Some, if not quite born trustees, are appointed as such at the outset of a trust. Some achieve trusteeship at some later stage. And some have some aspects of trusteeship thrust upon them.

Within this third category are strangers to a trust who “dishonestly assist” an express trustee in a breach of the trustee’s fiduciary duty. Through this dishonest assistance, the stranger will be liable to the injured beneficiary, even though no fiduciary relationship exists between them. Although not sued as fiduciaries, such strangers can be held liable to account in equity as if they were a trustee of the beneficiary. Commonly, for convenience (which more often leads to confusion), the stranger is called a “constructive trustee”.

Previously, there was some uncertainty as to the scope of the remedies available for dishonest assistance: specifically, whether the claimant-beneficiary could obtain an account of profits against the dishonest assister, even though no loss was suffered. The unanimous decision of the Court of Appeal in Novoship (UK) Limited & ors v Nikitin & ors [2014] EWCA Civ 908 confirms the availability of the remedy in claims against third parties for dishonest assistance and also the circumstances in which the remedy will be available, namely where there is a sufficient causal connection between the dishonest assistance and the profit and where it would be not be disproportionate to grant the remedy. Robert Hunter and Tom Wood consider the decision below.

robert.hunter@hsf.com_LGSRobert Hunter
Partner, Head of Fraud
+44 20 7466 2426
tom.wood@hsf.com_LGSTom Wood
+44 20 7466 2609


The appeal stemmed from a classic case of fraud carried out by an agent, charged with the responsibility of negotiating contracts for the benefit of his principal.

Mr Mikhaylyuk (M) was a former General and Commercial Manager of the first respondent, Novoship (UK) Limited (NOUK). One of M’s responsibilities was to negotiate the charter of vessels owned by companies in the NOUK group. In breach of his fiduciary duties to the NOUK companies, M orchestrated a series of dishonest schemes, whereby bribes were paid to him in connection with the chartering of the NOUK companies’ vessels.

In one such scheme, M directed a third party to pay substantial secret commissions both to himself and the first appellant (Amon) a company owned and controlled by the second appellant (N) (the Scheme).This was plainly a serious and dishonest breach of M’s duties as director and agent of NOUK and its ship-owning group companies. Clarke J at first instance concluded that N knew the commissions were coming to Amon (his company) at M’s direction and that the likely reason for the payments to Amon was that N had provided some benefit or advantage to M (or might provide some such benefit in the future). All in all, M and N were parties to a dishonest arrangement at the expense of M’s principals.

The Henriot Charters

At the same time as carrying out the Scheme, M and N also negotiated long-term charters of NOUK vessels to the third appellant, Henriot, another company owned and controlled by N (the Henriot Charters, also incidentally the subject of the long-running Fiona Trust litigation). The Henriot Charters were at commercial rates and not – in and of themselves – disadvantageous to the NOUK companies. However, because M was negotiating the Henriot Charters with N at the same time as carrying out the Scheme (which involved secret payments for the benefit of N), Clarke J concluded there was a realistic possibility of a conflict between M’s duty of loyalty to NOUK and M’s personal interest in favouring N in negotiating the Henriot Charters. M’s failure to seek NOUK’s consent – fully informed with knowledge of what was occurring under the Scheme – to negotiate the Henriot Charters with N was a further breach of M’s fiduciary duties to the NOUK companies.

As for N, Clarke J held that he was liable for dishonestly assisting M’s breach. As a third party benefiting under a fraudulent scheme against NOUK orchestrated by its agent, N must have known of M’s conflict of interest and that M could not be trusted to act with undivided loyalty to his principal in negotiating the Henriot Charters. N was “continuing a relationship which was corrupt at inception and had not been cleansed”.

Grounds of appeal: availability of account of profits

There were 12 grounds in the appeal, including several concerned with the judge’s findings of fact in relation to M’s breaches of duty and N’s dishonest assistance (all of which were dismissed by the Court of Appeal).

The most notable ground, however, concerned the scope of the claimant’s remedy. For his role in dishonestly assisting M’s breaches in respect of the Henriot Charters, Clarke J had held N personally liable to account to the NOUK companies for the profit of around $109 million that he and Henriot had made from the Henriot Charters. The respondents appealed on the question of whether an account of profits was available as a remedy for dishonest assistance. If the remedy was available, further issues were:

  • whether some causal connection was needed between N’s dishonest assistance and the profits which NOUK was seeking to claim; and
  • whether, in the circumstances, it would be right to order an account of profits in the case.

The Court of Appeal’s decision

The appellants’ first argument was that an account of profits should not be available against a third party who had never voluntarily assumed fiduciary obligations to the claimant: a dishonest assister is not a fiduciary, and has not promised expressly or inferentially to subordinate his interests to those of a principal. N sought to rely on the recent Supreme Court decision in Williams (Respondent) v Central Bank of Nigeria (Appellant) [2014] UKSC 10 (see our blog post on the decision), in which the court held that dishonest assisters and knowing recipients should be treated in line with common law fraudsters (rather than with defaulting trustees) for the purposes of limitation. Similarly, N argued, as the common law does not strip a fraudster of his gains, the remedy for dishonest assistance should be limited to what is necessary to compensate the claimant for his loss.

The Court of Appeal unanimously rejected this submission. Judicial and academic authority supported the view that an account of profits was available as a result of the dishonest assister being “accountable in equity” and “liable to account as a constructive trustee”. These expressions could not mean that the extent of the remedy should be limited to the principal’s losses. If a dishonest assister is liable to account “as if he were” a trustee, then it followed that he must be liable both (i) to make good the principal’s losses; and (ii) to give an account of profits (Cook v Deeks [1916] 1 AC 554 approved).

This conclusion was justified as a matter of policy, both in order to deter dishonest third parties from compromising the high standards of conduct expected from fiduciaries, and simply as a matter of equity. If a defaulting fiduciary should have to disgorge the fruits of his fraud, why not also a third-party accessory to that fraud?

Further, and mirroring the decision in respect of limitation in Williams v Central Bank of Nigeria, there was no reason to differentiate between knowing receipt and dishonest assistance for the purposes of deciding whether an account of profits was an available remedy. Relief was “founded on fraud” and ought not to turn upon whether or not the third-party accessory had received trust property (although it may of course be relevant to whether or not a proprietary claim is available).

A causal connection required

As for causation, the position of a dishonest assister was different to that of a fiduciary. A fiduciary’s liability to account for a profit does not depend on causation at all; it is sufficient that the profit received falls within the scope of the fiduciary’s duty of loyalty. Where a dishonest assister was concerned, however, in the absence of a duty of loyalty owed to the claimant-beneficiary, the common law rules of causation, remoteness and measure of damages should be applied by analogy to determine the scope of liability. This meant that a simple “but for” test of causation was not appropriate: there needed to be a sufficiently direct causal connection between the defendant’s dishonest assistance and the profit made.

In N’s case, the Court of Appeal held that the causal connection was insufficient. While it was true that “but for” entering into the Henriot Charters, N would not have made a profit, the “real or effective cause” of N’s profit was an unexpected change in the market which worked in his favour, as he had locked in favourable long-term charters against rising rates for freight. N had made a profit by playing the market, not by dishonest assistance.

Discretion where the remedy is “disproportionate”

Finally, the Court of Appeal noted that even if a causal connection had been made out, the decision whether or not to award an account of profits against a non-fiduciary was ultimately a matter for the court’s discretion (with reference to Attorney General v Blake [2001] 1 AC 268). An order to account was not an automatic consequence, in the way that it would be for a fiduciary. The remedy could therefore be withheld on the grounds that it would be disproportionate, as was appropriate in the case of N and Henriot.


The decision provides clarity on what was an uncertain issue. There was a growing body of case law at first instance recognising the court’s power to order an account of profits, even where the claimant-beneficiary had suffered no loss (see for example Ultraframe (UK) Limited v Fielding [2005] EWHC 1638 (Ch)), and the principle is supported in key textbooks, including Snell’s and Underhill and Hayton. Until the decision in Novoship, however, there was no Court of Appeal authority on the point and, given that the judgment confirmed the court’s general discretion to refuse the remedy in appropriate circumstances, the issue is unlikely to be referred to the Supreme Court any time soon. More generally, the decision is the latest in a line of authority unifying the principles governing knowing receipt and dishonest assistance (the previous being Williams v Central Bank of Nigeria).

There does seem to be some tension between the basis on which a dishonest assister is liable and availability of the remedy of an account of profits. A fiduciary’s liability to account for a secret profit arises on the basis that he owes a duty of loyalty to his beneficiary which compels him to account for the profit. It is his failure to perform this duty – and account to his beneficiary immediately upon receiving a bribe or commission – that gives rise to the remedy: equity enforces performance of the duty by making the trustee pay-over his ill-gotten gain. By contrast, a dishonest assister (or knowing recipient), never owes such a duty. Their liability to account only arises as a result of their dishonest involvement in another’s breach of duty. Liability is therefore founded on a wrong, rather than a duty. To some extent, this gap between a positive duty to account and the liability to pay it over is bridged by the requirement for a causal connection between the dishonest assistance and the profit, but this really only serves to link the defendant’s actions with the remedy. It does not provide a substitute for a duty which the defendant never owed. As the Court of Appeal concluded, however, the availability of the remedy is consistent with the overarching principle that dishonest assisters/knowing recipients should be held liable to account “as though they were trustees” (see Dubai Aluminium Co Ltd. v. Salaam [2002] UKHL 48).

It will be interesting to see how the requirement of causation is developed in relation to the remedy of accounts of profits. In claims for equitable compensation against dishonest assisters, the starting point is to identify what losses have resulted from the breach of trust which has been dishonestly assisted; the court will not concern itself with the precise causative significance of the dishonest assistance (see Grupo Torras SA [2001] CLC 221). That said, it is unlikely that the assister would be held liable for losses caused by a breach of trust which occurred before the assister became involved (for example, if the dishonest assistance was limited to attempts to cover up a prior breach of trust). Similarly, with a claim for profits, it would follow that the “sufficient causal connection” requirement will preclude a claim for profits made before all the elements sufficient to fix the defendant with accessorial liability have arisen (most often, a dishonest state of mind will be the missing ingredient).

As a final point, if relief really is “founded on fraud”, it is not clear as a matter of policy why the common law fraudster should be in a better position for the purposes of relief than a dishonest assister. Punishing a third party accessory seems a rather indirect way to uphold the high standards expected of fiduciaries (far more effective to this end is the Supreme Court’s recent decision in FHR European Ventures LLP and ors v Cedar Capital Partners LLC [2014] UKSC 45 regarding bribes received by fiduciaries). The court’s general discretion to withhold the remedy of account should help to ensure that it is not used oppressively in dishonest assistance cases. The confirmation of its availability, however, should serve as a strong deterrent to would-be fraudsters contemplating schemes with or involving fiduciaries.