Dr Stephen Daly

This blog and some published work which has emanated from it has focused upon what happens when HMRC gets it wrong – when the authority advises a taxpayer, but that advice turns out to be wrong for instance whether the advice is in the form of general guidance or in a more bespoke ruling issued directly to the taxpayer. In this blog, I will go one step further back to investigate and provide a theory behind HMRC’s power to get it wrong when it comes to collecting taxes.[1]

The starting point is that taxpayers have an obligation to comply with the tax code, just as citizens have an obligation to comply with the law generally: a citizen should not break the speed limit on a particular road for instance and a taxpayer should not fail to pay taxes that are due under the law. Compliance with the law is the responsibility of the person to whom the law is directed. Managing compliance with the law, on the other hand, will be the responsibility of the State, which in turn will be delegated to different public bodies and officials. Managing law and order is the responsibility of the police, local planning authorities manage planning, and managing compliance with the tax code is the responsibility of HMRC. Where there is a failure to comply with the law, the failure falls prima facie on the citizen (though enforcement in respect of that failure may not occur in the case where the public authority itself has led the citizen into non-compliance). The essential point is that there is a distinction between the responsibility for compliance (being on the taxpayer) and the responsibility for managing compliance (being on HMRC).

That distinction is what ultimately allows HMRC to sometimes “get it wrong”. To explain this, it is necessary to visit the Commissioners for Revenue and Customs Act 2005 which sets out HMRC’s primary responsibilities. Section 5 of the Act provides that HMRC is “responsible for the collection and management” of taxes. How that phrase should be interpreted is a matter of public law. A strict interpretation would lead to an absurdity – namely that HMRC would be regularly and involuntarily acting beyond its powers by failing to ensure that every single penny of tax due is paid (which in practice too is impossible to perfectly quantify). A different interpretation which looks to the purpose of the Act then should be adopted and for that reason the highest court in the UK has very occasionally explained what the purpose of the provision is: it is to allow HMRC to decide upon the “best means of obtaining for the national exchequer from the taxes committed to their charge, the highest net return that is practicable having regard to the staff available to them and the cost of collection” (IRC v National Federation of Self-Employed and Small Businesses Ltd [1982] AC 617, 637 (Lord Diplock)). It seems utilitarian – obtaining the most amount of tax from the most amount of taxing provisions. The courts then should defer to decisions that HMRC makes in respect of the collection of tax – if HMRC decides that it is appropriate to offer advice to taxpayers whether in a generalised form or in a bespoke manner, that is a decision that the Court will generally respect as it helps in the efficient collection of tax (see for instance R (Davies) v HMRC; R (Gaines-Cooper) v HMRC [2010] EWCA Civ 83, (2010) STC 860 (Moses J) at para 12).

But that respect that the courts will give to HMRC decisions is not without limits. HMRC must not use the “collection and management of tax” for an improper purpose. HMRC should take into account relevant considerations, and disregard irrelevant considerations either for instance. Also, HMRC must not act in a way that is unreasonable in the Wednesbury sense. Thus, where HMRC, under the guise of collecting taxes, does something so outrageously stupid or unfair, then the body will be held to have acted beyond its powers (see R v Inland Revenue Commissioners, ex parte Unilever plc [1996] STC 681).

The result is that HMRC may provide advice to taxpayers as this assists in the collection of tax and the courts will generally respect that decision to do so, but the courts will intervene to find that HMRC has not acted in accordance with its responsibilities where it gets the advice “catastrophically wrong” as in the case of Al-Fayed v IRC [2004] ScotCS 112, [2004] STC 1703. There, HMRC had an agreement with several taxpayers that it would not investigate their tax affairs provided that they paid a lump sum in “tax” each year. That advice and agreement was “ultra vires”: it was the antithesis of the collection and management of taxes as it sanctioned future non-compliance with the law. Similarly, HMRC may not get the law so wrong as to provide a concession to a taxpayer which Parliament clearly did not grant (R (Wilkinson) v IRC [2005] UKHL 30, [2006] STC 270 [20] (Lord Hoffmann)). Where HMRC gets the law wrong in its advice and is advised of the error, then HMRC too will have a responsibility to correct this (see: HMRC v Hely-Hutchinson [2017] EWCA Civ 1075, [2017] WLR(D) 517).

But between those two points – where HMRC provides advice to taxpayers and it is not unreasonably incorrect and no other public law constraint applies – then HMRC will be said to be acting within its powers. That its advice to taxpayers is incorrect does not mean that the body has acted in breach of its responsibility to collect and manage tax. The taxpayer who followed the advice may have failed to comply with the law, but HMRC has not failed in its responsibility to manage compliance. Within certain limits then, HMRC has the right to get it wrong.

[1] The issues that arise in respect of HMRC’s other responsibilities such as administering tax credits will not be dealt with here

A new peer-reviewed working paper from Dr Ulrike Spangenberg (Umea University), Prof Ann Mumford (King’s College London) and myself has just been uploaded here. The abstract for the piece reads as follows:

“This paper analyses existing tax competences in EU law, in the light of European and international obligations, and their evolution since the initial founding of the European Economic Community. It approaches this task not just from the perspective of competences, but examines values, objectives, obligations and actual governance capacities underpinning the implementation and realisation of these aims and duties in the field of taxation. The analysis is linked to discourses addressing sustainability gaps within EU taxation law and policy, in particular: a prevailing focus on economic growth; a lack of EU-level environmental taxation; an absence of tax measures that tackle, much less consider inequalities in income and wealth; and, persisting socio-economic inequalities between men and women.

The concept of sustainable development features prominently in the objectives of the European Union and is closely linked to the function of the internal market. The implications of sustainability for legal obligations, however, remain unclear, and particularly so in the context of taxation. This paper provides an outline of the current legal framework for positive and negative integration of taxation, and considers the evolution of values, objectives and obligations in European and International law with a particular focus on the legal concept of sustainable development. From this basis, it examines legislative capacities to address the economic, ecological and social dimensions of sustainable development in the area of taxation and offers preliminary options for the amendment of hard and soft law mechanisms, so as fully to support European values and objectives in the field of taxation.”

The lengthy piece was written as a report for the EU Commission as part of the FairTax project. The FairTax consortium consists of 11 partner universities from 9 countries which is undertaking to complete a series of Work Packages for the European Commission, the ultimate goal of which is to produce recommendations on how fair and sustainable taxation and social policy reforms can increase the economic stability of EU member states.

A college lecturer at the University of Oxford tutors students in a particular course, helping them on their voyage towards the exam (in addition to helping them pick up additional skills also). However, the college lecturer will generally not set the final exam nor mark the final scripts. Very few students will have been taught in a tutorial capacity by either the setters or markers. It could be said then that one of the roles of the college lecturer is to act as exam advisor. The ultimate responsibility for performing well in the exam lies with the student. The student will have legitimate grounds to complain if the advisor is not very good either because the teaching is not particularly clear or that some items ought to be taught which are not or that the student is taught something that in fact is incorrect. But these complaints will not result in a change to the exam mark that the student will eventually obtain. If these complaints were sufficiently serious, it would seem correct that the student should be put in a position the student would have been in but for the poor teaching (though of course, putting that principle into practice is marred by myriad interesting problems which would be the discussion for another day). Fairness would dictate too that a structure of accountability should be in place which could prevent, track and deal with such problems in the teaching. To this end, there are three broad things the student should expect in terms of the “exam advisor” – that the substance of advice or teaching should be of a sufficient standard (“substance”), that there are means for redress where there are failings (“remedies”) and that a structure of accountability exists (“accountability”).

The role of the college lecturer as exam advisor is similar to the role of HMRC when it performs the task of collecting and managing taxes and credits, or essentially managing compliance with the rules. It has not set the rules and is not the ultimate judge of whether the taxpayer has fully complied with the rules. The ultimate responsibility for complying with the rules rests with the taxpayer. HMRC could refuse to engage with taxpayers and let them navigate their way through the tax code, retrospectively challenging the taxpayer on any failings. But HMRC has long recognised that a better way of carrying out its task of managing compliance with the rules is to proactively engage with taxpayers. Taxpayers may approach HMRC individually to obtain advice on whether proposed transactions are in compliance with the rules (i.e. tax rulings). If enough taxpayers approach HMRC, it might think it wise to publish this advice in a generalised form (i.e. HMRC guidance). It may similarly produce guidance even without being approached if it believes that a sufficient number of taxpayers will likely struggle to properly understand the underlying rules and their application.

Just like the college lecturer then, HMRC acts as an advisor. And just like the college lecturer, the same principal issues in respect of substance, remedies and accountability arise. What if the advice is unclear, incorrect or not published? What remedies are in place where HMRC has been a poor advisor? Who holds HMRC to account when it acts as an advisor?

Some of these issues and their potential solutions I have written about here and elsewhere. The point of this blog is simply to bring these issues together under a single framework and to recognise the important role that HMRC plays as tax advisor.

A discussion paper by Tracey Bowler for the Tax Law Review Committee, a Committee of the Institute for Fiscal Studies, was released in November and focuses upon ‘new’ HMRC powers. Vanessa Houlder gives an overview of the paper in a piece for the Financial Times yesterday. Both the report and Houlder’s article are worth reading.

The report makes two important contributions that this post will focus on. The first is that the paper makes clear that the nature of HMRC powers has changed in recent years. The background to the report is that HMRC over the past 10 years has acquired new powers to tackle tax avoidance. But these powers differ from traditional HMRC powers which are directed towards obtaining information and imposing penalties for non-compliance. Instead these new powers seek to nudge taxpayers away from tax avoidance activities and also away from delaying the collection of tax by engaging in litigation. The report cites the Bank Tax Code as heralding a “new approach to HMRC’s powers” accordingly (though the DOTAS regime came in some years prior). The past three years has seen the placing of significant “nudging” powers in the hands of HMRC – for instance, to require the upfront payment of tax due (APNs and PPNs), to require taxpayers to follow judicial decisions or face a penalty for failing to do so (Follower Notices), to penalise taxpayers who fall foul of the GAAR (GAAR penalties), and to penalise ‘serial’ tax avoiders by naming and shaming, issuing fiscal penalties and restricting tax relief (serial avoiders scheme). These are the powers that the Bowler paper focuses on, though one could also examine under this heading of new powers other initiatives such as the DPT and POTAS regime.

Secondly, the paper considers the safeguards in connection with these new powers to be inadequate. It notes that:

“Traditionally, the main independent safeguard to control the power of an administrative body such as HMRC has been the granting of appeal rights to the tribunals and courts. The New Powers have avoided that approach. Although the taxpayer retains the right to appeal the underlying tax assessment, there is no appeal right in relation to the use of the New Powers”

This is an important observation. But what also follows from having broad, intrusive powers without traditional rights of appeal is that the courts will generally apply a restrictive reading of the powers. This has already occurred in the case of APNs and PPNs. Put another way, where legislation does not provide safeguards against powers, the courts will generally intervene to introduce them unless the statute is worded to specifically preclude the courts from doing so. The litigation in this area accordingly is likely to produce interesting assessments of the scope of HMRC’s ‘new’ powers.

On the 12^th of December 2017, the Court of Appeal handed down judgment in the joined cases of Rowe v HMRC and Vital Nut v HMRC, in which dozens of taxpayers challenged the decision of HMRC to issue to them notices requiring the upfront payment of disputed tax – ‘upfront’ in the sense of being prior to a determination of the amount being due by a court or tribunal. These notices requiring the accelerated payment are aptly called ‘Accelerated Payment Notices’ (APNs), though in the specific instances where they are issued to partners in a partnership are referred to as ‘Partner Payment Notices’ (PPNs). The notices can be issued where the following conditions are satisfied (see Finance Act 2014, s. 219, as well as section 228 and Schedule 32):

1. Either an enquiry or appeal are in progress;
2. A tax advantage accrues from the particular arrangements; and
3. A follower notice has been issued; the arrangements are DOTAS notifiable (FA 2004, s. 311); or a GAAR counteraction notice has been issued (FA 2013, Sch. 43, para 12).

Once issued, the disputed tax becomes payable within 90 days (though the taxpayer has a right to make representations to HMRC). These notices are being challenged through judicial review by 1,000s of taxpayers and it is easy to see why – taxpayers have no right of appeal against the notices themselves and, given the short time period in which the money must be repaid (which could well be into the 100s of £1,000s for tax schemes entered into a decade previously), will have little option but to fight the payment through judicial review. The High Court decision in Rowe was the first to be handed down in relation to accelerated payments (with several cases failing at the High Court on the basis of the judgment of Simler J –  see here, and here for instance) and several cases, or parts thereof, are being stayed behind the appeal (see here, here and here for instance). This general background underlines the importance of the judgment in Rowe and Vital Nut v HMRC.

Before discussing the case itself, I should also note the following things that have become apparent from the litigation in this area over the past two years. The first is that whilst there are explicit statutory conditions which must be complied with before an accelerated notice can be granted (as set out already), there are also inevitably implied conditions which too must be satisfied. This is simply a reflection of the fact that these notices are a ‘game-changer’ and significantly impact the taxpayers concerned. (This was predicted here, and confirmed here and here where the judges introduced implied conditions). Secondly, in order for a taxpayer to succeed in a challenge to the issuance of an accelerated payment notice, the taxpayer should point to some choice made by HMRC in the execution of the relevant legislation, rather than trying to challenge the legislation itself. This argument was explicitly made by Jonathan Peacock in the British Tax Review. The third is the seeming failure of the courts to properly apply the test for taking into account relevant considerations. Broadly, public law requires in a tax setting that HMRC, or the relevant person in HMRC, takes into account relevant considerations when making a decision (though a distinction is made between mandatory and discretionary considerations) and does not take into account any irrelevant consideration(s). If there is a failure for either reason, the question then is whether HMRC, or the relevant person, would inevitably have arrived at the same decision (see pages 101-106 here). In the case of Dickinson, when deciding whether to issue an APN, the fact that there was previously an agreement with the taxpayer to postpone the requirement to pay the disputed tax until after a tribunal determination was ignored. This seemed a relevant consideration, but the court did not deal with the point. At various points in the lengthy judgment of the Court of Appeal in Rowe and Vital Nut v HMRC, these three observations were confirmed.

In Rowe and Vital Nut v HMRC, the Court of Appeal summarised that the taxpayers sought to challenge the issuance of the notices on six broad grounds, finding against the taxpayers ultimately on all six. Judgment on the first four grounds was given by Lady Justice Arden, with Lord Justice McCombe giving judgment on the remaining two. Lady Justice Thirlwall concurred, save for considering whether the legislation was compliant with Article 6 of the ECHR, whilst Arden LJ added some thoughts of her own on the compliance of the notices with the Convention.

The first ground of appeal was that the notices were unreasonable (or disproportionate or otherwise unfair). The second was that the notices did not comply with the statutory conditions necessary for their issuance. These two grounds were taken together by Lady Justice Arden who dealt with the varied and complex arguments advanced by the taxpayers under these headings such as whether it was unfair to issue the notices long after the schemes had been entered into (unfair either because of the delay, or the fact that the legislation should only apply prospectively, as posited here) or whether it was unfair to issue notices in a period shortly before the hearing of the substantive appeals which underpinned the tax dispute. Arden LJ was minded that the power to issue notices, given its severity, called for “caution” (in line with my comment above):

“In a case such as Mr Rowe’s, if the provisions of the FA 2014 are applied without limitation, the result may be that Parliament imposes a disadvantage on citizen A in order to deter citizens B, C, D, E and F from acting in a similar way. That is on the face of it a remarkable result. In principle, it is possible for Parliament to impose such an obligation, but the court will expect the legislation to be expressed in clear language if it is to achieve that effect. I approach the issues of statutory interpretation arising on this appeal on that basis.” [paragraph 50]

Ultimately, however, the Court was satisfied that the notices were issued by HMRC within Finance Act 2014, consistently with its statutory purpose, following a fair procedure specifically prescribed by the legislation, pursuant to rational and proportionate exercise of HMRC’s discretion and the notices did not involve any unlawful interference with the appellants’ rights under the Convention.

One wonders however whether the taxpayers might have been more successful if their argument had been based on a constitutional principle, such as access to justice, which had been infringed by deliberately sending the notices so soon before the substantive appeal was due to be heard. Courts are more likely to entertain an access to justice point after the success of the argument in the Supreme Court this year in Unison. In the case of the taxpayers in Rowe, HMRC offered a settlement opportunity to the taxpayers prior to the issuance of the notices [see paragraph 42], with the settlement expressly providing that those taxpayers who did not accept the offer would later receive an APN. The combination, it could be argued, was an aggressive means of forcing tax settlement and severely restricting the taxpayers’ access to the court. To recall, an APN requires upfront payment, but a settlement would not and could be staggered (though it is claimed by HMRC that “around 95 per cent of requests for extra time to pay accelerated payments debts have been agreed”). Objectionable here is the fact that HMRC are requiring payment of the maximum amount due (by issuing the APN) despite knowing that less is in all likelihood receivable (by issuing the settlement opportunity). In order for a judicial review of HMRC to be successful, it is necessary for the taxpayer to show that she has been “shafted” in some way (my words), or as Arden LJ put it – “[t]here must be some material factor which makes [the exercise of a discretionary power] unfair.” Could it be argued that this is a material factor – asking for more up front, when knowing it is not due, in a bid to restrict the taxpayers’ access to the courtroom when an appeal is pending?

The third ground of appeal was that the issuance of the notices breached the principles of natural justice because taxpayers were not provided with a proper opportunity to rebut the claims of the asserted tax liability before the notices were served. Arden LJ held that the right to make representations to HMRC after the initial issuance of the notices must include the right to make representations as to the effectiveness of the underlying tax scheme. This again deviates from the strict wording of the statute and reads in a further requirement for HMRC. In the circumstances however, the taxpayers were well aware of the basis of the dispute and that HMRC questioned the effectiveness of the underlying scheme.

The fourth ground of appeal was a technical procedural point – that the notice did not relate to an “understated partner tax” as required by section 228 and Schedule 32 of the Finance Act 2014. In the case of PPNs, the notice must state the “understated partner tax” which is the amount which becomes “due and payable” following determination by the designated HMRC officer of the amount of tax advantage to be denied. The argument here was that no amount was “due and payable” because HMRC never opened an enquiry into the relevant tax return for the right year of the lead taxpayer in Rowe. The taxpayer filed his return for the year and then later, in a separate claim, filed for the relevant relief. HMRC challenged the relief, it was argued, but not the return and thus there was no tax due and payable. This ground of appeal is a direct fallout from the Supreme Court judgment in Cotter (though the taxpayer also had to try to distinguish the Court of Appeal judgment in De Silva, the appeal of which was unanimously refused by the Supreme Court). Arden LJ rejected this argument. HMRC had made an enquiry into the return of the partnership for the loss year, this operated as a deemed enquiry into Mr Rowe’s tax return, including the statement of his share of the relevant loss for the same period. Therefore HMRC did not need to open any other enquiry into the standalone claim for relief.

The fifth ground was that there was a breach of the taxpayers’ human rights (in respect of Article 1 of the First Protocol as well as Articles 6 and 7 of the Convention). The Article 7 ground was not argued before the Court. Whilst McCombe LJ had doubts about the conclusion of the High Court judges that Article 1 Protocol 1 was not engaged, he was happy nevertheless that any interference was in accordance with law and proportionate. As for Article 6, McCombe LJ was satisfied that the availability of the procedure for the making of representations against the issue of notices, backed by judicial review of any decision made, was sufficient.

The sixth ground was also a technical procedural point that HMRC must be satisfied that the disputed amount is in fact owed. The legislation requires that the notice specify the amount of disputed tax (section 221(2)(b) of Finance Act 2014), which should be determined by a designated officer (section 221(3) of Finance Act 2014). Charles J in the High Court found that this required that the officer is not satisfied that the underlying tax scheme is effective. The Court of Appeal however reversed this onus of proof, finding that it in fact required that the designated officer was positively satisfied that the scheme was ineffective. This is an important change as HMRC had argued that it could send out notices for instance whenever a scheme was DOTAS notifiable, unless it was as clear as a “slam dunk case” that the taxpayers would succeed in challenging HMRC as to the effectiveness of the underlying scheme.

Notwithstanding the fact that the incorrect test was applied, the Court of Appeal found that it was “highly likely that the same decision would have been reached by the designated officers in these cases, even if the correct test had been applied by him/her in specifying the sum to be paid.” This, it could be argued, was the incorrect test to be applied if the language of relevant considerations is used. The designated officer should have taken into account whether the scheme was in fact ineffective (which is an importantly distinct consideration from whether the officer was not satisfied that it was effective). That consideration ignored, the test then is not whether the same decision was highly likely, but rather whether it was inevitable. This then is where the court erred – by using the incorrect test of likelihood rather than inevitability to determine whether the officer’s decision was vitiated. But the court did not do so, and it does not appear to have been argued in such a way. But as noted already, it is becoming a common trend to overlook the proper application of the relevant considerations test. Whilst the taxpayers did come close to making an argument on the basis of failing to take account of relevant considerations, this was done so on the basis that HMRC had formulated an overly strict policy for the exercise of its discretion which in turn did not allow for exceptional cases in which the schemes might be effective. (The Court found that HMRC’s policy did in fact allow for exceptional cases).

Some time ago, I predicted that one of the ’notices’ cases would make it to the Supreme Court and this looks likely to happen with Rowe and Vital Nut. Over the past few years, HMRC have acquired important powers which ‘nudge’ the actions of taxpayers – the DPT, APNs, the GAAR, POTAS and so on. These powers are distinct from the traditional powers that HMRC exercised when investigating taxpayers. The public interest in having the Supreme Court consider a case like Rowe and Vital Nut accordingly lies in the fact that it would be the first opportunity to pronounce upon the proper interpretation and scope of these new ‘nudging powers’. This case for instance highlights the role that the courts can play in restricting the wording of statutes and the litigation has highlighted some misconceptions that HMRC had about its power to issue the notices.

Before anything can actually be assessed, there must be information that can form the basis of the assessment. This is true in tax as it is in any other walk of life. A further question arises in respect of the source of that information. Should it be purely from taxpayers? Or should tax authorities be entitled to look at information gathered from other sources also? It seems clear that in order to crosscheck taxpayer’s information, the latter will be necessary. But to effectuate this end, it will be necessary for the tax authority to be granted the requisite powers, and wherever this occurs, the issue of taxpayer protections necessarily arises. James Madison recognised this as long ago as 1788 when he wrote: “you must first enable the government to control the governed; and in the next place oblige it to control itself”. For this reason additionally, the need for tax authorities themselves to act transparently towards taxpayers becomes obvious – it provides for control of the authority by providing clear evidence and means of investigation to concerned parties to either ensure that the authority is acting within its prescribed limits or to obtain remedies where the body fails to act within them.

Against this background, I have just uploaded on to SSRN a report I have prepared on tax transparency for the Annual Congress of the European Association of Tax Law Professors. The report is prepared in response to detailed questions and as a result it does not flow like an academic article pursuing a particular thesis.

The report itself deals with three primary aspects of tax transparency: what information HMRC may acquire and how it may be used, taxpayer protections, and the transparency of HMRC towards taxpayers and the public generally. It may be downloaded here.

Judgment in the case of R (Aozora) v HMRC [2017] EWHC 2881 (Admin) was handed down yesterday. Once again, a taxpayer sought to rely upon the doctrine of legitimate expectations. Once again, the taxpayer lost. The case however flags up important issues in respect of the doctrine of legitimate expectations, in particular the effect of HMRC guidance containing an error of law.

The taxpayer had been issued closure notices, the effect of which was to deny the taxpayer relief under section 790 of the Income and Corporation Taxes Act 1988 in respect of withholding tax imposed by the US. This was on the basis that, as HMRC contended, s. 793A of the Act operated to prevent the availability of the relief. The taxpayer however contended that HMRC guidance (in this instance HMRC’s international manual) gave rise to a legitimate expectation that it would be taxed in accordance with the guidance.

In this case, HMRC’s guidance set out the meaning of the legal provisions, but then also, significantly, added that “At 1 April 2003 the only provisions to which s.793A applies is Article 24(4)(c) of the new UK/US DTA”. This was removed in 2011. Article 24(4)(c) of the UK-USA Double Taxation Convention is a very specific provision which the judge read “several times, in a futile endeavour to understand its purpose” but, in essence, is an anti-avoidance provision in respect of tax on dividends. It was this limitation to section 793A that the taxpayer was seeking to rely upon in the case as creating a legitimate expectation.

With regard to the issue of legitimate expectation, the standard formula set out in tax cases is that you first consider whether there is a legitimate expectation created by the public authority, and secondly whether frustrating that is so unfair as to amount to an abuse of power. An important feature of this case was the idea that HMRC’s guidance had incorrectly stated the law, something which similarly arose in the case of Hely-Hutchinson (which this blog has covered extensively. For instance, see here). The judge quoted Lady Justice Arden’s formulation therein that “it is well established that it is open to a public body to change a policy if it has acted under a mistake. The decision whether or not to do so is not reviewed for its compatibility in the public interest: the question is whether or not there has been sufficient unfairness to prevent correction of the mistake”. In other words, it is legitimate for HMRC to seek to correct past mistakes in respect of guidance and it is only where doing so would be particularly unfair that the body will be estopped from doing so in an individual case.

Applying that law to the case at hand, the judge looked at three issues in respect of the legitimate expectation claim: first, whether there was a relevant representation from HMRC; second, whether there was reliance upon that representation; and finally, whether there was resulting conspicuous unfairness.

The judge was satisfied that there was a relevant representation. This was despite the fact that it came from an internal manual and that it was qualified that it might not cover the particularities of each taxpayer’s case. The finding that an internal manual can create a legitimate expectation, even though it is not per se directed towards taxpayers, is entirely orthodox. As for the qualified language of the guidance, the most prudent of ordinarily sophisticated taxpayer would have interpreted the guidance as meaning that section 793A only applied in respect of Article 24(4)(c).

But the judge did note two further matters in respect of representations. The first was that counsel for HMRC at one point “came uncomfortably close to asserting that HMRC could not in law be prevented in any case from resiling from a representation that could later be shown to be an incorrect interpretation of the applicable law”. Interestingly this is not the first time that it has been suggested by HMRC that it is bound to apply the law correctly and collect the tax due. In the High Court judgment of Hely-Hutchinson, Mrs Justice Whipple noted that HMRC there “came close to characterising the duty to collect tax as a trump card which prevails over all other considerations”.

In truth, HMRC may be precluded from applying a correct interpretation by reason of a legitimate expectation. But there are clearly “conceptual difficulties” with that proposition. This is the second further matter the judge briefly notes. HMRC is under a duty to collect tax due. How can it be the case that it would be empowered to act beyond this express duty by purposefully not collecting that tax which is due? This gives rise to circular conundrum: what authority does the body have to act outside its authority? This issue is addressed in Paul Craig’s Administrative Law 7th edition, at pages 701-716 and also in an article of mine in Public Law.

What is unclear from the judgment however, and the Court of Appeal in Hely-Hutchinson also did not elaborate upon this point, is how much weight to put on the fact that HMRC guidance might contain an error of law.

Turning to the matter of reliance upon the representation in the manual, the judge found that there was no reliance by the taxpayer. This was either because, as a matter of fact, the taxpayer was not told about the HMRC guidance by the advisors or because the advisors actually came to the conclusion about the applicability of section 793A independently, without recourse to the guidance. It might be queried whether it makes sense that it might be necessary to show that an actual taxpayer, rather than her adviser (which could then be imputed to the taxpayer), did in fact rely upon the HMRC guidance in question.

The judge held on the third point that even if he were wrong in respect of reliance, the taxpayer could not demonstrate that it was conspicuously unfair for HMRC to resile from its representation. In order to do so, the taxpayer would have to be able to show that “but for the advice that unilateral tax credit was available, it would not have made the business decision that it did, but would have made a business decision that was more favourable from a tax point of view”. But there was no evidence to support this proposition. This finding however raises a broader conceptual issue. The doctrine of legitimate expectations pursues different objectives. On the one hand, it undoubtedly protects citizens where they have changed their position in reliance upon a representation from a public authority. But the doctrine also seeks to ensure that public authorities act consistently with their policies. Where this is in play, the fact that a particular citizen may not have relied upon the representation is not determinative of the claim. However, the judge in this case only approached the matter as if the doctrine is only relevant in so far as a taxpayer has changed her position in reliance upon an expectation.

Over the past few days, I have had the pleasure of marking around 50 tax essays. And I do mean it has been a pleasure, because in the process of marking you must confront the fundamentals underpinning the questions that have been asked of students. A common issue throughout the questions has been how tax law distinguishes between two things, whether that is the difference between debt and equity; income and capital; an individual and a couple; employed and self-employed or accounting rules and tax rules. In each instance, in the paradigm case, it is easy to see how the two are different. An apple is clearly different from a tree if your business is selling apples. But at the margins unsurprisingly, the two are difficult to separate. If I buy a factory, but sell it very shortly after – is it an asset that has been sold off, or is it stock?

Much energy can be expended in trying to clarify the distinction. The task of doing so is not fruitless and can have obvious benefits to those that are seeking to plan their lives in accordance with the rules.

But there is a more fundamental point to the idea of these distinctions. Why does the particular distinction even matter? It will matter because there will usually be some meaningful difference between falling on one side or other of the line. In tax, the difference will be money (whether that is direct in that less tax must be paid or indirect in the sense of a lower administrative burden) and the result will be that people, companies, entities will seek to place themselves on the favourable side of the line. That is the significance of the distinction – remove that, and the fact that the line might be blurry becomes largely irrelevant. When viewed in this way, the question is not about how to clarify the distinction but whether, as a policy choice, that distinction is justified. If the distinction is justified, then we must live with the unfortunate consequences at the margins and try to clarify the differences. If it is not justified, then the task should be to minimise the difference in treatment between falling on either side of the line.

Bringing this to real world distinctions today that are, or at least should be, testing policymakers, is there a sufficient justification behind distinguishing between different categories of persons engaged in work for the purposes of employment law? What about the distinction in treatment between employed and self-employed persons for tax purposes? Although the Paradise Papers brought together myriad different issues, one oft-cited benefit of offshoring is its use for investment funds. If there is good reason for this, then should there be a distinction in treatment between offshore and onshore investment funds?

It is these questions on the justification underpinning any distinction that our attention should be focused upon. It is only after answering them that our minds should turn to clarifying the distinctions. (For an example of interrogating the justification for a distinction and looking at ways to clarify the distinction, see Professor Freedman’s work on the employee/self-employed distinction most notably here).

*Postscript: Apparently this two stage approach is in fact the approach adopted internally by HMRC (h/t Heather Self @hselftax)

In a recent article for the British Tax Review (see here for Peacock, ‘The “Margin of Appreciation” Afforded in the Tax Tribunals: is there any Limit to Judicial Deference?’ (2017) BTR 404), Jonathan Peacock QC explores recent caselaw where taxpayers have sought to challenge actions by HMRC where a “margin of appreciation” is present. In other words, those cases where the challenge directed itself towards policy choices made by the government or Parliament. In his article he concludes that:

“challenges to primary legislation are almost certain to fail; challenges to operational decisions made by HMRC might succeed…[and importantly] the taxpayer needs to point to some other cardinal principle offended by the actions of HMRC which the courts can balance against the incredible weight on the other side of the balance”

Peacock is absolutely correct. Challenges to policy choices by government or Parliament face an almost insurmountable hurdle in the tax arena, where it is orthodox for the courts to respect those choices (an interesting exercise would be to compare this to for instance challenges to immigration policy choices). This deference is entirely understandable and there are good reasons to support it – there are difficult policy choices which need to be made in respect of tax for which our politicians are ultimately accountable and these are made with the benefit of the best available evidence. Judges simply are not institutionally competent to lightly second guess in such instances. At the same time, one does query the degree to which there is genuine scrutiny of legislation as it passes through the House of Commons and or serious deliberation over the way in which policy choices are formulated in legislation – the recent case of Vrang should give pause for thought on whether the implications of policy choices are fully appreciated by those responsible for their promulgation.

Peacock is right too to point out that taxpayer challenges will be more successful if they point to some particular choice made by HMRC in the operation of the particular legislation, rather than being directed to the legislation itself. This is something that I have raised here on this blog in the context of the current lot of challenges to Accelerated Payment Notices (see here and the links therein). Taxpayers should try to argue for instance that, even though the express statutory conditions attached to the discretionary power to issue an APN may have been satisfied in the case, there may be some other implied condition which has not (see here), or that HMRC failed to take into account a relevant consideration (or took into account an irrelevant consideration) when exercising the power (see here), or that the exercise of the power in the circumstances breached some important constitutional principle, such as access to justice (see here). Even then, taxpayers must still overcome the hurdle that courts in practice will in turn give a “margin of appreciation” (in Peacock’s words) to actions taken by HMRC pursuant to its discretionary powers (see here).

The article is a thoroughly enjoyable read and to be recommended both for its substance and also because it displays Peacock’s unrivalled ability to distill complex issues into simple, discrete points.