CL-2018-000297, CL-2018-000404, CL-2018-000590, - [2025] EWHC 2364 (Comm)

There was therefore room for the possibility that Mr Shah might have been, if anything, better placed as a result of his relative isolation from the forensic process for two years to give a reasonably unvarnished account, as best he could, of what he perceives to be recollection of contemporaneous events and thought processes. Even then, of course, it would have been testimony liable to have been rendered unreliable by the passage of time and Mr Shah’s more active involvement in litigating the case prior to May 2022.

Regrettably, but without any hesitation, I concluded that by his voluminous and detailed witness statements, adopted as his evidence in chief, and in his cross-examination, Sanjay Shah did not attempt to give an unvarnished account, as best he can, of what he thinks he can remember. In my judgment, his evidence of ‘fact’ on contentious issues was for the most part argument, taking and attempting to defend positions he thought helpful to his defence, or clever, or both. One of Mr Shah’s shortcomings, in my judgment, is a smugness about what he did that is closely allied to a belief that he is a lot cleverer than he is.

His testimony, as a result, was riddled not so much with the ordinary errors of unreliable recollection that one might expect, although no doubt there were such errors too, but rather more with implausible claims and obvious lies. I do not consider it safe, in general, to treat anything Mr Shah says for himself or about the Danish dividend tax refund factory he created as reliable evidence of fact. So far as contested points of fact are concerned, his lengthy witness statements are in my view largely worthless, and his oral evidence on numerous occasions, during a long cross-examination, consisted of saying what it occurred to him to say to try to dodge or wriggle out of possible difficulties or giving argumentative answers he thought, in the moment, might best serve a line of defence. The primary attitude was to deny whatever SKAT alleged, because SKAT alleged it, and provide attempted justifications for the denial, rather than to give simple factual recollection (if he had any). The result was a large number of attempted explanations of contemporaneous events that were mostly unsatisfactory, often nonsensical, incoherent or inconsistent.

Most starkly – and I linger on this now because it was not only a major theme in Sanjay Shah’s testimony, but a pervasive feature of the case on the defence side – the Danish dividend tax refund business created by Sanjay Shah and those working for or with him (and likewise the Maple Point and Klar versions in which Sanjay Shah had no involvement) involved extensive and deliberate pretence, obfuscation and collateral dishonesty. However, instead of accepting that, Mr Shah made and maintained, for the most part, implausible claims that everything was done transparently and in honest fashion such that there was no behaviour that might be considered questionable and nothing that might call for explanation.

Behaviour like that, it has been said, is ‘a badge of fraud’, but generalities of that kind are no substitute for a balanced decision, on the whole of the evidence, as to whether any particular cause of action asserted by a claimant has been made out. The stand-out feature of the share trading activity in this case is that it did not involve meaningful independent decision-making by the participating entities, apart from an initial decision whether or not to sign up to the scheme. It was trading with the sole commercial purpose of generating a Danish dividend tax refund claim to be submitted on behalf of the tax advantaged buyer, supported by a CAN issued to it on the basis of the trading, where the exact terms of every transaction entered into were in reality set for the parties by those running the scheme. That I consider to be what Sanjay Shah wanted to keep hidden, and engaged in dishonesty to hide. He, likewise Mr Horn and Rajen Shah, thought that SKAT would or might well dispute claims, if aware of those full facts, and if SKAT had taken the view that claims based on contrived trading of that kind would not be paid, and had queried or scrutinised claims on that basis, rejected claims would not have been pressed (which, in practice, would have required a case to be taken to a Danish Tax Tribunal). Those scheme architects also, in my judgment, envisaged that the FCA (as regulator of the custodians and brokers involved) might disapprove of the activity, and were willing to be untruthful about what it involved in order to avoid that disapproval.

However, in the event SKAT did not allege that the contrived nature of the trading, in itself, rendered the resulting dividend tax refund claims invalid, or otherwise gave rise to any cause of action. It was relied on in support of SKAT’s allegation that the trading was entirely sham, i.e. that the parties were not trading at all, rather they were all just pretending to trade as a pretext for the generation of a CAN whereby to dupe SKAT into making payments. It was also said to undermine claims by some of the defendants that they were not aware at the time that the trading scheme involved share-less settlement loops. The sham trading allegation, in my judgment, encouraged Sanjay Shah (and some others) to adopt and maintain indefensible positions that the trading was not contrived, or that there was not, as I have just described it, pretence, obfuscation and collateral dishonesty designed to conceal the contrivance from SKAT. It does Mr Shah (and those others) no credit to have responded in that way; but that does not prove SKAT’s case, although of course I took it into account as capable of lending some support to it.