When tax avoidance schemes fail, professional negligence claims against the scheme promoters will surely follow. And, because HMRC’s investigations of such schemes typically proceed at a ponderous rate, it is likely that limitation defences will be raised, fuelled by the long lapse of time between the advice being given and the failure of the scheme.
Clients in such cases will eventually suffer loss. But when does the loss occur? On entry into the scheme? When the scheme is closed? Or at some other intermediate point? This question is crucial in determining when the limitation period started to run, and often arises as a key factual issue.
These arguments are being played out once again in the case of Evans v PricewaterhouseCoopers LLP, an ongoing matter before the Commercial Court. The claimants were trustees of a trust resident in the UK. In 2000, they wanted to sell shares in a private company held by the trust and were looking for a way to mitigate the capital gains tax liability that would arise on the disposal.
The defendant (PwC) suggested a “round the world” scheme, whereby the trust would become resident in another jurisdiction for part of the tax year in which the disposal would occur. The jurisdiction selected would be one in which no (or little) capital gains tax would be payable. Mauritius was initially suggested (and approved by tax counsel), but PwC recommended that Canada be substituted.
That (the claimants allege) was a fatal error. The UK-Mauritius tax treaty would have allowed the trust to be taxed in Mauritius based on its residency there for part of the year, a crucial feature necessary to escape the UK tax charge. By contrast, the UK-Canada double-tax treaty gave the UK and Canada the authority to choose where the trust would be taxed. As it transpired, in 2013, they agreed that the trust would be taxable in the UK, triggering the full UK capital gains tax charge on the disposal.
In 2016, the claimants issued a claim against PwC in respect of their losses, including the UK capital gains tax payable. PwC responded with a strike out / summary judgment application, arguing that the claimants had no prospect of overcoming the limitation points. After all, the decision to use Canada rather than Mauritius had been taken as far back as 2001.
However, PwC’s application was refused. The judge:
- Accepted submissions that there were (arguably) continuing breaches until around 2011 (when PwC ceased to act).
- Accepted, following Law Society v Sephton & Co, that it was at least arguable that any loss suffered by the claimants was “purely contingent” until the decision was taken to subject to trust to UK tax in 2013.
Accural of cause of action: where are we now?
Evans is one of a number of cases that have grappled with the limitation arguments, and the results are not always consistent. The key difficulties appear to coalesce around two questions (although they are not dealt with in precisely these terms).
First, does the cause of action accrue when the claimant has suffered an objective loss (that is, they are worse off in monetary terms) or is it sufficient for them to show a loss that is purely subjective (that is, while no worse off in monetary terms, they have lost or missed out on something that they were expecting)?
It seems now to be accepted that a subjective loss is sufficient to found the cause of action. This is clearest in Halsall v Champion Consulting, in which HHJ Moulder said:
“… [the claimants] suffered damage, not because the shares were then worth less than the claimants paid for them, but because the purchase of the shares would not give the claimants the result that the defendants had assured the claimants would result, i.e. the tax relief.”
Language referring to a loss or restriction of “commercial options” is also used to express the same idea.
The second issue is, does exposure to an unwanted risk count as a sufficient loss? Here the cases do not speak with one voice. On the one hand are Nykredit Mortgage Bank Plc v Edward Erdman Group Ltd (No.2) and Sephton, where the House of Lords/Supreme Court held that there was no loss until the risk in fact eventuated. Nykredit concerned exposure of a lender to a shortfall in the security for their loan (no loss until the value of the security falls below the sum lent plus applicable interest); Sephton concerned the risk that the Law Society would have to pay compensation out of its central funds (no loss until valid claims for compensation were made).
On the other hand is Shore v Sedgwick Financial Services Ltd, in which the transfer of assets into a pension scheme more volatile than that intended was held to constitute an immediate loss. I am not convinced that this is correct: if the claimant’s pension had increased exponentially in value, would they still have had a complete cause of action? Nonetheless, Shore was followed in Pegasus Management Holdings SCA v Ernst & Young and Halsall.
Evans appears to be on a collision course with this second question; the essence of the complaint against PwC is that they introduced an unwanted element of risk by selecting Canada. It remains to be seen whether the trial judge ultimately feels that this constitutes an immediate loss, or one that arises only if and when the risk eventuates.