Stamp duty land tax (SDLT) avoidance: HMRC win in Supreme Court

On 13 June 2018, the UK Supreme Court delivered its judgment in the Project Blue SDLT case (read our preview of the decision here). The case was decided by majority decision in HMRC’s favour, overturning the previous Court of Appeal decision. It is a huge victory for HMRC in tax terms and a significant one for advisers going forward. Lord Hodge (on behalf of Lady Hale, Lord Hughes and Lord Lloyd-Jones) delivered the majority view, with only Lord Briggs dissenting

Factual Background

The transaction at the heart of the case was the sale of the site of the former Chelsea Barracks in London by the Ministry of Defence to the sovereign wealth fund of the State of Qatar, funded in part by a Quatari bank. The total consideration actually paid was one of the issues in the case but at its highest could have been £1.25 billion, so potential SDLT of some £50 million. And that doesn’t take account of any cases which were waiting for the outcome of this one. The quantum of tax at stake and the parties concerned would probably have been enough to make this case high profile, but it also marked the first time the courts had seriously been asked to consider the potentially wide ranging SDLT anti avoidance legislation. So a keenly awaited decision.

SDLT issues

The transaction in question was structured in such a way that it appeared to string two SDLT relieving provisions together to result in no SDLT being payable. This was achieved by a sale from the Ministry of Defence (‘MOD’) to a ‘Project Blue’ company (‘PBL’) owned by State of Quatar, followed immediately by a subsale of the land to the funding bank. The funding bank leased the land back to PBL and at the same time put/call options were granted over the freehold between the bank and PBL. In theory this meant that the initial transfer to PBL was not taxed due to the rules allowing subsale relief in place at the time and that the further transfer and leaseback were also not taxed due to the exemptions in place to allow Shari’a compliant funding. The first key question was whether this scheme worked.

Following on from the question of whether the scheme worked was the question of if or how the SDLT anti-avoidance legislation would apply.

Did the scheme work or not?

The majority judgement overruled the Court of Appeal’s finding that that the scheme did NOT work. The Court of Appeal had concluded on the basis of previous SDLT case law (HMRC v DV3 RS LP [2013] EWCA Civ 907) that the effect of the subsale provisions was that no land interest moved to PBL. This in turn meant PBL had no interest to transfer on for the purposes of the Shari’a exemption and therefore the structure could not qualify for exemption. So the first technical question before the Supreme Court was whether to follow the DV3 logic on the effect of the subsale.

Readers with long memories and a particularly keen interest in SDLT will perhaps recall that certain commentators (the author included) believed at the time that the reasoning behind the DV3 decision was flawed. Rather than disregarding the entire contract, a better view would have been that the legislation removes the taxing point for this contract. SDLT applies on substantial performance or completion and the wording of the subsale rules was to ignore these taxing points in relation to the original contract. That is not the same as denying the effect of the contract entirely. So this case would have been an excellent opportunity to ‘correct’ DV3. Unfortunately, although there are passages where the Lords come close to re-analysing the old subsale provisions (Lord Briggs comes closest at para 116), they never go so far as to dismiss DV3. As it turned out they didn’t need to. Despite the subsale having the technical effect of not transferring the interest to PBL for SDLT purposes, the Lords found that on an analysis of the Shari’a exemption, it was sufficient that PBL was the bank’s customer and in real terms had the right to subsell the land. So the net effect of these findings was an effective SDLT saving scheme.

Readers who have not been following Project Blue closely would be forgiven for thinking that an effective SDLT scheme is good news for the taxpayer. But in this case the opposite is true. The Court of Appeal had found that the scheme did not work, but this turned out to be good news for the taxpayer because it meant that HMRC were now out of time to go after the funding bank having been ‘relentlessly pursuing the wrong taxpayer’ (to quote counsel for the taxpayer and Lord Briggs at para 129 of the Supreme Court judgment) from close to the outset. It also meant that the anti-avoidance legislation never came into play because there had been no effective scheme. So now that the Supreme Court had reversed the position by finding that there WAS an effective scheme, this opened the gateway to the anti-avoidance legislation.

SDLT anti-avoidance legislation

This is the bit that advisers were waiting for. The first real test of HMRC’s perceived ‘nuclear’ weapon, the SDLT anti-avoidance provisions.

One of the curious things about the SDLT anti-avoidance is that there is no mention in the statute of any motive test. We know it is anti-avoidance because it says that at the heading of the provisions, but that aside there is no mention of motive or purpose. This contrasts with most anti-avoidance where the legislation makes it clear that it’s only supposed to catch you if you’re being naughty. Some observers may have been hoping that the court might take the opportunity to ‘read in’ some type of purpose test (it is anti-avoidance legislation after all) but most (including HMRC) were anticipating that in this respect the Supreme Court was likely to follow the obiter dicta comments of the Court of Appeal in finding that motive/purpose is simply not relevant when applying the provisions. And indeed, this proved largely to be the case. The question of whether this structure was put in place deliberately to avoid SDLT or whether the saving arose fortuitously is never really discussed by the Supreme Court although parts of the dissenting judgement appear to apply that the effect was achieved innocently, notwithstanding the findings of the First Tier Tribunal that PBL had NOT established that it had entered into the Shari’a compliant financing for religious reasons. Either way, motive didn’t matter and the anti-avoidance legislation still applied.

The result of this was that the Supreme Court gave us a guide as to how to apply the untested anti-avoidance. The entire judgement is of interest but the majority of the debate centred around who should be identified as the purchaser for the notional transaction that the anti-avoidance demanded and what amount of consideration was to be taxed. This still mattered a great deal as HMRC had only pursued PBL for the tax, so a finding that the bank had to be seen as the notional purchaser would have constituted a result for the taxpayer. Perhaps unsurprisingly, once the court had engaged the anti-avoidance provisions they found without too much difficulty that PBL was the correct purchaser. The taxpayer was therefore on the hook.

To make matters worse, the amount that should properly be treated as the consideration was the highest possible amount payable. In theory this would have been £1.25 billion, but because that amount was never in fact drawn down, the highest amount actually paid (after currency conversion had been taken into account) was £959m as this was the initial tranche of consideration provided to secure the transfer by the MOD. Still a hefty tax bill.

Other arguments

The court also addressed human rights arguments around whether the effect of the anti-avoidance provisions was to discriminate against Shari’a compliant funding and a line of argument that had come out of the blue (pardon the pun) following an article written by Julian Farrand (partner of Lady Hale) that the entire transaction could/should be recharacterised as a simple mortgage for the purposes of applying the SDLT provisions.

On the human rights arguments, the leading judgment concludes that there is no discrimination and that if there were it could be objectively justified. Also that the anti-avoidance legislation has a built in exemption to prevent it being triggered by a structure which relies purely on the Shari’a compliant SDLT provisions.

On the Farrand article it is clear that neither of the parties argued in support of the property law analysis, so it would have been a stretch for the court to take that line. In fact the judgment prefers the explanation that parliament chose to recognise the separate steps of a Shari’a compliant transaction in putting together the SDLT provisions.

Points to note

As stated above, this is a significant judgement, so there is a lot to take on board. In addition to the aspects discussed above, there are a few points which people are perhaps not yet focusing on closely enough.

DV3 is still good law. It concerns provisions which are no longer in force, and there are parts of the Project Blue Supreme Court judgment which cast some doubt on its reasoning. But overall, cases which settled on the basis of DV3 have not been given grounds for revisiting those settlements. For cases which were awaiting the outcome of Project Blue, the situation may not be clear cut. Many of those will concern similar schemes or schemes which relied on subsale relief and another exemption provision other than Shari’a financing. Those that relied on a different exemption provision may take heart from this decision. After all, it was crucial to the ratio that the Shari’a exemption provisions looked to a ‘real world’ rather than ‘SDLT world’ analysis of who the vendor was. This might be difficult to say of certain other exemptions. So there could be other schemes that were put in place where the conclusion of applying DV3 is that the scheme could not have worked and therefore that if HMRC did not pursue the correct taxpayer, the anti-avoidance may not be available to help them.

There really is no motive test in the SDLT anti-avoidance legislation. This will spread fear amongst taxpayers and extreme caution amongst advisers who will start to be anxious about a mechanical application of the anti-avoidance wherever a multi-stage transaction leads to an SDLT saving compared to a transaction which could have transferred a land interest more directly. The need for HMRC guidance on this is now keener than ever, but if the mechanical application is demanded by the Supreme Court then will the guidance be able to offer much comfort? Certainly the so called ‘safety valve’ (para 77 of the judgment) by which the Treasury can disapply the anti-avoidance (including retrospectively) will not really reassure taxpayers.

One positive answer may come from the clear aim of the court to apply the anti-avoidance ‘purposively’. In their discussions over the identification of the purchaser in particular it can be seen that they are looking to ascertain the aim of the legislation and to apply it accordingly. So perhaps the correct approach to applying the provisions should be to ask whether parliament would have intended the anti-avoidance provisions to apply to a given set of steps. Clues can be found in the exemptions which are built into the anti-avoidance (such as the exemption aimed at transferring property owning companies) but the answers will still not be straightforward and will require experienced advisers.

Finally, there is still frustration about the increasing dichotomy of tax legislation from the real world. This judgment depended on taking a ‘real world’ rather than ‘SDLT world’ approach to the Shari’a financing provisions whilst at the same time noting that parliament had respected the various steps and legislated for the SDLT impact of those. It is a great shame that we have to consider in great detail legislation about transfers of land – ‘real’ property – in those uncertain terms