Our Firm

On 10 November 2021, the Supreme Court delivered its judgment in Lloyd v Google LLC [2021] UKSC 50, finding unanimously in favour of Google and rejecting the claim brought by Mr Lloyd on behalf of himself and over 4 million other iPhone users in respect of alleged breaches of data protection law.

The judgment will come as a significant relief to data controllers of all sizes who, following the Court of Appeal’s decision, faced the prospect of large-scale data claims. The Supreme Court’s discussion of the representative claim procedure will also be of interest to anyone involved in multi-party litigation.

Key points

1. The Court rejected Mr Lloyd’s argument that data subjects are entitled to compensation for the mere “loss of control” of their data, without proof of financial loss or distress. In any event, the claim could not proceed as a representative claim, as facts relating to each represented person were needed to prove any entitlement to damages.
    
2. The case was decided under the data protection laws applicable at the time, which have since been superseded. Whilst it is possible that a different result might be reached on the “loss of control” issue under the new regime (based on potentially broader statutory wording), the need for individual facts would likely still cause issues for a representative claim.
    
3. Despite rejecting Mr Lloyd’s claim, the Court emphasised the representative procedure as “a flexible tool of convenience in the administration of justice”, particularly at a time when digital technologies have greatly increased the potential for mass harm and, therefore, the need for collective redress. It nevertheless considered that a detailed legislative framework would be preferable, although it remains to be seen whether Parliament will take any action.
    
4. As the legal test for a representative claim was not met, the Court did not have to consider whether to exercise its discretion to prevent the claim proceeding on that basis. It therefore did not need to engage with the desirability of this type of action, in particular whether it was “officious litigation” as the High Court had held (with which the Court of Appeal disagreed).

The facts

Mr Lloyd alleged that for several months in 2011-2012, Google had used a cookie to secretly track the internet activity of millions of iPhone users without their consent and sold that information to advertisers. This allegation was not new and had already given rise to substantial settlements in the US.

Mr Lloyd, with the backing of a commercial litigation funder, sought to bring his claim on behalf of everyone in England and Wales who had been affected, with an estimated value of £3bn.

The law

Data Protection Act 1998

The data protection law applicable at the relevant time was contained in the Data Protection Act 1998 (“DPA 1998”).

Section 4(4) DPA 1998 required ‘data controllers’ (such as Google) to comply with various data protection principles, which Mr Lloyd claimed had been breached.

Section 13 DPA 1998 gave individuals a right to compensation where they suffered “damage” due to a breach.

Representative claims (CPR 19.6)

The UK, unlike some other countries, does not have a general mechanism for bringing ‘class actions’. Instead, there exist a variety of procedures by which collective redress can be sought.

Mr Lloyd attempted to use the representative procedure in CPR 19.6, which allows a claim to be brought by (or against) one or more persons as representatives of others who have the “same interest” in the claim.

A key feature of this procedure is that it is ‘opt-out’ and, as such, represented persons do not need to take any positive step, or even be aware of the existence of the action, in order to take advantage of the outcome.

The Supreme Court’s judgment

It had previously been held in another case that “damage”, for the purposes of section 13 DPA 1998, includes (i) material damage (essentially financial loss) and (ii) distress. In either case, an individual assessment of the harm is needed, precluding use of the representative procedure (as the represented persons would not have the “same interest”).

Mr Lloyd, however, sought to “break new legal ground” by arguing that claims under section 13 DPA 1998 could also be made for the “loss of control” over personal data which, he said, inevitably results from a breach of the Act and which all members of the class had suffered. In other words, that damages were available for the breach itself, without the need to prove material damage/distress.

The Court rejected this argument on the basis that:

1. Such a reading of section 13 DPA 1998 was “untenable” as a matter of domestic law and was not required by EU law.
    
2. The fact that the tort of misuse of private information, for which damages can be awarded for the loss of autonomy itself, and the data protection legislation both sought to protect a person’s right to privacy did not mean that they must do so by affording identical remedies. In fact, there were material differences between the two regimes which made the analogy “positively inappropriate”.

The Court went on to hold that, even if “loss of control” did constitute “damage” for the purpose of section 13 DPA 1998, the claim could still not have proceeded as a representative claim, as it would have been necessary to establish, in each case, the extent of any unlawful processing by Google in order to determine the amount of damages (if any) to be awarded. Relevant factors would include the quantity and nature of the data, the period of time over which it was taken and the use to which it was put. This would inevitably differ from one individual member of the class to another.

Mr Lloyd had sought to overcome this issue by disavowing reliance on any facts relating to individual class members and instead claiming an “irreducible minimum harm” suffered by each of them for which a uniform sum of damages could be awarded. This, however, failed because the limited common facts on which Mr Lloyd sought to rely (only that each person had a cookie unlawfully placed on their device) were insufficient to establish anything more than trivial or de minimis harm and, therefore, any entitlement to damages.

The Court similarly rejected Mr Lloyd’s alternative method of assessment based on ‘user damages’ (a hypothetical sum which data subjects could have charged Google for releasing it from its duties), as, even if such damages were available (which was not the case), the sum which Google would pay to place a cookie on a user’s device, but not to collect or use any of their data, would be zero.

The future of representative claims

Whilst the judgment highlights a significant constraint on the ability to use the representative procedure to claim damages, the Court nevertheless held that this would be possible where the damages could be calculated on a basis common to all the persons represented (e.g. where they were each wrongly charged a fixed fee) or where the loss suffered by the class as a whole could be calculated without reference to individual claims.

The Court also considered that, where damages would require individual assessments, a bifurcated (two-stage) process might be used, whereby a declaration of liability is sought through a representative claim, followed by individual (or presumably group) claims for damages. This approach could have been adopted by Mr Lloyd but wasn’t, “doubtless”, the Court held, because pursuing the individual claims would not have been cost-effective.

Also, whilst the Court held that the representative procedure was not available in this case, it nevertheless endorsed a liberal approach to it and, in particular, the “same interest” requirement, which will no doubt influence lower courts for years to come. In particular, it held:

1. Contrary to the approach taken in some other cases, the “same interest” requirement does not impose a tripartite test (namely a common interest, common grievance and remedy beneficial to all). Instead, it should be interpreted purposively in light of its rationale and the overriding objective of the CPR.
    
2. The purpose of the requirement is to ensure that the representative can be relied on to promote and protect the interests of all represented persons. Whilst this would not be possible where there was a genuine conflict of interest between them, there was no issue with class members having “merely divergent interests”, in that an issue arises in respect of some of the claims but not others. Such concerns as may have once existed about whether representatives could be relied on to pursue vigorously lines of argument not directly applicable to their case are misplaced in the modern context, where proceedings are typically driven by lawyers/litigation funders, with the representative claimant merely acting as a figurehead.

The full text of the Supreme Court’s judgment is available here.

On 12 July 2021, the First-tier Tribunal (Tax Chamber) (“FTT”) released its decision in Scanwell Logistics (UK) Limited v HMRC [2021] UKFTT 261 (TC), rejecting the taxpayer’s claim for onward supply relief (“OSR”).

Whilst OSR is now limited, post-Brexit, to goods imported into Northern Ireland for onward supply to the EU, the FTT’s discussion of agency under section 47 of the Value Added Tax Act 1994 (“VATA”) is of broader interest.

The case serves as a reminder of the significant financial consequences that can result from errors in tax planning, as Scanwell was ultimately held liable for £5.7 million in unpaid import VAT despite the fact that the imported goods almost immediately left the UK (which, if properly planned, could have meant Scanwell was relieved from liability to import VAT).

The Facts

Scanwell acted as an ‘import agent’ in relation to the import of goods from China into the UK (which at the time was still an EU member state) and their onward transport to end customers in other EU countries.

On importing the goods into the UK, Scanwell claimed exemption from import VAT under the OSR provisions. HMRC, however, considered that the relief was not available to Scanwell and subsequently assessed Scanwell to import VAT. Scanwell appealed the assessment to the FTT.

The Law

Where goods are imported into the UK, import VAT is typically charged under section 1 of VATA.

At the relevant time, OSR provided a relief from import VAT where the goods imported into the UK were subsequently re-despatched to a VAT-registered entity in another EU country, generally within one month of importation.

Scanwell accepted that it had never acquired title to the goods from the Chinese suppliers. Instead, it argued that, since it acted as agent for the end customers in the EU, it fell to be treated as part of the supply chain under section 47 VATA, so as to meet the requirements for OSR.

The FTT’s decision

After careful consideration of the legislative provisions, the FTT (Judge Hellier) concluded:

1. In order to qualify for OSR Scanwell had to make, or be deemed to make, supplies of goods;
2. A supply of goods required the transfer of title in goods. Scanwell did not make any actual supply of goods;
3. Scanwell could potentially be deemed to make a supply by section 47;
4. Scanwell’s activities did not fall within section 47(1), therefore they could potentially fall within section 47(2A);
5. Scanwell’s activities did not fall within section 47(2A) because it did not bring about the supply of goods and did not act in its own name in relation to any supply;
6. As a result Scanwell was not entitled to OSR.

Section 47(2A) VATA

Whilst the FTT considered various provisions within section 47 VATA, its main focus was on section 47(2A), notably rejecting HMRC’s contention that it was limited to purely domestic transactions.

Section 47(2A) provides that “where… goods are supplied through an agent who acts in his own name, the supply shall be treated both as a supply to the agent and as a supply by the agent”.

The FTT decided that this did not apply on the facts, as Scanwell was neither (i) an agent through which goods were supplied nor (ii) did it act in its own name for the purposes of the provision. Specifically, it held:

1. A person is an agent through which goods are supplied “only if he has authority to give rise to a transfer of title to goods to his principal or to cause title to his principal’s goods to be transferred to another”. It was not sufficient that Scanwell be an agent in the sense of having authority to receive and take custody of the goods as they arrived in the UK.
   
   Whilst the agreements with the end customers appointing Scanwell as an ‘import agent’ clearly envisaged and intended that Scanwell would be able to claim OSR (and might therefore be construed as granting it the necessary authority), there was nothing in the tasks assumed by Scanwell under those agreements which involved it actually making, or using its authority to give rise to, a transfer of title;
    
2. For an agent to be acting in its own name, its actions “must create a legal relationship between [it] and the third party” (i.e. give rise to rights and obligations), specifically in relation to the transfer of ownership in the goods. The condition is not met where the agent’s actions in its own name are limited to collateral matters, such as haulage, customs clearance and inspection.
   
   Scanwell had no contact with the third-party Chinese suppliers and there was no evidence of any agreement with them under which Scanwell had acted. Whilst Scanwell did receive invoices from the Chinese suppliers (albeit addressed to it as agent), the terms of the agency agreements with the end customers made it clear that Scanwell was not required to pay these, making any obligation under them nugatory.

The FTT dismissed Scanwell’s concern that the denial of OSR would lead to double taxation on the supply to the end customers: import tax in the UK and tax again in the destination member state. This was because there existed an alternative procedure by which the goods could have been transported which would have avoided double taxation, but which had not been used. There was therefore no need to construe the legislation in the appeal any differently. 

The full text of the FTT’s decision is available here.

On 2 July 2021, the Supreme Court delivered its judgment in R (on the application of Haworth) v HMRC [2021] UKSC 25, finding unanimously in favour of the taxpayer and upholding the Court of Appeal’s decision to quash the follower notice issued to him.

What it Means for Taxpayers

The Supreme Court held that HMRC can only issue a follower notice where they consider that there is “no scope for a reasonable person to disagree” that an earlier judicial decision would deny the taxpayer the advantage claimed. This test is both (i) more precise and (ii) imposes a higher threshold than the Court of Appeal’s formulation that required HMRC to only have “a substantial degree of confidence in the outcome”. This reformulation of the test is apt considering the severe consequences which such measures can have for taxpayers.

The Court’s decision is likely to be of primary relevance to cases with less uniform fact patterns or where the issues involved are fact sensitive. Tax avoidance schemes which are mass marketed are likely to be harder to distinguish from those addressed in earlier decisions and thus remain more susceptible to follower notices. Nevertheless, taxpayers may still be able to point to differences in the legal arguments raised, so a thorough assessment of such arguments would be sensible.

The Court also confirmed that follower notices are not automatically invalidated by defects. Taxpayers should therefore be wary of ignoring such notices simply because they consider them to be non-compliant, particularly on formal or technical grounds.

The Follower Notice Regime

The follower notice regime is contained in Part 4 of the Finance Act 2014 (“FA 2014”). It applies where HMRC contend that an advantage claimed by a taxpayer depends on a particular interpretation of a taxing statute which a court has already decided is wrong.

Where HMRC issue a follower notice, the taxpayer has two options: they can either (i) accept HMRC’s interpretation, concede the advantage and pay tax on that basis, or (ii) refuse to do so and maintain the claim. If they do the latter and HMRC are ultimately proven to be correct, the taxpayer may be liable not only for the tax owed but also an additional and substantial penalty calculated by reference to the value of the claimed advantage.

Furthermore, the existence of a follower notice forms one of the bases on which HMRC can issue an accelerated payment notice, which requires the taxpayer to pay the disputed tax to HMRC on account, in advance of the substantive issues being determined.

Background to the Case

The taxpayer sought to make use of what is commonly referred to as a ‘round the world’ scheme to avoid the payment of UK capital gains tax on the disposal of shares by a trust of which he was the settlor. It aimed to do this by taking advantage of provisions in the Taxation of Chargeable Gains Act 1992 and the UK/Mauritius double tax convention to ensure that Mauritius (which did not charge capital gains tax) had the sole taxing rights.

HMRC issued the taxpayer with a follower notice on the basis that the scheme was materially the same as the one which had been held to be ineffective by the Court of Appeal in the prior case of Smallwood v HMRC [2010] EWCA Civ 778 (“Smallwood”).

The taxpayer’s challenge by judicial review was initially rejected by the High Court but upheld by the Court of Appeal, which quashed the follower notice. The case then came to the Supreme Court to finally resolve the issue.

The UK Supreme Court’s Judgment

The main issue

Before HMRC can issue a follower notice, they must be of the opinion that “the principles laid down, or reasoning given, in the [earlier] ruling would, if applied to the chosen [tax] arrangements, deny the asserted advantage or part of that advantage” (emphasis added) (sections 204(4) and 205(3)(b) FA 2014).

The main issue before the UKSC concerned the degree of certainty that this test requires from HMRC as to the application of the prior ruling. The UKSC held that it is not sufficient for HMRC to opine that the earlier ruling is likely to deny the taxpayer’s advantage; instead, they must consider that there is “no scope for a reasonable person to disagree” that it would. Since this threshold was not met, the follower notice was quashed.

The Court’s decision was based, in part, on the fact that the follower notice regime restricts taxpayers’ constitutional rights to have their case determined by a court by imposing the risk of a significant financial penalty. As such, the provisions had to be interpreted narrowly to reduce the interference with those rights to the minimum extent necessary, whilst still being consistent with the aim that Parliament wanted to achieve by enacting the regime, namely to reduce the resources needed to deal with unmeritorious claims. It was therefore appropriate to give full weight to Parliament’s use of the word “would” in the legislation.

Having set out the relevant test, the Court identified four factors that it said would be relevant to whether HMRC can reasonably form the opinion that the earlier ruling would deny the claimed tax advantage. These were: (i) how fact sensitive the application of the previous decision is (i.e. whether a small difference in the taxpayer’s facts as compared with those of the earlier decision would prevent it from applying); (ii) HMRC’s view of the truthfulness (or otherwise) of the taxpayer’s evidence; (iii) whether the taxpayer raised any legal arguments not considered in the earlier decision; and (iv) the precedential value of the earlier decision (e.g. whether the taxpayer was legally represented and the reasoning in the decision was clear).

The remaining issues

The Court also determined three other issues, holding that:

1. HMRC had overstated the Court of Appeal’s conclusion in Smallwood and therefore misdirected themselves as to its relevance to this case;
2. Factual findings do form part of the “principles laid down” or “reasoning given” in a prior decision for the purpose of section 205(3)(b) FA 2014; and
3. Whilst the follower notice was defective as it did not contain all of the information required by section 206 FA 2014, it remained valid as the legislation does not provide that any defect in the notice will render it invalid and the defects in this case did not do so.

The full text of the Supreme Court’s judgment is available here.

First published in CCH Daily incorporating Accountancy Live. 

On 3 April 2017, Marcus Smith J delivered judgment in the High Court in Jazztel plc v HMRC [2017] EWHC 677 (Ch), the test case in the Stamp Taxes GLO. The claimant here sought recovery of UK stamp duty reserve tax (SDRT) that it had paid on the issue of shares to clearance houses and in exchange for depository receipts. Under ss 96 and 93 respectively of the Finance Act 1986, such transactions gave rise to a charge to SDRT at a higher rate of 1.5%, as compared with the 0.5% charge that applied to a standard transfer of shares.

Continue reading on CCH Daily incorporating Accountancy Live (subscription required) 

The case concerned claims made by certain investment trust companies (“ITCs”) for refunds of VAT which they had paid to investment managers on the supply of management services, which later turned out not to be due since the supplies in question were exempt under EU law.

On discovering that the VAT was not in fact due, the investment managers made statutory refund claims and then passed on the refunded VAT and interest to the ITCs. Such statutory claims were however (1) subject to a three year limitation period and (2) incapable of recovering the input VAT which the managers had deducted from the amounts paid on to HMRC. The ITCs therefore did not receive the full amount of the VAT that they had been charged and, as a result, brought claims in unjust enrichment against HMRC claiming that they should be able to recover the gross amount of VAT paid and that their claims should not be limited to the three year statutory period.

The Supreme Court held that the ITCs’ claims in respect of the input tax failed and that HMRC’s enrichment was limited to the amount which they had actually received from the investment managers. The enrichment did not include the amounts retained as input tax credits since the supplies should have been exempt, and so there had been no entitlement to deduct input VAT. The Court went on to hold that the ITCs had, in any event, been unable to make out a claim in unjust enrichment since there had been no direct transfer of funds from them to HMRC and, therefore, HMRC’s enrichment could not be said to have been at their expense. As a result, the ITCs’ claims for VAT paid outside of the three year statutory period also failed. Finally, the Court held that the statutory repayment scheme created an exhaustive code of remedies and, as such, was effective in excluding any common law claims which the ITCs may otherwise have had.

This article appears in the JHA April 2017 Tax Newsletter, which also features:

1. AG Opinion in Austria v Germany: concept of ‘income from rights or debt-claims with participation in profits’
2. Volkswagen Financial Services (UK) Ltd v Commissioners for Her Majesty’s Revenue and Customs

VAT exemption not of direct effect

The CJEU held here, contrary to the decisions of the UK First-tier and Upper Tax Tribunals, that Article 13A(1)(n) of Sixth Council Directive 77/388/EEC, which exempts from VAT the provision of “certain cultural services… supplied by bodies governed by public law”, was not sufficiently clear and precise so as to have direct effect.

The significance of this was that The British Film Institute was not able to rely on the directive itself to recover tax paid in relation to periods prior to the UK implementing it into domestic law. The decision provides useful guidance as to the court’s approach where a Member State fails to implement a directive correctly or within the prescribed time-frame and the potential risks which taxpayers may face in such circumstances.

This article appears in the JHA February 2017 Tax Newsletter, which also features:

1. C-317/15 X v Staatssecretaris van Financiën by Joseph Irwin

CJEU allows restriction on free movement of capital

The CJEU here considered the scope of the exception to the free movement of capital contained in Article 64(1) TFEU. This allows restrictions on capital movements that were adopted prior to 31 December 1993 and which involve direct investment, establishment, the provision of financial services or the admission of securities to capital markets. The court notably gave a broad interpretation to Article 64(1) and held, contrary to the findings of the Dutch Regional Court of Appeal, that the exception was not limited to restrictions on the movement of capital which relate solely to the categories referred to in Article 64(1). As such, it could apply to general measures, such as the extended time-period for the recovery of tax on foreign income here, provided that the particular case involved one of the prescribed types of capital movement.

This article appears in the JHA February 2017 Tax Newsletter, which also features:

1. C-592/15 British Film Institute v HMRC by Joseph Irwin