What a carry-back!

When working on the Tax Law Rewrite Project at the turn of the millennium, I was told that there had been a recent delegation to the Inland Revenue from the Japanese tax authority. According to the story I heard, the Japanese were asked how long it took for their country to get used to their self-assessment regime. On hearing that it took about ‘four to five years’, the UK hosts expressed considerable relief (given that Self Assessment had been introduced in the UK in April 1996). However, one of the Japanese visitors then clarified his answer: he had said ‘45 years’.

As we are now in the third decade of Self Assessment in the UK, it is possible that the Japanese experience is being repeated here. Of course, much of Self Assessment is working as intended and without too many hiccups. However, it appears that difficulties are cropping up in cases where something slightly out of the ordinary is happening.  

In this context, ‘out of the ordinary’ does not necessarily mean exceptional or unusual. For example, it is widely known that opening an enquiry into a partnership (Taxes Management Act 1970 s 12AC) also deems there to be an enquiry into each partner of that partnership (s 12AC(6)). However, there is still no clarity as to the scope of such a deemed enquiry; indeed, this is something that the Upper Tribunal is due to consider later this month.

Another situation which keeps returning to the courts is the process by which taxpayers carry back relief from one tax year to another. That has already seen three Supreme Court decisions in the past six years, yet the matter is still not going away, with it being recently reconsidered by the Court of Appeal in the case of Knibbs and others v HMRC [2019] EWCA Civ 1719.

The facts of the case

Stated at its simplest, each of the taxpayers claimed to have sustained losses in one tax year (year 2), some or all of which they carried back to an earlier tax year (year 1).  

The taxpayers consider that those carry-back claims did not form part of the tax return for year 2; and therefore, that any HMRC challenge to those claims should have been effected by way of enquiry under TMA 1970 Sch 1A, rather than under the normal enquiry procedures found in TMA 1970 s 9A. (The two different enquiry procedures are more or less identical in practice. However, they operate in parallel: s 9A for enquiries into tax returns themselves, including any claims made within the tax return; and Sch 1A for any claims made outside a tax return.) However, no Sch 1A enquiry was opened into the claims. As a result, the taxpayers consider that those carry-back claims must now be treated as final.

On the other hand, HMRC consider that they have the right to revisit these claims through the s 9A procedure, provided that they open such an enquiry into the taxpayers’ tax returns for year 2. This is because the losses claimed related to year 2.

The point has already been determined in favour of HMRC by the Supreme Court in R (oao De Silva) v HMRC [2017] UKSC 74. However, a number of doubts have been expressed about the correctness of the Supreme Court’s conclusion. The taxpayers therefore wanted an authoritative statement confirming the position. Accordingly, they made a claim direct to the High Court seeking a declaration as to the correct legal position.  

Although there is a defined procedure for pure questions of law to be determined by the High Court, HMRC argued that it was not the appropriate forum for the taxpayers to have used in this case. They therefore asked the court to strike out the claims as an ‘abuse of the court’s process’.  

The High Court agreed with HMRC both on the question of abuse and also on the substantive question; the taxpayers appealed against that decision to the Court of Appeal.

The court’s decision

The case came before Lord Justices David Richards, Henderson and Moylan who gave a joint judgment.

Dealing with the abuse point first, the court noted that the starting point is that TMA 1970 contains a detailed code for challenging HMRC decisions (the appeals process). Accordingly, taxpayers trying to embark upon a different litigation strategy (such as the High Court claims in the present case) would be abusing the court’s process. The taxpayers had tried to argue that there were exceptional circumstances permitting the court to depart from this conclusion – being that some of the taxpayers were now out of time to challenge closure notices issued by HMRC in relation to the s 9A enquiries into year 2. However, the court said that this fact emphasised the importance of following the statutory scheme and did not justify any departure therefrom.

In response to the taxpayers’ arguments that the s 9A enquiries are themselves invalid, the court made the point that the taxpayers can raise that argument in the course of any appeal against the subsequent closure notices.  

The taxpayers also argued that they needed to commence proceedings so as to avoid the risk of their challenges becoming time-barred under the Limitation Act 1980. However, the court dismissed this reason as well. The first reason given was that if tribunal proceedings led to HMRC learning that they had misunderstood the effects of a closure notice, it would be ‘startling … that … HMRC would decline to give effect to the FTT’s decision and refuse to pay the tax that HMRC had no right to retain’. The second reason given by the court was that, in any event, time would not start to run until the First-tier Tribunal’s determination of 
the appeal.

For these reasons, the court considered that the taxpayers’ claims had to be struck out as an abuse of the process of the court.

However, the court then proceeded to consider the substantive question as to whether the carry-backs to year 1 must be challenged through Sch 1A or whether HMRC can instead use the s 9A enquiry procedure into the returns for year 2. The argument focused on the fact that (for some of the taxpayers at least) the cases concerned the years covered by the Income Tax Act 2007, where there was a more prescriptive set of rules determining the correct tax liability for a tax year, whereas the De Silva case concerned the pre-2007 code. However, the court considered that there was nothing in the rewrite process (that led to the 2007 Act) which would suggest that there was any intention to change the law from that, as since explained by the Supreme Court, in De Silva.

Furthermore, the court concluded that the De Silva approach remained correct and dismissed the taxpayers’ appeals.

Commentary 

In the course of their arguments, the taxpayers had identified a line of case law which showed that citizens do not need to take proactive action to challenge certain decisions by public authorities (i.e. by commencing judicial review). In short, the case law clearly demonstrates that (in the absence of any statutory appeals process) citizens are entitled to await enforcement action being commenced by the public authority and then make a collateral challenge against the public authority’s decision in the course of their defence. The court considered that this line of cases was not relevant to the present cases. I am tempted to agree with the court on this point.

However, the court also added that none of those other cases had been in the context of tax. And that statement is wrong. The most obvious example of a tax case following that line of authorities is Pawlowski (Collector of Taxes) v Dunnington [1999] STC 550. (This concerns the transfer of an unpaid PAYE liability from employer to employee under what is now regulation 81 of the PAYE regulations – that case being heard at a time when the taxpayer could not appeal against such a decision.) However, the same point is also implicit in at least three cases concerning information notices (now of the sort given under the Finance Act 2008 Sch 36 where the tribunal has pre-authorised the notices), most notably Kempton v Special Commissioners and IRC [1992] STC 823.

So far as the substantive point is concerned, it is hardly surprising that the court followed the Supreme Court’s decision in De Silva, as it would not have been permitted to depart from it unless it was convinced that the Supreme Court had got it wrong. Nevertheless, I was surprised that the Court of Appeal was so emphatic in the way that it upheld the De Silva approach. In particular, there still remain a number of questions as to the route taken by the courts to get to the current view of the law.

These questions boil down to just two words found in TMA 1970 Sch 1B. Schedule 1B was specifically written to deal with cases where claims concern two different tax years. In particular, para 2 deals with cases (such as the present) where relief arises in one tax year and is then carried back to an earlier tax year. Paragraph 2(6) explains how such relief should be given: ‘Effect shall be given to the claim in relation to the later year, whether by repayment or set-off, or by an increase in the aggregate amount given by s 59B(1)(b) of this Act [effectively, by treating the tax relief as if it were a payment on account for the later year], or otherwise.’

What was suggested by the Supreme Court in De Silva was that these words ‘or otherwise’ envisaged the claims being effected through an adjustment to the self-assessed tax liability for the later year. The Supreme Court then proceeded to say that these words also meant that HMRC were then fully entitled to amend a taxpayer’s Self Assessment so as to correct any relief wrongly claimed.

In the present case, the Court of Appeal expressed surprise that, on the taxpayers’ argument, the tax relief claimed need not feature in the tax return for the later year, given that the relief is statutorily stated to ‘relate to’ that later year. However, it is my view that this is not in fact so surprising. First, the words ‘relate to’ have a clear statutory purpose and this is to determine the time limits in Sch 1A for opening any enquiry into the claim. Secondly, that link to Sch 1A (which axiomatically concerns claims made outside a tax return) reinforces the view that Parliament was not actually expecting the claims to feature on the tax return at all. Thirdly, if inclusion on the tax return were envisaged, there would be no need to give effect to the claim as a deemed payment on account. Fourthly, one might have thought that such route to relief would have been stated slightly more explicitly than merely in a sweep-up phrase tagged at the end of a sentence. Indeed, it has long been HMRC’s published view that those two words had no real meaning.

However, leaving aside these objections, there is yet a further reason why I am still not convinced by the court’s conclusion. Paragraph 2(6) explains how effect is given to a claim made by a taxpayer. In other words, when a taxpayer makes such a claim, how does the taxpayer get the benefit of the tax relief being claimed? I am prepared to accept that the words ‘or otherwise’ are wide enough so as to allow the taxpayer to give effect to the claim through the Self Assessment machinery for year 2 (or arguably for any other year). But if a taxpayer has not actually claimed the relief in that way, it is my view that para 2(6) becomes irrelevant. That is because para 2(6) does not concern how HMRC may correct claims made by taxpayers. HMRC’s methods of correction are dealt with in the various provisions concerning the conclusion of enquiries, etc. Furthermore, if a taxpayer has not claimed relief in a return, it cannot be possible for the return to be ‘corrected’ by the removal or amendment of the claim from or within the return.

In short, I remain unconvinced about the views expressed so far by the courts and hope that the matter can be revisited by the Supreme Court.

What to do next

The court’s decision on abuse of process serves as a reminder of the need to follow the correct procedures when challenging decisions taken by HMRC. As the court noted, ‘both appeals to the FTT and applications for permission to pursue judicial review are subject to short time limits [and] it makes no sense at all’ that such time limits can be sidestepped by taxpayers seeking to adopt an alternative route of challenge.

As far as the substantive issue is concerned, I suspect that the issue has not gone away for good. However, any taxpayer still wishing to run the argument will have to expect a long trawl through the courts in the hope that the Supreme Court will have an opportunity to reconsider its position as stated in De Silva. Because of the need for permission at each level of appeal, there is a possibility that a case will never get that far.