the stay prevented BTI from reviving it. It will be a rare case in which a claim against a non-party to an earlier claim will be an abuse of process by reason of the outcome of that other claim. That is particularly so if the claim had already been issued and there was a bona fide attempt to have the two claims heard together. As the issues and evidence in the claim against PwC will not be identical to those in the Sequana claim, there is no collateral attack and no abuse of process.

PwC’s alternative argument is that the claim is doomed to fail because there can be no material evidence that will make a difference to the outcome before Rose J or establish that any breach of duty caused the Directors to act as they did, and that therefore summary judgment should be given now. I have already addressed and rejected the argument that there can be no materially different evidence bearing on the question of what the Directors would have done in December 2008 and May 2009. I turn now to the argument based on causation.

Assuming for these purposes that PwC was in breach of duty, it argues that the Directors did not rely on its audit reports and that there is no arguable case that any further information, advice or expert evidence would have made a difference to what the Directors did. Before the audit of the 2007 accounts, the Directors had formed the intention (contained in “the September plan”) to reduce AWA’s capital, declare substantial dividends in favour of Sequana to remove the inter-company receivable and then sell AWA, in order to remove any exposure of Sequana to liability in respect of the Lower Fox River and Future Sites.

At all relevant times, submits PwC, the Directors had full knowledge of the exposure and risks and of various assessments of the extent of liability for these sites. Nothing that PwC could tell the Directors, either directly or in the form of external expert advice, would have made any difference to what the Directors knew or would have done. The Directors formed their own view based on full information at the time. PwC also point out that BTI plead that the Directors had an improper plan to shelter Sequana from risk and that, in relation to the May dividend, they were found to have acted in order to put assets beyond the reach of creditors.

PwC reminded me that BTI will have to prove that the breaches were a substantial cause of the losses claimed. Mr Salzedo referred to the cases of Berg Sons v Adams [1992] BCC 661 and Galoo v Bright Graeme Murray [1994] 1 WLR 1360 in support of the proposition that the (assumed) negligent misstatement by PwC must be a cause of the loss and that accordingly it must be shown that the Directors relied on PwC’s advice or audit report. If the Directors already knew the full picture and were determined to proceed in order to shelter Sequana, it cannot be said that the Directors relied on any misstatement by PwC.

I reject the argument that there is no realistic possibility of BTI proving reliance on a misstatement contained in the 2007 or 2008 audit reports. Rose J found that the Directors were acting conscientiously in seeking to make appropriate provision in the accounts for the Lower Fox River liability. It is not possible to say, on the basis of their evidence to Rose J, that they would have disregarded material information that they were given and ploughed ahead with the December (or May) dividend regardless. It cannot be concluded on this application that, if PwC had questioned the amount of the provision, or required disclosure of a risk of greater liability, or declined on the basis of the proposed provision to give an unqualified audit report on a going concern basis, the Directors would not have changed the contents of the accounts. Whether, in those circumstances, they would have made changes and would have declared the dividends that they did must be an issue for trial. It is not possible to say that if advice had been given that AWA should make provision based on the highest figure in a range of estimates of liability, that was something that the Directors already knew.

The Directors did in fact have various assessments of liability from management and connected, non-independent advisors, but they did not have independent advice from an expert lawyer or consultant about the risks or the validity of the assumptions that they were making, or about what would be a prudent provision to make. Nor had they any advice from PwC about qualifications or disclosure that were appropriate in the accounts. It is therefore not possible to conclude, without hearing the evidence at trial, that nothing that the Directors learnt could have made a difference.

Further, it is clear from the evidence given to Rose J that the Directors required a clear audit certificate from PwC in order to proceed as they did. The September plan required this. The timing of the submission of the draft accounts, the issue of the audit certificate and the meeting at which the May dividend was declared shows that PwC’s audit was the final matter to be put in place to allow the Directors to proceed – which they did immediately upon receipt. There is clearly an arguable case that the Directors relied on the terms of the PwC audit report and on the absence of any challenge in approving the 2007 and 2008 accounts and declaring the dividends, and if they did so they relied on any misstatement contained in PwC’s reports.

PwC contends that, as a matter of law, loss suffered by AWA by the payment to Sequana of the disputed dividends is not loss that PwC owed AWA a duty of care to prevent. While acknowledging that in several cases it has been held that unlawfully paid dividends are recoverable as damages from a negligent auditor (see, e.g., Leeds Estate, Building and Investment Co v Shepherd (1887) 36 ChD 787 at 809; Re London and General Bank (No.2) [1895] 2 Ch 673 at 686-688), PwC submits that those decisions were made before the landmark decision of the House of Lords in South Australian Asset Management Corporation v York Montague Ltd [1997] A.C. 191 (“SAAMCO”), which mandates a different approach to the assessment of the scope of a professional’s duty of care and whether the loss suffered is attributable in law to the negligence in question.

The approach in SAAMCO has recently been considered and applied by the Court of Appeal in Manchester Building Society v Grant Thornton UK LLP [2019] EWCA Civ 40; [2019] 1 WLR 4610, and by the Supreme Court in Hughes-Holland v BPE Solicitors [2017] UKSC 21; [2018] A.C. 599.

As a result of those decisions, PwC submits that the first question is whether it was retained to provide information or advice on the basis of which AWA would decide whether to declare dividends (an “information case”), or whether it was retained to advise AWA’s directors on what decision they should take (an “advice case”): see SAAMCO at p.214, per Lord Hoffmann, and Manchester Building Society at [49][53], per Hamblen LJ. If it is an information case, PwC is not responsible for all the consequences of the dividends being paid but only for the foreseeable consequences of its audit opinion or advice being wrong. The relevant question is: what loss flowed from the fact that the information or advice was wrong, not what loss flowed from approving the accounts and declaring the dividends. The way to identify what loss flowed from the incorrectness of the information or advice is to ascertain what loss would have been suffered in any event, regardless of the correctness or incorrectness of the information. This is expressed by Lord Hoffmann as being “losses which would have occurred even if the information which he gave had been correct …”.

It is generally accepted (and BTI accepts in principle) that audit work carried out by an accountant is an “information case”, not an “advice case”. This is so even though the work that an auditor does will involve advice about the content of the accounts (see per Lord Sumption in the Hughes-Holland case at [39]-[41] on the correct distinction between the categories). It is not an advice case because the auditor is not advising the company whether or not to approve the accounts and enter into a transaction in reliance on them. The auditor provides opinions, information and advice about the propriety of the accounts that the directors prepared. BTI in fact contends that PwC should be taken to have assumed responsibility for the dividends paid, by reason of its proximity to the decision to distribute almost all of AWA’s reserves to Sequana and then sell the company, but it is not necessary for me to consider that part of its case further.

On the assumption that this is an “information” case, PwC argues that it is not liable for losses that would have been incurred even if its audit opinion had been correct. On that basis, it submits that if the accounts in question had indeed shown a true and fair view then the dividend would have been declared in any event, and accordingly

In my judgment, this reasoning, while apparently consistent with Lord Hoffmann’s explanation in SAAMCO, misapplies the test that SAAMCO provides.

In SAAMCO itself, the negligent surveyors over-valued the security. As a result, the lenders lent money but the security was worth less than the surveyors said. The surveyors were only liable to the extent of the difference between their valuation and the true value of the property. Any losses over and above that would still have been suffered, by reason of a fall in the market, if the security had indeed been worth the surveyors’ valuation. Lord Hoffmann’s reference to “losses which would have occurred even if the information which he gave had been correct …” makes perfect sense in that context. If the valuation had been correct, the lenders would still have suffered substantial losses.

How does that translate into a claim that a company suffered loss by paying dividends on the basis of accounts approved after a negligent audit report? PwC submits that the relevant comparison is between the dividend that was paid in fact and the dividend that would have been paid if the audit report had been correct. Self-evidently, says Mr Salzedo, the same result would follow, because the directors would do what they in fact did, as the incorrect audit supported the same figures in the accounts. But that will always be the result and negligent auditors can never be liable in such circumstances if the focus is on the question of whether the dividend would have been paid. That result seems intuitively wrong if, in the event that the audit report had been non-negligent, the company would have accounted for higher liabilities and so would have had lower distributable reserves. If, in such circumstances, there were still sufficient reserves and the dividend would still have been paid then no relevant loss has been suffered; but if accounting for higher liabilities would have reduced the reserves, so that only a lower dividend would have been paid, then the difference between the higher and the lower dividend ought to be recoverable as loss referable to the negligent audit.

Mr Salzedo submits that that is the incorrect counterfactual when considering whether losses fall within the scope of the duty. He says that the correct counterfactual is simply that the information provided is correct, not that different information was given. As to that, the SAAMCO principle would be directly applicable if the loss that AWA suffered, in paying the dividends, was partly attributable to PwC’s negligence and partly to some other cause for which PwC bore no responsibility, e.g. a negligently prepared (but apparently reliable) consultant’s report on the extent of indemnity liabilities. To the extent that that loss would have been suffered in any event, even if PwC’s report had been correct, PwC cannot be liable for that loss. But that is not the situation that arises in this case. Either the Directors would have proceeded with the dividends in any event, in which case PwC’s breach of duty is not a cause of any loss, or the only relevant cause of loss that AWA suffered was PwC’s breach. There is no need to apportion the loss that AWA suffered between matters for which PwC bore responsibility and other causes, and no need for the SAAMCO filter to be applied.

Even if one were to apply Mr Salzedo’s counterfactual approach, the position would then have been not only that the dividend would have been paid but also that AWA

The argument on scope of duty that PwC now advances is novel. It is contrary to previous decisions of this court (the two 19^th Century cases referred to above; Sasea v KPMG [1999] BCC 857 at 869 and (obiter) Assetco plc v Grant Thornton UK LLP [2019] EWHC 150 (Comm) at [967]-[969]) in which dividends paid as a result of negligent audit reports have been held (or stated) to be recoverable in full. In Equitable Life v Ernst & Young [2003] PNLR 23 at [88]; [2004] PNLR 16 at [96], Langley J and the Court of Appeal held that this principle was not confined to a case where a dividend was unlawfully paid, as long as it would not have been paid had the report been non-negligent.

PwC’s argument is also contrary to learned commentary on this area of the law, including Mr Salzedo’s and Mr Singla’s own textbook, Accountants’ Negligence and Liability, in which the authors comment that the observation of Collins J as to the recoverability of an unlawful dividend in the Sasea case was plainly right; state that the strike out of the claim in the Galoo case was justified only on a pleading point; and having cited the observations of Langley J in the Equitable Life case comment:

“It is thus established in English law that it is at least arguable that a company which was at the relevant time both solvent and in possession of distributable profits may reclaim a dividend from a negligent auditor to the extent that it can establish that it would not have paid the same dividend if had known the true position of its finances.”

The fact that Mr Salzedo and Mr Singla have expressed this considered view in their book does not of course mean that it is necessarily right, much less that PwC is prevented from arguing the contrary. It does however neatly illustrate that the argument now advanced is contrary to the general understanding of the law as well as being open to doubt as to its applicability and correctness, for reasons that I have already mentioned.

This is a notoriously difficult area of the law. Despite the clarification of the principles by the Supreme Court in the Hughes-Holland case, it is the application of those principles to different facts that can be difficult. In those circumstances, it would be wrong, in my judgment, to decide the issue of law as a general proposition, on a summary basis, without having the benefit of clear factual findings in the context of which the arguments and their consequences can be properly tested. That is particularly so given the very close connection that there appears to be between the assumed negligence in giving a clear audit report and the payment of the May dividend. The scope of PwC’s duty and the extent to which losses are referable to breaches of that duty need to be carefully examined in this case in the light of the true facts.

As previously noted, BTI claims the whole of the December and May dividends by way of damages.

PwC contends that that basis of claim is wrong in principle, in that sums of money equal to the dividends were not paid out by AWA but were set off against the receivable in AWA’s accounts and the debt in Sequana’s accounts. As a matter of principle, therefore, AWA cannot have lost more than the value of an acknowledged debt equal in amount to the dividends; indeed, no more than the difference between the value of such a debt and the value (whatever that might be) of a disputed claim against Sequana to recover the same amount of money. That, however, is something of a pleading point, so far as BTI’s claim for damages is concerned, and PwC eschews reliance on pleading points at this stage.

The more substantial point, as it emerged in the course of argument, is that PwC contends that the value of the debt falls to be assessed at the date of trial. At the date of trial, Sequana was insolvent with no prospect of any payment to unsecured creditors. With the benefit of hindsight, what AWA lost is nothing, since at trial a claim for repayment is worth as much as an unpaid debt and in any event Sequana in all likelihood has no distributable assets for unsecured creditors. PwC points out that BTI has not pleaded that the debt would have been repaid by Sequana prior to its insolvency.

BTI disputes the correctness of that analysis. It contends that the value of the debt – if that is the correct characterisation of its loss – falls to be assessed at the date or dates on which the debt ceased to exist, namely December 2008 and May 2009. It further contends that any plea about what would have happened to the debt is only necessary and appropriate in response to any specifically pleaded defence of PwC.

The authority on which PwC relies for the proposition that the debt is to be valued at the date of trial is Golden Strait Corpn v Nippon Yusen Kubishika Kaisha (The Golden Victory) [2007] 2 AC 353. Charterers repudiated a time charter 15 months before the outbreak of a war that would have entitled them to terminate the charter lawfully. The owners accepted the repudiatory breach. The issue was whether damages were payable only for the period up to the outbreak of the war or up to the (later) end of the charter period. The answer given by the arbitrator was: up to the outbreak of war. He found that at the time of the repudiatory breach the outbreak of war was not probable but only a possibility.

The House of Lords held, by a bare majority, that the principle that damages should be assessed at the date of breach was not inflexible and that the damages awarded should be no more than the value of the contractual benefits of which the owners had been deprived. The assessment of damages should reflect the possibility at the breach date that subsequent events might reduce those benefits; and where such events had in fact occurred before the damages were assessed a court should take account of the reality, so that it did not have to estimate the chance of the events occurring. Accordingly, no damages were recoverable for the period after the outbreak of war.

With the benefit of hindsight, the value of the acknowledged debt that AWA lost depended on the prospect of its being repaid by Sequana, either in whole or in part, before the insolvency of Sequana in 2019. As at the dates of breach, the debt (or part of it) would had very substantial value since, ex hypothesi on BTI’s case, the Directors (who are appointed by Sequana as sole shareholder) would have decided to leave that asset in AWA so that AWA could meet its liabilities. (If PwC establish at

In my judgment, in these circumstances, The Golden Victory does not stand for the proposition that the debt that was lost must be valued at the trial date. Rather, the debt must be valued at the date of breach but taking into account the prospects of Sequana becoming insolvent before repayment or otherwise refusing to repay it. That requires an assessment to be made of when repayment would most likely have been sought by AWA and of whether it would have been made by Sequana. Hindsight enables one to say that it would have had to have happened before 2019. It is plainly not possible to say, on a strike out application, that that assessment will produce a nil sum. On the contrary, since the debt was retained to meet likely liabilities of AWA and since those liabilities arose from 2010 onwards, there must be, in retrospect, a real possibility at least that repayment would have been made by Sequana, in whole or in part.

I am satisfied that the right answer is not that AWA suffered no loss. The loss currently pleaded by BTI may not be the correct measure of loss, but the claim should not be struck out on the basis of a pleading point of that kind. It is really PwC’s case - which it will now plead - that any breach of its duty to AWA caused no loss, or alternatively only caused loss less in amount and/or to be assessed on a different basis from that alleged by BTI.

For all the reasons given above, I therefore dismiss PwC’s application to strike out the claim form and for summary judgment.