Alison L. Young: Fact/Law – a Flawed Distinction?

young_alison-l2If prizes were awarded to ‘Distinctions in English law’, then a good contender for the ‘lifetime achievement’ award would be the distinction between ‘law’ and ‘fact’. Whilst adventurers have their Swiss Army knife, and the Dr has his sonic screwdriver, lawyers have the multi-purpose malleable ‘law/fact’ distinction which is just as capable of opening or closing avenues of review, or providing a deus ex machina ‘get out of jail free’ card – or so a perusal of two recent decisions of the Supreme Court might have us believe.

On the 13th March, the Supreme Court delivered its judgment in HMRC v Aimia Coalition Loyalty Limited [2013] UKSC 15 concerning VAT payments on the nectar card reward scheme. The nectar card reward scheme relied on a series of transactions. If company B wished to join the scheme, it would contract with Aimia to supply points to customers, C, buying goods and services from B. Aimia would agree to redeem these points to C. In order to do so, Aimia contracted with other companies supplying goods and services, D, to pay for the goods and services supplied by D to C when C used their points rewarded to them from buying goods from B. B would pay Aimia for the points it issued and Aimia would pay VAT on this transaction. In addition, Aimia pays D an amount per point to enable D to supply the goods and services to C using their points. VAT would be paid on these transactions too. Just to confuse matters further, in practice customers would often redeem their points with the company that had supplied the points in the first place – in other words B and D were often the same person, though need not be. The question arose as to whether Aimia could deduct the VAT it paid on transactions with D from the VAT due from transactions with B.

The issue in Jones (by Caldwell) v First Tier Tribunal [2013] UKSC 19, decided by the Supreme Court on 17th April, concerned a tragedy far-removed from the complex tax arrangements of loyalty schemes. Jones was seriously injured when the vehicle he was driving was involved in a collision with an articulated lorry. The collision was caused when a man, H, ran out from near a car parked on the hard shoulder into the path of the lorry. The lorry driver braked, but was unable to avoid hitting and killing H. In the process of braking hard, the rear end of the lorry had swung into the path of Mr Jones’s vehicle. Acting through his mother, Mr Jones applied for compensation from the Criminal Injuries Compensation Authority (CICA). Compensation would only be due if the injuries suffered by Mr Jones were due to a ‘crime of violence’. If H had committed a ‘crime of violence’ then Mr Jones would be able to obtain compensation from the CICA. If not, then tragic and devastating as Mr Jones’s situation would remain, he would not be able to obtain compensation from the CICA.

In both cases the issue on which the case turned can be expressed fairly easily. Is Aimia paying D for services received from D to Aimia, and can therefore deduct this VAT, or is Aimia merely a third party to the contract between D and C and so cannot deduct the VAT? Was Mr Jones seriously injured because H had committed a crime of violence when he ran out on to the road? In both cases, whether this issue, or elements contributing to its determination, was on of law or ‘law’ or ‘fact’ would also be determinative. In Aimia the VAT implications of the nectar card loyalty scheme had a long history. The case had been referred to the Court of Justice of the European Union (CJEU) under Article 267 (TFEU) by the House of Lords in 2008. When hearing the reference, the CJEU had joined the case with another that also concerned VAT issues from loyalty card schemes – Baxi v HMRC. The CJEU concluded that payments from companies like Aimia, who facilitated loyalty schemes, were consideration payments as regards contracts to which they were third parties and, as such, VAT could not be deducted. It would appear, therefore, that the outcome in the Supreme Court is straightforward. If the CJEU concludes that Aimia cannot deduct VAT then the national court must also conclude that Aimia cannot deduct the VAT, after all, in the words of Lord Carnwath in Aimia, ‘Luxembourg has spoken’ [2013] UKSC 15 [120]-[123]. However, things are not always as they seem. Lord Carnwarth was not expressing his opinion as part of the majority, but in dissent. The majority, however, did not reach their conclusion by deliberately defying the CJEU. Rather, their conclusions rested on an interpretation of the scope of their obligations to follow a preliminary ruling of the CJEU. Article 267 TFEU distributes differing tasks to domestic courts and the CJEU. Whilst the CJEU determines the issue of European Union law, it is for the domestic courts to apply this determination of the law to the facts. The majority in Aimia concluded that the ‘law’ was fairly clear and that the real issue here was how the law applied to the distinct factual arrangements in the nectar card loyalty scheme. Although the preliminary reference in Aimia was joined to that in Baxi, the majority felt that the two schemes differed on their facts. Baxi involved schemes where A contracted with B to administered loyalty schemes by B to its customers. In effect, C received points from B and redeemed them for goods and services from B, the scheme being administered by A. The nectar card scheme in Aimia was different. C received points from B and redeemed them to obtain goods and services from a range of companies, sometimes this would be B, but it need not be from B and could be from D, E, F etc. This difference in facts meant that the Supreme Court was not bound by the conclusion of the CJEU that VAT could not be deducted.

In Jones, appeals from the first tier tribunal to the upper tribunal, and from the upper tribunal to the courts, can only be for  issues of ‘law’ and not for ‘fact’ (and, post Cart, not all appeals on matters of ‘law’ from the upper tribunal will be heard by the Court of Appeal).  When determining whether H had committed a crime of violence, there was the need to determine the legal definition of a crime of violence and to determine whether the facts about H’s conduct satisfied the legal definition. When H walked into the road, leading to a chain of events that caused Mr Jones’s injuries, had he committed grievous bodily harm under section 20 of the Offences Against the Person Act 1861 and, if so, would this be a crime of violence? Problems arose as to whether H had the requisite mens rea. The Supreme Court concluded that whether section 20 of the Offences Against the Person Act 1861 was a ‘crime of violence’ was an issue of ‘law’, deciding in addition that the section 20 offence was clearly a ‘crime of violence’. Whether H had the requisite mens rea was a question of fact. Lord Hope and Lord Carnwarth, with whom the rest of the Court agreed, went further, concluding that the distinction between ‘fact’ and ‘law’, at least in these circumstances, was best understood pragmatically. For Lord Hope, ‘[a] pragmatic approach should be taken to the dividing line between law and fact, so that the expertise of tribunals at the first tier and that of the Upper Tribunal can be used to best effect. An appeal court should not venture too readily into this area by classifying issues as issues of law which are really best left for determination by the specialist appellate tribunals.’ [2013] UKSC 19 [16]. Lord Carnwarth, referring to the decisions of Lord Hoffmann in Moyna v Secretary of State for Work and Pensions [2003] UKHL 44 and Lawson v Serco [2006] UKHL 3, quoted from his own academic commentary on these judgments to conclude that ‘the division between law and fact … is not purely objective, but must take account of such factors as ‘expediency’ or ‘policy’ [2013] UKSC 19 [46]. For Lord Carnwarth, these factors of policy and expediency include ‘the utility of an appeal’, the ‘relative competencies’ of the tribunal of fact and the reviewing court and must have regard to the ‘development of the law in the particular field.’ [2013] UKSC 19 [46]. Although not wishing/needing to give a full account, it was clear that, for Lord Carnwarth, the definition of a ‘crime of violence’ was ‘a task primarily for the tribunals, not the appellate courts’ [2013] UKSC 19 [47]. For Mark Elliott, Lord Carnwarth’s approach, where an error of law is nevertheless deemed to be an error of fact in order to ensure that it is not subject to appeal, demonstrates that the distinction between law and fact, ‘appears to be positively liquified’.

However, the question remains whether it is the distinction between law and fact that is flawed, or the use to which we put the distinction. There are ‘facts’ about the nectar card scheme that, though complex, nevertheless are ‘facts’ about the series of contractual relationships underpinning the loyalty scheme. The difficulty in Aimia was in determining how the VAT Directives would apply to these facts once determined. There are ‘facts’ about H’s behaviour which, though tragic, are nevertheless ‘facts’ and whether a difference in facts was sufficient to distinguish the preliminary ruling of the CJEU on the VAT implications of loyalty schemes. The difficult in Jones was that it would not be possible to determine some of those facts – nobody could ask H about his state of mind when he took the tragic decision to walk onto a busy carriageway.  In both of these scenarios, if errors were made, logically they could only be errors as to the facts, or as to the definition of the law. To use a more simple example, if legislation enabled clothing allowances to be paid to tall public law academics, then an error would have been made if the clothing allowance was paid to the author of this blog post. This error could either be a factual error of a legal error. Either my height was measured incorrectly (maybe I learnt from my daughter and stood on tiptoe, or found a dodgy tape measure) or the legal definition of ‘tall’ applied to my correctly recoded height was wrong (how can anyone a couple of inches below average height be considered ‘tall’?).

But to recognise this is to miss the deeper issue. We are concerned about the facts of loyalty card schemes and of the scope of application of VAT Directives because we need to ascertain the amount of VAT owed by Aimia to HMRC. We want to know the facts about what happened on one tragic day because we want to know if the injuries caused to Jones were caused by a crime of violence and so should give rise to compensation from the CICA.  To answer these questions does require an assessment of facts and an ascertainment of legal definitions and errors can be made with regard to both. But nothing is ever that simple. We can define legal terms with varying levels of precision. Often our definitions will refine themselves and become more precise when ascertained against the backdrop of particular facts. I am not ‘tall’, but if for some reason public law academics tended to be shorter than other academics, such that all public law academics were below average height, maybe I would be a ‘tall’ public law academic and therefore could receive the extra clothing allowance even if our abstract refinement of ‘tall’ would not include those below average height.  The definition of ‘tall’ depends on your comparison and so may well be more contextually sensitive than the definition of a ‘crime of violence’.

It may be easy to determine whether a legal or factual error was made. In Aimia the facts about the nectar loyalty scheme were subtly different from the other loyalty schemes in Baxia. Yet the preliminary reference dealt with them in the same manner. The ‘error’, therefore, was ‘factual’. However it may be difficult to know whether the refinement of the scope of a legal definition is best understood as the determination of precision that arises from a clear understanding of the factual background, or whether this process of refining the legal definition is a process of abstract reasoning about the law. Do we determine the precise meaning of a ‘crime of violence’ by looking at the range of possible crimes that could give rise to claims before the CICA to ascertain its meaning against this factual backdrop, or do we refine its meaning by reasoning in the abstract, thinking more generally about what is meant by a ‘crime of violence’?  The nectar card loyalty scheme may be factually distinct, but is this distinction such that it is no longer covered by the law as interpreted by the CJEU?

Yet the law treats the distinction as providing simple and straight forward answers to conclusions as to the relative powers between multi-levels of tribunals and courts. The CJEU interprets the law and domestic courts apply the law to the facts. Tribunals are better at fact-finding, courts are more specialised at determining the law. Whilst this may be accurate as an assessment of whether an appeal to a court from a tribunal is better able to correct a legal or factual ‘error’, it is less accurate when assessing whether the refinement of a legal definition requires input from those with expertise in the specific factual background or those with expertise in abstract legal analysis. This suggests, in turn, that it is not the distinction that is liquefied, but its application. Whilst it may reflect assumptions as to the relative expertise of tribunals or courts to correct a factual or legal error, the assumption that the refinement of legal definitions is always or is better done through a process of abstract legal analysis fails to reflect the way in which some legal definitions can only be, or may be better refined through an analysis of the context in which they are applied. Therefore, the refinement of some legal definitions may be better performed by those with relative knowledge and expertise of this context.

Where does this leave the law/fact distinction? It is not capable of clearly pigeon-holing issues into those that should be resolved finally by the courts or by tribunals. But this does not mean that it is defunct and, for the purposes of English administrative law, should be replaced by a return to the equally malleable, though some would argue more theoretically justifiable, distinction between jurisdictional and non-jurisdictional errors. Maybe the law/fact distinction can provide a prima facie means of determining whether courts should or should not correct earlier assessments of tribunals, with expediency and pragmatic considerations applied, where necessary, to recognise areas where legal definitions require more contextual refinement which may best be performed by those with particular expertise in a specific area of the law. This in turn may mean that these issues are not corrected by the courts – either through being classified as errors of ‘facts’ that cannot be so corrected, or through maintaining their classification as errors of ‘law’ but modifying the stringency with which they are reviewed. There may also be arguments in the opposite direction, where we are concerned about a lack of independence of the original decision-maker requiring more detailed control by courts as an independent and impartial tribunal.

I would go further and argue that it is time to reassess the relationship between administrative bodies, tribunals and courts. Despite the deceptively straightforward wording and interpretation of article 267 TFEU, the inter-relationship between domestic courts and the CJEU is far from straight-forward. The complexities of this inter-relationship sometimes leads to the facilitation of dialogue and the development of complimentary principles and sometimes appears like nothing more than defiance (I’ll leave it to the reader to classify the response of the Supreme Court in Aimia to the preliminary reference of the CJEU). A re-assessment of the relative roles of tribunals and courts post the Tribunals and Courts Enforcement Act 2007 can hopefully allow for the flexibility that leads to greater co-operation, although the wording of the Act which only allows a right of appeal from the first tier tribunal to the upper tribunal, or from the upper tribunal to the Court of Appeal on a point of law may well require judicial sleights of hand to facilitate this objective.

Alison L. Young is a Fellow of Hertford College, University of Oxford.

Suggested citation: A. L. Young, ‘Fact/Law – a Flawed Distinction?’  U.K. Const. L. Blog (21st May 2013) (available at