Statutory interpretation after Brexit: implications from a case study of VAT

Abstract

As the UK left the European Union, a new body of UK law, labelled ‘retained EU law’, was introduced to save and convert certain parts of EU law into UK statutes. This paper explores the impact of Brexit on statutory interpretation in the UK in the context of VAT. In particular, it looks at whether, and the manner in which, UK courts and the Court of Justice of the European Union (CJEU) will move in different directions when interpreting what is essentially the same law. The paper predicts the post-Brexit evolution of statutory interpretation in UK courts based on an empirical study of cases concerning VAT referred by UK courts to the CJEU between 1973 and 2020, augmented by a doctrinal analysis of selected cases. The methodology is built on the premise that past case decisions may provide an indication of the nature of possible future divergence. A case study of VAT may offer wider implications as to departure from the CJEU jurisprudence in other legal areas in the coming years.

Introduction

The UK legal system found itself in a new European environment in 1973, following the UK's accession to the European Economic Community (EEC), as the European Union (EU) was known at the time. One of the central problems of accession was statutory interpretation, with the accession fundamentally transforming the role of UK courts and their judicial practice. The UK's 47 years of EU membership officially ended on 31 January 2020, followed by a transition period that expired on 31 December 2020. The European Communities Act 1972 that established the supremacy of EU law in the UK was repealed by the European Union (Withdrawal) Act 2018 (EUWA 2018), which provides a new constitutional framework for the continuity of certain parts of EU law in the UK to avoid a ‘legal vacuum’ upon Brexit. The EUWA 2018 fundamentally changed the relationship between the EU law and UK law and created new challenges for UK judges tasked with interpretating the new body of UK law that was derived from EU law. On the one hand, the EUWA 2018 ended Parliament's subjection to a foreign court, the Court of Justice of the European Union (CJEU), and restored the UK's domestic judicial supremacy. On the other hand, the requirement for legal continuity entailed continuing observance of the judgments of the CJEU.

Will these different approaches to the CJEU authority endorsed by the EUWA 2018 lead to diminishing influence of the CJEU jurisprudence on the interpretation of retained EU law in UK courts? This paper explores the impact of Brexit on statutory interpretation in the UK in the context of VAT. In particular, it looks at whether UK courts’ interpretation of unmodified EU-derived domestic legislation will diverge from the CJEU jurisprudence. Tax law, which is largely statutory in origin, provides an ideal vehicle to study questions of statutory interpretation.¹ VAT law is perhaps one of the tax or legal areas most affected by the reversion of the highest court of appeal to a national court. Adopted to replace the predecessor turnover tax in the UK, Purchase Tax, in 1973 as a condition of the UK's accession to the EEC, VAT was one of the first imports from EU law into UK law. The UK's primary VAT legislation was enacted to implement an EU directive, designed to harmonise the tax across the EU. During the UK's 47 years of EU membership and the Brexit transition period, a total number of 672 cases were referred to the CJEU by UK courts,² among which 109 concern VAT matters. The relative large number of VAT cases as a proportion of the UK referrals may simply reflect the fact that UK VAT is mostly a product of EU law. This clearly shows the significance of European influence in VAT matters, making VAT an excellent area for studying the impact of Brexit on statutory interpretation in the UK.

The paper adopts a new approach to the question of post-Brexit statutory interpretation, and in particular the future relevance of CJEU jurisprudence in UK courts.³ To date, this question has been addressed from a public law⁴ or a comparative law perspective.⁵ However, there has been a dearth of literature that seeks empirical evidence from UK courts’ past experience with interpreting EU legislation in predicting how UK courts will respond to the legal challenges brought by Brexit. Moreover, despite the substantive and statistical significance of tax law cases, the use of empirical methods to study systematically judicial opinions in tax, compared to other areas of law, is particularly scarce.⁶

This paper fills in these gaps and surmises the impact of Brexit on statutory interpretation in the UK through an empirical study of cases concerning VAT matters referred by UK courts to the CJEU between 1973, when the UK joined the EEC, and the end of 2020, when EU law ceased to apply in the UK, supplemented by a doctrinal analysis of selected cases.⁷ The methodology is built on the premise that past case decisions may provide an indication of the nature of possible future divergence. A combination of empirical study and doctrinal analysis addresses the weaknesses of either research method undertaken on its own, while maximising the strengths of each method.⁸ The empirical study provides the most systematic evidence of whether and how UK courts’ decisions differed from those of the CJEU, whereas the doctrinal analysis of selected cases affords a deeper understanding of why the decisions of the CJEU and UK courts differed.

The findings of the empirical study suggest that the CJEU and UK courts differed with respect to their interpretative approaches, court procedures and the handling of precedents, which can largely be attributed to the tension between civil and common law traditions of these courts and the different institutional roles they have in their respective legal systems. This indicates that, although CJEU case law will continue to be relevant in UK courts, UK judges may pursue a separate path when interpreting EU-derived domestic law.

The remainder of the paper is structured as follows. Part 1 sets out the contexts of VAT law and statutory interpretation in the UK and the EU. This is followed by an explanation of the empirical methodology in Part 2. Part 3 then discusses the empirical findings and implications. Finally, the conclusion summarises the key findings.

1. Background

(a) Legislative background

An explanation of the legislative background is necessary for the discussion of statutory interpretation. The idea of a VAT was developed independently by Thomas S Adams, an American economist, and Carl Friedrich von Siemens, a German industrialist, in the 1920s. The Great Depression and the Second World War, however, distracted governments around the globe from serious consideration of the tax. At the time, an important source of revenue for continental European countries was a turnover tax imposed on business sales. These taxes applied at low rates to every sale along the production and supply chains, leading to serious cascading and consequent economic distortions.

In cross-border sales, the turnover taxes operated on a destination basis: on the one hand, exports were exempt from the tax, and the tax already charged on exports was rebated in the exporting country; on the other hand, imports were taxed in the same way as domestic products in the importing country. A consequence of the cumulative nature of the taxes is that export rebates could not be calculated accurately because it was never possible to know exactly how much tax had been paid on a product. In the earliest stages of European integration, Member States realised that the distortion of the compounding turnover taxes with respect to cross-border sales was a serious obstacle to the completion of the Common Market, which required that domestic products and goods imported from another Member State be taxed on the same basis.⁹

The solution was adoption of a VAT, which removes the cascading taxation by rebating, through a credit mechanism, all tax included in the price of acquisitions by registered businesses.¹⁰ VAT is imposed on the consideration payable for supplies of goods and services, but registered businesses are allowed to deduct the tax they paid on input purchases from the tax they collected on output sales, leaving the net balance to be remitted to the government. The tax burden thus falls on final consumers only and the tax on consumption is exactly proportional to the price of the goods or services. The adoption of VAT was central to the economic integration of the EEC because the operation mechanism of the tax facilitates border adjustments and ensures that domestic sales and imports are taxed in the same manner.¹¹

In 1967, the Council issued two Directives that mandate the adoption of VAT in all Member States.¹² As a result, all Member States had a VAT in place but the details of the taxes varied significantly. Genuine progress towards a common system of VAT was only achieved a decade later, in 1977, with the adoption of the Sixth Directive that harmonised the VAT base.¹³ That Directive and successive amendments were consolidated in 2006 into what became known as the Principal VAT Directive (VAT Directive), which is now the principal legislative instrument governing the common VAT system.¹⁴ The UK adopted VAT, as a condition of entry to the EEC, to replace its predecessor Purchase Tax in 1973. Enacted specifically to implement the requirements of the VAT Directives, the VAT Act (VATA) contains the substantive legal rules regulating the UK VAT regime.¹⁵

This situation changed after the end of Brexit transition period, which eventually became the ‘exit day’. The EUWA 2018 governs the post-Brexit relationship between EU law and UK law and created a freestanding body of UK law, labelled ‘retained EU law’, to convert or preserve most of EU law as it stood on exit day into domestic law. Retained EU law consists of three main components: EU-derived domestic legislation; direct EU legislation; and directly effective rights, etc.¹⁶ EU directives do not, by themselves, fall within the scope of retained EU law. However, domestic legislation implementing EU directives forms part of EU-derived domestic legislation and is preserved in domestic law.

The VATA is not affected by Brexit because it is an Act of Parliament and, therefore, remains effective unless it is changed by Parliament. However, as the EUWA 2018 regulates how retained EU law is to be interpreted by UK courts, it is necessary to determine whether the VATA falls within the scope of retained EU legislation. As mentioned, the VATA was implemented to conform to the requirements of the VAT Directives and, prima facie, falls within the scope of EU-derived domestic legislation. However, not all provisions of the VATA were enacted to give effect to VAT directives. Measures concerning the collection of VAT, rather than the levying of the tax, for example, were not governed by EU law. There are also rules derived from the predecessor Purchase Tax and survived in the VATA by way of derogations from VAT directives, most notably the zero-rating of food.¹⁷ These rules should not be treated as EU-derived domestic legislation. Therefore, substantial parts of the VATA could fall within the definition of retained EU law but some parts could not.

(b) Statutory interpretation

The theoretical debate about statutory interpretation broadly comes down to the tension between textualism and purposivism.¹⁸ The tension was particularly acute in UK courts shortly after the UK became an EEC member, thanks to the different legal traditions in which continental European countries and the UK have their origins. One of the most fundamental ways in which the UK was distinguished from continental European countries is the style of legislative drafting and judicial interpretation approaches.

The EU legal system comes within the Romano-Germanic family that favours general principles. EU legislation, therefore, tends to be expressed in abstract terms and leaves the judges to work out the details. The continental European legislative approach undoubtedly reflects the influence of the legal traditions of the original six Member States.¹⁹ However, the generality of EU legislation should also be understood in the unique context of European integration. First, EU instruments are products of bargaining and negotiation between all Member States.²⁰ The political compromises and the need to achieve agreement within a narrow time frame inevitably favour the use of ambiguous language and a lack of precision.²¹ Secondly, the multilingual nature of EU law offers a further reason for the use of broad terms. It would have been an impossible task to draft a watertight legislation with a high degree of specificity in 24 languages that are equally authentic.

This style of drafting necessitates a diverse and flexible approach to the interpretation of EU law. The CJEU deploys a range of interpretative methods, including purposive interpretation in the light of the other relevant provisions (‘contextual’), the objectives of the legislation (‘teleological’) and the recorded intentions of the legislators (‘historical’), as well as the literal approach.²² In the Court's own words, ‘in interpreting a provision of European Union law, it is necessary to consider not only its wording but also the context in which it occurs and the objectives pursued by the rules of which it is part’.²³ There is no clear hierarchy between the CJEU's interpretation methods and often the CJEU uses a combination of different methods in reaching its conclusion. However, it is believed that the purposive approach generally prevails over literal interpretation, especially in the event of linguistic divergences between different language versions.²⁴

In contrast, English legislation, which founded the Anglo-American legal tradition, is commonly drafted in exceedingly detailed and specific language, placing considerable constraints on judges’ discretion to interpret the law.²⁵ Accordingly, the English judiciary was traditionally more faithful to the literal or plain meaning of the statutes. The strict literal approach was considered even more justifiable in the context of tax law.²⁶ There was a popular belief that taxpayers are only obliged to pay the taxes which have been unambiguously imposed by statute.²⁷ Furthermore, it was argued that tax law has no purposes besides the raising of revenue to fund public expenditure.²⁸

Since the second half of the twentieth century, however, there has been a clear shift away from the literal towards purposive interpretation in UK courts.²⁹ Both internal and external factors may have contributed to the change in judicial attitude. First, taxation has increasingly been used as an instrument for social and economic policy. Tax laws, as a consequence, are equipped with an array of social and economic purposes, in addition to their primary purpose of raising revenues. Judicial approaches to interpretation have evolved in response to the use of tax laws for non-revenue purposes.³⁰

Secondly, the significant tax rises after the two world wars incentivised taxpayers and their advisers to devise sophisticated tax avoidance schemes that frustrate the objectives of tax legislation without falling foul of the legislative language.³¹ The strict interpretation approach that favoured taxpayers prevailed in UK courts in the first half of the twentieth century and allowed avoidance schemes to further flourish, forcing courts to move to a more purposive approach to the construction of tax legislation.³²

Thirdly, EU membership had influenced judicial interpretation in the UK. The UK VATA, implemented to conform to the requirements of VAT directives, was drafted in the European fashion that features broad terms. This drafting style, along with the requirement of confronting interpretation under EU law, required UK judges, accustomed to traditional English canons of interpretation, to employ a different approach to the interpretation of VAT law. The challenge faced by UK judges in construing legislation with a EU dimension was expressed by Lord Denning, who described the EU law as ‘an incoming tide’ flowing up the UK legal system, soon after the UK's accession to the EEC.³³

While the internal factors remain unchanged, Brexit, once again, changes the landscape of interpretation of legislation originating from the EU in UK courts. An important consequence of Brexit is that UK courts are neither required nor permitted to refer questions of EU law to the CJEU through the preliminary ruling procedure.³⁴ The EUWA 2018 preserved a body of case law, including any principles laid down by, and any decisions of, the CJEU and a UK court as to the interpretation and application of EU law as at exit day, into domestic law. Whilst becoming part of UK law, CJEU decisions made before exit day have the status of UK Supreme Court (UKSC) decisions, which are binding on the First-tier Tribunal (FTT) and the Upper Tribunal (UT). However, in England and Wales, the UKSC and the Court of Appeal (CA) are not bound by any retained CJEU or domestic case law.³⁵ UK courts are not bound by, but may have regard to, post-exit CJEU case law.

2. Empirical methodology

Content analysis methods³⁶ are best suited to answer the questions of whether and how the decisions of the CJEU and UK courts differed in the same cases. The empirical results, presented in Part 3, are complemented by doctrinal analysis to predict the possible divergence between UK courts’ and the CJEU's interpretations of VAT law after Brexit.

Data are comprised of all the VAT cases referred by UK courts to the CJEU between 1973 and 2020.³⁷ This ensures all relevant cases are included and selection biases, a major concern with traditional legal analysis, were avoided. Figure 1 provides an overview of the coding rules.

Figure 1: Coding rules

The cases were coded in two steps. The first step focused on the outcomes of the CJEU decision and the decision of the UK court below the referring court (hereafter ‘UK decision’) in each case. This step aims to reveal whether and the extent to which the UK decisions differed from the CJEU's decisions in the same cases by way of a comparison of the outcomes of these decisions. The cases were classified into three groups: ‘Affirmation’, ‘Reversal’ and ‘Other’. ‘Affirmation’ includes cases in which the CJEU affirmed the UK decision.³⁸ ‘Reversal’ includes cases in which the CJEU reversed the UK decision (hereafter ‘reversal cases’). The reversal cases, viewed collectively, provide evidence of past divergence between the CJEU and UK jurisprudence.

Three categories of cases were coded as ‘Other’. The first is the cases in which a preliminary ruling was sought before a UK court decision was made, most of which are cases referred by a first instance court. However, in a few cases referred by the first instance court, the court expressed an opinion in favour of a party without deciding the case.³⁹ The opinions were treated as decisions and the cases were coded as either ‘Reversal’ or ‘Affirmation’. The second is the two cases pending before the CJEU at the time of writing. The third is the cases in which the CJEU offered a hypothetical conclusion based on the facts set out in the order for reference by the national court, leaving it to the national court to find the outcome in the actual case.⁴⁰

The second step of the coding process focused on the reasonings in the reversal cases to study why the UK decisions diverged from the CJEU's decisions. The reversal cases were classified into two categories: ‘Interpretation approaches’ and ‘Other’. A case was coded as ‘Interpretation approaches’ if the different decisions were a result of different interpretation approaches adopted by the CJEU and UK courts; otherwise it was coded as ‘Other’. The interpretation approaches used by the CJEU and UK court in each case within the former category were recorded as ‘literalism’ if the decision was justified on the literal approach, or ‘purposivism’ if the court grounded its decision on a purposive approach. A purposive approach is defined broadly to include all the interpretation methods that went beyond a reliance on the plain meaning of words, including teleological, contextual and historical interpretations mentioned above.

It is common that courts employ a combination of different interpretation methods in arriving at their decisions. In these circumstances, only the primary approach is recorded, disregarding other approaches employed to provide support to the conclusion reached through the use of the primary approach. This involved the author's evaluation of the importance judges placed on each interpretation approach used and may introduce biases in the research. The potential biases, however, are inevitable in any detailed and meaningful study of judicial opinions and, therefore, do not compromise the suitability of the chosen method in answering the research question.⁴¹

The cases coded as ‘Interpretation approaches’ were subdivided into two groups, ‘Provisions’ and ‘Facts’, based on the interpretative focus of each case. The differences between the judicial approaches used by the CJEU and UK courts were primarily understood on the basis of the distinction between the literal and purposive interpretations of ambiguous statutory provisions. Nevertheless, a large number of VAT cases are not directly concerned with the interpretation of any particular statutory provision, but rather, how a set of facts should be characterised to fall within the clear words of provisions. Interpretation in these cases does not concern what the statute means, but how this statute applies to the facts. In the cases that involve a series of transactions, the conflicting decisions of the CJEU and UK courts may have been driven by the different weights they gave to the private law consequences (or the economic reality) of the transactions.

The emphasis on economic reality inevitably involves consideration of the underlying economic purposes of the legislation.⁴² In contrast, courts that decided on the basis of legal relationship tend to confine themselves to the question whether the legal forms of the transactions fall within the language of the legislation. For these reasons, the interpretation approaches were recorded as ‘literalism’ where the court focused on legal forms and ‘purposivism’ in the cases in which the court placed its emphasis on the economic reality of the transactions.

One limitation of the coding rules is that the reversal cases do not offer a full picture of the differences in judicial approaches used by the CJEU and UK courts. This is because it is common that differently constituted domestic courts make different interpretative choices and even different judges in the same common law court often adopt different reasonings that lead to conflicting decisions in the same case. As such, if a preliminary ruling had not been sought, a decision of an non-apex UK court may as well be reversed by an upper court. Moreover, the reversal cases do not capture all the circumstances in which UK courts and the CJEU used different approaches. In rare cases, the CJEU and UK courts adopted different approaches but arrived at the same conclusion.⁴³ Despite the caveats, the reversal cases afford a direct and powerful test of the differences between the CJEU and UK courts in their interpretation approaches and represent the best data source available to address the research question.

3. Approaches to statutory interpretation

Whilst UK domestic law remains intertwined with EU law through the creation of ‘retained EU law’, the key question is whether UK courts will depart from the CJEU jurisprudence when interpretating retained EU law.

(a) Unintentional divergence

(i) Reversal cases: the literal versus purposive division

A comparison of the interpretations of EU VAT legislation by UK courts and the CJEU in the UK references suggests that the decisions of UK courts and the CJEU often yielded different results. Among the cases referred by UK courts, the CJEU reversed 30 UK decisions and affirmed 32 UK decisions. In 26 out of the 30 reversal cases, the different decisions were a result of the marked differences in interpretation approaches adopted by the CJEU and UK courts, with one court using the literal approach and the other applying purposive interpretation.

Table 1 summarises the interpretation approaches used by the CJEU and UK courts in the reversal cases. Seven out of the 13 cases with an interpretative focus on provisions of legislation concern exemptions that exclude certain supplies from the VAT base, a feature mandated by the VAT Directive. Harmonisation of the VAT base was considered necessary because of the decision in 1970 that included VAT as the main source of the Community's own revenue. The Directive mandates two classes of transactions that Member States must exempt in their VAT systems. The first is ‘activities in the public interest’, including postal services, medical and dental services, education, welfare and social services and cultural activities.⁴⁴ A feature common to this group of transactions is that they were of perceived social and political importance in the founding Member States. The second group is ‘other activities’ such as financial and insurance services, gambling and transactions involving immovable property.⁴⁵ These exemptions are commonly explained with the technical difficulty of taxing these supplies under VAT and a simple solution was to exempt these supplies from VAT.⁴⁶

Table 1: summary of interpretation approaches

CJEU (purposivism) versus UK court (literalism): In eight cases, the CJEU applied a purposive approach to the construction of a term or a provision of EU law, whilst the UK court adhered to the literal meaning of the term or provision. In D'Ambrumenil,⁴⁷ for example, the question arose whether certain services provided by a medical practitioner were exempt from VAT. The services in dispute include conducting medical examinations of individuals for employers or insurance companies and giving certificates of medical fitness for travel purposes or medical conditions for purposes such as war pension entitlement. The issue was whether the services could be classified as ‘provision of medical care’ within the meaning of Article 13A(1)(c) of the VAT Directive.

The VAT and Duties Tribunal (‘Tribunal’) expressed a preference for characterising the supplies as exempt supplies based on the ordinary meaning of the term ‘medical care’. It nevertheless referred the question of the proper construction of the term to the CJEU because the scope of the exemption is unclear and the question is of Community-wide importance.⁴⁸ The CJEU held that medical services ‘carried out for a purpose other than that of diagnosing, treating and, in so far as possible, curing diseases of health disorders’ cannot benefit from the exemption.⁴⁹ The CJEU in its purposive approach to the construction of the term took the view that excluding the services in question from exemption is not contrary to the objective of the exemption of medical care, which is to reduce the cost of health care and to make it more accessible to individuals (‘teleological’).⁵⁰

The different interpretation methods adopted by the Tribunal and the CJEU led to conflicting views on the interpretation of the concept. Notably, the CJEU emphasised the importance of the purpose of a medical service in determining whether it should be VAT exempt, concluding that medical services that fall within the scope of the exemption must have a therapeutic aim. This view was specifically rejected by the Tribunal before the case reached the CJEU. The Tribunal said ‘to equate care with treatment would be to restrict its meaning and in a way for which no justification is to be found in the wording of the provision’ (emphasis added) and concluded that whether medical services carried out by a doctor are classified as the provision of care should not be determined on the basis of the purpose of the service.⁵¹

CJEU (literalism) versus UK court (purposivism): Although, as mentioned, it is widely argued that the dominant interpretative method used by the CJEU is the teleological approach,⁵² the CJEU relied (primarily) on literal interpretation in five cases in which the UK court decided on a purposive approach. An example is British Film Institute (BFI). ⁵³ The taxpayer in this case argued that their supplies of rights of admission to showings of films constituted exempt cultural services under Article 13A(1)(n) of the VAT Directive, which the UK failed to transpose into its domestic law in the period in question. The provision refers to ‘certain cultural services’ without specifying which cultural services Member States are required to exempt. It was not disputed that the services in question were cultural services for the purpose of Article 13A(1)(n). The question was whether the Directive provision was of direct effect, with the result that the taxpayer can rely upon the provision directly to benefit from the exemption in the absence of transposition by the UK. This question then turns on whether the expression ‘certain cultural services’ is to be interpreted to include all cultural services and does not allow the Member States a discretion in determining which cultural services to exempt, and is therefore sufficiently clear and precise for the provision to have direct effect.

Both the FTT and UT decided that Article 13A(1)(n) had direct effect before the CA referred the case to the CJEU.⁵⁴ The UT commenced its interpretative task by noting that the EU legislation requires a teleological approach to interpretation and a semantic approach is not appropriate.⁵⁵ It quoted a trite statement of the CJEU in the established case law that ‘the purpose of the exemptions is to avoid divergences in the application of the VAT system as between one Member State and another’.⁵⁶ The UT held the view that interpreting ‘certain cultural activities’ based on its ordinary meaning would inevitably lead to divergences between the tax treatment of cultural activities in different Member States (‘teleological’). The UT, steeped in the doctrine of precedent, justified its decision by reference to prior CJEU decisions. In particular, the CJEU held in another case that the term ‘certain services closely linked to sport or physical education’ in Article 13A(1)(m) requires Member States to exempt all services closely linked to sport.⁵⁷ The UT reasoned that it could not have been the likely intention of the EU legislature to ascribe different meanings to ‘certain’ in two sequential paragraphs of the same provision (Article 13A(1)(m) and (n)) (‘contextual’).

The CJEU, however, endorsed the literal approach in this case, considering that interpreting ‘certain cultural activities’ as referring to all cultural activities does not correspond to the ordinary meaning of the term ‘certain’.⁵⁸ In support of its use of literal approach, the Court looked for the legislative intention in the historical documents that paved the way to the adoption of the Directive (‘historical’). It pointed out that the European Commission's original proposal that contains an exhaustive list of the cultural services was not accepted by the EU legislature, indicating that the legislature intended to leave it to Member States to determine which cultural services were to be subject to the exemption.⁵⁹ It is worth noting that the same legislative proposal was considered by the UT in support of a different conclusion (‘historical’).⁶⁰ This shows how the use of legislative history as an aid to interpretation may complicate the interpretative process and give rise to greater legal uncertainty, given that any historical material is also subject to interpretation and the language employed in it is often capable of more than one interpretation.

Other: There are four cases in which the different decisions of the CJEU and UK courts were not a result of different interpretation methods. In AXA UK plc, for example, the CJEU reached a decision different from that of the domestic court based on an interpretation of part of a provision that was not relied upon or raised by the parties at any point in the domestic proceedings.⁶¹ The different decisions were a result of the different trial processes in the CJEU and UK courts, based on the inquisitorial and adversarial systems respectively. The civil law systems that hold sway in continental Europe typically have an inquisitorial system, with the court assuming a proactive role in identifying the issues and examining the case.⁶² The CJEU, operating an inquisitorial system, has the power to identify points of law or issues not raised by the parties for addressing the dispute. The UK and other common law countries, in contrast, have an adversarial system of legal procedure, under which the judges adjudicate on the basis of the facts and arguments presented to them by the opposing parties.⁶³ It is unsurprising that the procedure in the CJEU in this case was perceived to be unfair by the UK taxpayer when the case returned to the domestic court.⁶⁴ While this case is no doubt unusual, it shows that the different legal procedures may also affect the choice of applicable law and give rise to different outcomes.

(ii) Implications for post-Brexit interpretations

Post-Brexit, the new set of legal arrangements, set out in the EUWA 2018, that governs the relationship of the UK and EU laws requires UK courts to take a new approach to the interpretation of retained EU law. The EUWA 2018 converted the pre-exit CJEU case law into domestic law to ensure legal continuity. However, this objective may be compromised if the CJEU's interpretation methods are not preserved, with the result that UK judges might head in a different direction when interpreting retained EU law.⁶⁵ The explanatory notes to the EUWA 2018 provide that UK courts are required to take a purposive approach to the interpretation of unmodified retained EU law.⁶⁶ Resorting to the recitals and preambles of an EU instrument and materials that led to the adoption of the law is referred to as an example of how the purpose of the law can be sought.⁶⁷ Importantly, the requirement to take a purposive approach is an example of fulfilling the UK courts’ obligations under section 6(3) of the EUWA 2018, which provides that the meaning of unmodified retained EU law will be determined in accordance with the pre-exit CJEU case law and general principles of EU law. This indicates that the intention of the EUWA 2018 is to preserve the CJEU's interpretation methods, which are conceived to be broadly purposive.

A challenge for UK judges is to ascertain the purposes of retained EU law. Does retained EU law inherit all the purposes of EU law? The purposes of EU law are partly inherited through preserving general principles of EU law and the pre-exit CJEU case law. This seems to be logical in relation to VAT not only because of the need to preserve legal certainty but also because arguably a large bulk of the UK's VAT legislation does not have its own purposes. However, it might be going too far to argue that the UK law should inherit all the purposes of EU law, especially where the purposes of EU law are associated with European integration. The introduction of the EU VAT, as noted, was driven more by the need to remove cross-border fiscal obstacles than the revenue-raising objective. The effectiveness of EU law partly relies on uniform application and interpretation in all Member States. Requiring UK courts to interpret retained EU law, which is now solely UK domestic law, in a way that achieves the EU's Common Market objectives or ensures uniform application by the UK and its EU counterparts clearly is not fit for purpose in a post-Brexit legal system. Therefore, with Brexit changing the purposes of UK law, it is not unreasonable to expect further divergence between the CJEU and UK courts in their interpretations in legal areas that were crucial to European integration.

To the extent that the purposes of EU law are preserved in retained EU law, are UK courts able to give effect to the purposes of retained EU law without external assistance? The differences between the approaches used by UK courts and the CJEU discussed earlier indicate that UK courts were not always able to find the purposes of EU law and assistance from the CJEU was necessary on occasion. This does not mean that UK courts were less determined than the CJEU in searching for legislative intention. While the expansive definition of ‘purposivism’ adopted in this paper captures all the interpretative methods that extend beyond the plain meaning of words, a closer reading of the cases suggests that the purposive approach used by UK courts is more literalist than that adopted by the CJEU. This is a result of the starkly different roles of the CJEU and UK courts in their respective legal systems.

UK courts’ strict literal approach is derived from the doctrine of parliamentary sovereignty that has its emphasis on the absolute law-making authority of Parliament.⁶⁸ While English judges have been moving towards purposive interpretation over the past few decades, they seem to have applied the purposive approach insofar as the intended purpose can be discerned from the wording of the legislation. In D'Ambrumenil, as noted, the Tribunal rejected an alternative interpretation, which was later invoked by the CJEU, on the basis that no justification for that interpretation could be found in the wording of the provision. In one of the reversal cases, the FTT fell back to the plain meaning of the words after concluding that it was not possible to discover an evident purpose based on the wording of the provision in question.⁶⁹

The CJEU, in contrast, often does not hesitate to fill in the gaps in the legislation or even depart from the literal meaning of words in seeking the purposes of EU law, which might be seen by English judges as ‘a naked usurpation of the legislative function under the thin guise of interpretation’.⁷⁰ While the different legal traditions have undoubtably influenced the choice of interpretative methods in the CJEU and UK courts, another significant reason that explains the differences in judicial approaches is the political dimension to the CJEU's role as a supranational court that distinguishes it from orthodox courts. The CJEU is often criticised for being a ‘mission’ engine of European integration.⁷¹ The flexible interpretative methods provide the tools it needs to expand the competence of EU law through judicial interpretation into areas where the EU's legislative competence is limited.⁷²

This appeared to be the case in areas where the policy purposes of EU law are unclear. Justifications for a purposive approach are built on the presumption that every statutory provision has an underlying legislative purpose. However, most of the exemptions, for example, were introduced not on the basis of clear rationales but as a matter of path-dependency.⁷³ Certain broad categories of goods or services were untaxed in the predecessor turnover taxes in the founding members of the EEC and it was politically difficult to remove the concessions.⁷⁴ The only possible way to achieve harmonisation of the tax base at the time was to grandfather these historic curiosities in VAT. The CJEU, when asked about the meaning of a provision that lacks an identifiable purpose but which was introduced as a result of political compromises, had at times engaged in judicial policy-making in its attempts to uncover the hidden rationales for the provision.⁷⁵

Another unfortunate consequence of the political compromises is that EU law contains conflicting provisions. The exemptions provide an example of conflicting provisions in the VAT Directive. While the Directive mandates broad categories of exemptions, it gives Member States discretion to lay down the conditions in accordance to which the exemptions shall apply.⁷⁶ The Commission noted in its report to the Council that ‘it seems paradoxical to introduce cases of compulsory exemption and leave the substance to the discretion of each Member State’.⁷⁷ Nevertheless, its proposal of an exhaustive list of supplies subject to exemptions was not accepted by Member States.⁷⁸

The broad and ambiguous terms are left to be interpreted by national courts, and ultimately, the CJEU. In several cases, the CJEU went beyond ascertaining the purposes of exemptions and altered the scope of exemptions by adding conditions not found in the wording. For example, in D'Ambrumenil, the CJEU concluded that exempt provisions of medical care must have a therapeutic aim. It held in Institute of the Motor Industry that, for their supplies to fall within the scope of exemption, trade unions must have their main aims as defending the collective interests of its members and representing them vis-à-vis the appropriate third parties, including the public authorities.⁷⁹ By holding that a supply is exempt only if it has a specific aim, the CJEU bolted additional conditions on to the exemptions, which is within the discretion explicitly given to Member States in the VAT Directive. The English judiciary, in contrast, would not encroach upon what is seemingly a legislative domain.⁸⁰

The CJEU, as shown in Table 1, did not always commit itself to a purposive approach. In five cases with an interpretative focus on provisions, the CJEU relied on the literal meaning of the provisions, whilst UK courts applied purposive interpretations. In three out of these cases, the UK courts grounded their choice of a purposive approach in the CJEU precedents.

Why did the UK decisions differ from those of the CJEU in these cases? This is partly because of the difficulties faced by UK courts in applying the CJEU precedents, as illustrated by BFI. In addition to the different sets of interpretation tools available to the CJEU and English courts and the different legal procedures, deeply embedded in these courts’ respective civil and common law traditions are their preferences over markedly different judicial styles. One obvious distinction between the styles of judgments of the CJEU and English courts is that CJEU judgments are considerably shorter and more terse than fully reasoned judgments of English courts. CJEU judgments contain far less legal analysis and reasoning, which sometimes may even not be regarded as reasoning by common lawyers.⁸¹ It is therefore not surprising that UK judges often find the judgments of the CJEU difficult to understand and follow.⁸² However, in a common law system, previous decisions are often decisive in assisting a court's task of choosing the most appropriate interpretative method, especially where different interpretative methods would lead to conflicting decisions. By way of contrast, although the CJEU, with the influence of common law tradition, also makes references to its own precedents to maintain an impression of coherence, it is not bound to its own previous decisions. Post-Brexit, although UK lower courts are bound by pre-exit CJEU jurisprudence and may have regard to post-exit CJEU case law, the CJEU case law may be distinguished, explained, and developed in a different way in UK courts.

Table 1 also demonstrates that half of the reversal cases concern the construction of relevant facts. The Loyalty Management UK Ltd (LMUK) decisions,⁸³ discussed below, provide an example of cases in which the focus of the interpretative inquiry is the choice of facts, which in turn determines the choice of applicable law. The data reveal that neither the CJEU nor UK courts systematically favoured the economic reality approach over the strict literal interpretation or vice versa. In any case, it is more difficult to identify trends in the cases with an interpretative focus on facts or explain why a court focused more on the economic reality or legal forms in individual cases. This is because these cases, which usually involve complex commercial arrangements, are highly fact-specific compared to cases in which the interpretative focus is on legislative provisions.⁸⁴

The UK's modern approach to statutory interpretation based on a purposive interpretation restrained by the wording of statutes seems to have achieved a satisfactory compromise between its traditional formal interpretation and the CJEU's flexible interpretation. The change in judicial attitude in UK courts, starting in the earlier stages of EU membership, and the decades of experience UK courts had in interpreting EU law led many to believe that UK courts and the CJEU were moving towards the same destination.⁸⁵ However, although cases like BFI clearly show the CJEU's influence on UK courts’ choices of interpretation approaches, the divergence between the CJEU and UK courts’ interpretations, evidenced by the reversal cases, has not been narrowed over time. UK courts, whilst adjusting to the European interpretation style, had not abandoned the traditional English rules of interpretation. Before Brexit, where the UK decisions were overturned by the CJEU, UK courts were bound to follow the CJEU rulings. With the end of the general jurisdiction of the CJEU on UK courts, UK courts are likely to embark on a separate path.

(b) Intentional divergence

The preliminary ruling procedure transformed the judicial hierarchy of domestic courts by allowing the lower courts to engage directly with the CJEU. Figure 2 shows that the overwhelming majority of UK references came from the tribunals. On the one hand, this reflects the fact-finding tribunal's view that a reference ‘should be made at the earliest opportunity’⁸⁶ so as to save time and costs for both parties. On the other hand, the large number of references made by the tribunals may suggest that tribunal judges were not so confident about the meaning of EU law and the preliminary ruling procedure offered them an easy means of resolving the disputes before them. Whatever the explanations may be, it is unquestionable that the preliminary ruling procedure provided a ‘fast channel’ through which a large number of cases concerning questions of EU law were resolved without having to climb through domestic appellate courts. It can be expected that appellate courts in the UK will hear more VAT proceedings, with the end of the wave of preliminary ruling requests coming from the UK tribunals to the CJEU.⁸⁷

Figure 2: UK-originated CJEU case law by referring court⁸⁸

Although the UKSC and CA are not bound by retained EU case law, they are not expected to depart from this body of case law easily. The EUWA 2018 requires that in determining whether to depart from retained EU case law, they must apply the same test that the UKSC would apply in deciding whether to depart from its own case law.⁸⁹ In the domestic context, the House of Lords (HL) and the UKSC both stated that departing from their own precedent is ‘a power to be exercised with great caution’.⁹⁰

So far there have been no decided CA or UKSC cases exploring the possibility of departing from retained EU case law in VAT. The first decided case dealing with the question of departure was TuneIn, a CA case concerning copyright infringement.⁹¹ The CA, in that case, refused the appellant's invitation to depart from the CJEU jurisprudence. Sir Geoffrey Vos indicated that the reasons for not exercising their new found power to depart from the CJEU jurisprudence include that ‘the CJEU's approach … is neither impeding nor restricting the proper development of the law, nor is it leading to results which are unjust or contrary to public policy’.⁹²

Although there are no VAT cases on issues of departure from retained EU case law, some implications can be sought from pre-exit referrals by the HL. The last referral made by the HL, LMUK, concerned the right of a loyalty scheme manager, named LMUK, to deduct input tax in a customer loyalty scheme.⁹³ Customers signed up for the loyalty scheme earned points with purchases from participating retailers (‘sponsors’) and redeemed their accumulated points for supplies with either the same or other participating retailers (‘redeemers’). Sponsors paid LMUK for the right to issue points to customers and LMUK paid a higher amount to redeemers for supplies that were redeemed by customers using accumulated points. The issue in dispute was whether LMUK has the right to deduct input tax in respect of its payments to redeemers. Application of the law in this case required a proper characterisation of the transactions.

The case wound its way through the Tribunal, High Court and CA before the HL requested a preliminary ruling from the CJEU. The Tribunal and CA characterised the payments as consideration for redemption services made by redeemers to LMUK, entitling the taxpayer to input tax deduction and freeing rewards from taxation. However, the CJEU, relying on a legalistic categorisation of the payments as consideration for supplies to the customers paid by LMUK on behalf of the customers, denied LMUK's right to deduction.

When the matter returned to the UK, the UKSC⁹⁴ (by a majority of 3-2) used its common law technique of distinguishing⁹⁵ and dismissed the CJEU's decision on the ground that the CJEU considered an incomplete set of facts.⁹⁶ The CJEU looked at the transactions in isolation and limited its analysis to the transactions between LMUK, redeemers and customers, without regard to the undisputed transactions between LMUK, sponsors and customers. The UKSC, in contrast, adopted a purposive approach, considering that it is necessary to look at the transactions between all the parties as a whole in order to determine the economic reality.⁹⁷ The CJEU approach, as Lord Reed observed, would lead to an element of double taxation: tax would be imposed on both the reward and the initial supply made by sponsors to customers that includes the right to receive the reward. As a matter of economic reality, the payments made by LMUK to redeemers should be regarded as its business cost of securing that rewards are provided to customers in exchange for their points. LMUK should therefore be authorised to deduct the VAT borne by that cost. This conclusion, Lord Reed said, is consistent with the fundamental principle of VAT being a general tax on consumption that is completely neutral in its application to all economic activities.

An important implication of the UKSC decision in LMUK is that the CA and UKSC, and in particular the latter, will exercise the power to depart from prior CJEU jurisprudence where CJEU interpretations have led to unreasonable outcomes. The UKSC's approach to departure from the CJEU jurisprudence in LMUK is in line with the post-exit approach to departure outlined by Sir Geoffrey Vos in TuneIn.

Conclusion

The 47 years’ of EU membership has no doubt had an irreversible influence on judicial interpretation in UK courts. However, the empirical study of the VAT cases referred to the CJEU by UK courts shows that the interpretations by the CJEU and UK courts were considerably diverged, giving rise to different outcomes. In almost half of the cases heard and determined by both the CJEU and UK courts, the UK approaches contrasted with those of the CJEU in terms of methods of interpretation, court procedures and reliance on precedents. The differences in approach between the CJEU and UK courts can broadly be traced back to their respective civil and common law traditions and their different roles as a supranational or national courts. While UK courts’ shift towards purposive interpretation has to some extent tempered the tensions between English and European legal traditions, the clash of the two legal systems still existed at the end of the UK's EU membership and will continue to affect legal outcomes after Brexit. Freed from the shackles of the CJEU jurisdiction, UK courts will probably move in a different direction when interpreting retained EU law. Moreover, there may be further divergence between the CJEU and UK courts’ interpretations of legislation in areas that were at the heart of the European integration project, given that post-Brexit UK legislation does not necessarily share the same purposes as EU law.

While the UKSC and CA are empowered to depart from CJEU precedents under the EUWA 2018, it can be expected that they would be very circumspect in exercising this new found power. However, the UKSC decision in LMUK suggests that the UKSC had already started to pursue its own path seven years before Brexit where the CJEU decision was at odd with the explicit principles of the VAT legislation. The UKSC decision in LMUK may herald further divergence between the CJEU and UK appellate courts’ interpretations of VAT law. Therefore, divergence from CJEU jurisprudence may occur at all levels of judicial hierarchy in the UK.

This paper has provided a preliminary answer to the question of the impact of Brexit on statutory interpretation in the UK. While it is not difficult to foresee that what is essentially the same law will be interpreted in different ways on opposite sides of the Channel, the prediction will ultimately have to be tested by taxpayers and HMRC on a case-by-case basis, which may take many years and give rise to prolonged legal uncertainty. The methodology and the general findings of this paper may be replicated in other legal areas, not least those dominated by legislation enacted to implement EU directives.