21 Sept 2020

Tax and the City Review for September 2020

Mike Lane and Zoe Andrews look at recent decisions in the Irish Bank case; Dodika; and Axa. They also consider the new economic crime levy.

In the Irish Bank case, the Court of Appeal agrees with the Upper Tribunal that attribution of notional capital to a UK PE of an Irish bank (rather than looking at actual capital) is compatible with the UK-Ireland double tax convention. A new tax, the economic crime levy, is intended to raise £100m a year from 2022/23 and is going to be payable by the AML-regulated sector, subject to thresholds and exemptions to be decided. Lessons can be learnt from recent High Court decisions involving M&A disputes: Dodika emphasises the necessity for a buyer to comply strictly with any contractual notice provisions when bringing a claim against a seller in respect of a tax liability; whereas Axa is a reminder of documenting parties’ intentions clearly. Responses to the consultation on the notification of uncertain tax treatment question whether such a measure is necessary and if it is, suggest clarity in the definition of uncertainty is essential to make it workable.

Irish Bank case: attribution of notional capital to permanent establishment

HMRC continues its winning streak in Irish Bank Resolution Corporation and another v HMRC [2020] EWCA Civ 1128, an important case on the interaction of UK domestic legislation with double tax treaties.  The Court of Appeal agrees with the Upper Tribunal (UT) that HMRC’s attribution of notional capital to a UK permanent establishment (PE) of an Irish bank pursuant to UK domestic legislation is compatible with the UK-Republic of Ireland double tax convention (UK/RI DTC).

The taxpayers, IBRC and Irish Nationwide, were companies resident in the Republic of Ireland and both traded in the UK through a PE at the relevant time. The taxpayers were, therefore, chargeable to UK corporation tax on the profits attributed to their respective UK PEs.

Each PE borrowed from the respective taxpayer and paid interest on the borrowing. Each taxpayer claimed deduction of interest expenses paid to them by their respective PE. The amount of interest deductible depends on the level of borrowing by the PE, which in turn depends on the level of capital attributed to the PE. HMRC disallowed the interest on the basis of ICTA 1988, s11AA(3) (now CTA 2009, ss 21(2)(b) and 30). These provisions:

  • require an assumption to be made that a PE is attributed a notional level of capital expected of a distinct and separate enterprise dealing wholly independently with the non-resident company; and
  • disqualify for deduction interest and other costs which would not have been incurred if the assumed level of capital was in fact held by the PE.

The taxpayers appealed to the First-tier Tribunal (FTT) arguing that the UK/RI DTC obliged HMRC to compute a PE’s profit by the PE’s books of account, including the capital actually attributed to the PE, rather than attributing a notional level of capital. The taxpayers argued that s11AA(3) was precluded by the DTC on the basis that the DTC overrides a contrary provision in domestic legislation. The FTT, and then the UT, rejected the taxpayers’ interpretation of Article 8 (Business Profits) of the UK/RI DTC and the Court of Appeal concluded that the UT had been correct to do so.

The Court of Appeal confirmed that in the interpretation of treaties you can take into account OECD materials and foreign case law, to the extent that it is relevant, but you cannot take into account HMRC’s prior practice as this is unilateral. Unilateral practice cannot alter the meaning of a treaty. Academic materials may be of some interest but, as they express the views of their authors, they are not in any sense authoritative in relation to the legal issues of construction before the Court.

The Court of Appeal concluded that the words relied on by the taxpayers, when read in context, do not impose the restriction for which they contend. To construe the phrase “same or similar conditions” as requiring the PE’s actual ratio of free to borrowed capital to be applied would be self-defeating as it would rob Article 8(2) (attribution of profits to a PE) of any real ability to depart from the accounting treatment of the PE which the overseas company might choose to adopt. This would make the application of a uniform test of attribution impossible.

The Court of Appeal’s conclusion is consistent with the 2008 Model commentary but, unlike the decisions of the FTT and UT, is not dependent on it. Article 8(2), and the equivalent provisions of the OECD Model convention as it varied over time, are not intended to lay down precise or exhaustive rules – contracting states have been given a measure of flexibility in deciding how to implement them and it would be inconsistent with this approach to interpret “same or similar conditions” in the way the taxpayers suggested.

This case will be of interest to other UK DTCs with similar business profits articles based on the pre-2010 OECD Models. The wider points about what can be taken into account for the interpretation of DTCs will be of more general interest, however, such as the conclusion that the unilateral past practice of a tax authority, no matter how well informed, is irrelevant to interpretation of a tax treaty.

Whilst the principal judgment was delivered by Patten LJ, the additional comments by Singh LJ may also be of wider note. He observed that ‘some confusion has crept into this area of law because insufficient attention has been paid to the important distinction between different parts of the state’. What was before the Court was a question of what the UK Parliament, as the legislature, had done and not HMRC, as part of the executive. It would be worth bearing in mind the following words the next time you see HMRC attempting to address a perceived problem with primary legislation through guidance: ‘Even if there is a legitimate expectation created by the past practice of HMRC it cannot prevent HMRC giving effect to the will of Parliament; indeed, it is the duty of HMRC to give effect to that will’.

Economic crime levy

It is clear from the consultation document that the government wants more money to combat economic crime than it is prepared to raise from taxpayers in general and that it considers that ‘those who contribute towards the risks within the UK economy should pay towards the costs of addressing those risks’. As money laundering is one of the key economic risks the Economic Crime Plan seeks to address, a new levy, the economic crime levy, raising an additional £100 million a year will be charged from 2022/23 upon the AML-regulated sector. There is also a call for evidence on options for funding a fraud response which may extend beyond the AML-regulated sector.

The consultation, which is open until 13 October 2020, seeks views on the principles of the levy, what the funds raised will be used for, how to ensure transparency over spending, how to calculate the levy, which businesses should pay, and how the levy will be collected and enforced. The government proposes there will be an annual report on the use of the levy and a review of the levy will take place every 5 years.

The government favours a levy based on UK revenue – but it is not clear yet whether this will look at total UK business activity or only AML-regulated activity. It depends whether people respond that it is too onerous to separate out these amounts. There will be a different metric from revenue for deposit-taking institutions: this is likely to be either total income or net operating income.

There is clearly a tension between simplicity and proportionality. To be truly proportionate, the levy would need to reflect the money laundering risk of each business as well as its revenue but finding a suitable metric to measure this is proving difficult. So depending on the results from the consultation, the levy might start off being solely revenue based, with a more refined money laundering risk metric being developed in advance of the first 5 year review.

The rate of the levy will depend on the threshold chosen for liability but the rate is likely to be between £100 and £200 per £1m of revenue. This would give a minimum levy payment of £1000 to £2000. As the aim is to raise £100m a year, the rate is likely to be adjusted each year accordingly. Although as experience with another levy (the bank levy) has shown, once it is on the books it is easy not just to raise the rate to maintain the targeted take, but also to raise the targeted take too!

It has not been decided yet whether the levy will be collected by a (new) single agency, or if it should be collected by AML supervisors (such as the FCA and HMRC). There are pros and cons of each model to be considered.

Quasi-hypothecated taxes and levies – a new tax or levy to raise funds by reference to a particular purpose without actually ring-fencing those funds for that purpose - are quite fashionable at the moment (another such example is the non-resident SDLT surcharge linked to money spent on the Rough Sleeping Strategy to help reduce homelessness in the UK). But this sort of policy, with different (bitty) taxes being imposed ostensibly to fund different projects, adds more complexity to the UK tax system at a time when business calls for simplification and competitiveness. One of the key AML-regulated sectors to be subject to the new economic crime levy, of course, is financial institutions but as many of these are already subject to the bank levy and the bank surcharge, they could surely do without yet another separate charge to calculate and pay.

Lessons to be learned from recent contractual claims cases: Dodika and AXA

Dodika Ltd and others v United Luck Group Holdings Ltd [2020] EWHC 2101 (Comm) is the latest in a line of cases which illustrates the importance of following the correct notice procedure to bring a tax covenant claim. In the context of a claim by some of the sellers for summary judgment to release funds held in escrow, the judge had to consider whether the buyers had validly given notice of a tax covenant claim in respect of a transfer pricing issue under investigation by the Slovenian tax authorities.

The High Court concluded that the notice provisions in the share purchase agreement had not been fulfilled because although the notice gave reasonable detail of the nature of the claim, it failed to give reasonable detail of the matter which gave rise to the claim as required. The notice failed to set out the facts, events or circumstances giving rise to the claim. As the buyers had no reasonable prospect of succeeding to argue in a trial that they had given such details, summary judgment was given to allow the escrow funds to be released.

The High Court construed the phrase in the contract ‘the matter which gives rise to such Claim’ as referring to the factual basis – this meant facts unearthed during a tax authority investigation from which a tax liability might result rather than the mere existence of such an investigation.

The current trend is for courts to side with sellers in cases like this, even though to the untrained observer it may seem that the seller is getting off the hook on a technicality. The notice provisions are part of a seller’s contractual protection and the courts expect them to be followed to the letter, so it is of paramount importance that a buyer making a claim provides all the information required pursuant to the contract, in as much detail as possible, within the prescribed time limits.

Axa SA v Genworth Financial International Holdings LLC and others [2020] EWHC 2024 (Comm) concerned a claim under a share purchase agreement in respect of losses resulting from mis-selling of payment protection insurance. One aspect of this case concerned the interpretation of the gross-up for tax on the receipt of a payment. The judge had to choose between two rival interpretations of the phrase ‘subject to Taxation in the hands of the receiving party’. The first option, looking at a theoretical tax liability at the jurisdiction’s headline rate, was rejected. The phrase ‘subject to Taxation’ in this context meant ‘actually taxed’. So, the gross-up would compensate the recipient only to the extent that it was subject to an enforceable obligation to pay tax.

Parties should ensure that the gross-up clause is carefully drafted to achieve the intended result as it is clear that a court faced with rival constructions will settle on the construction that is more consistent with business common sense.

Notification of uncertain tax treatment of large business

This consultation made us question whether an additional compliance and reporting burden on large business to address the legal interpretation tax gap is appropriate or necessary. Our view is that the existing legal framework of tax compliance already adequately incentivises taxpayers not to take overly aggressive positions in their returns. HMRC’s existing policies and powers to investigate large business taxpayers, combined with the cooperative relationships many large businesses have with HMRC, ought to be sufficient to ensure any such legal interpretation gap is minimal.

However, if this measure does proceed (and HMRC are always keen to add to their armoury!), we hope that a definition of when uncertainty exists will be clear and easy to apply and that there will be adequate exclusions from the regime. Otherwise the regime will be unworkable. In particular, HMRC needs to move away from a regime that will require notification of positions that ‘HMRC is likely to challenge’, regardless of technical merit, to objective triggers such as filing contrary to HMRC’s published view on a position.

Taxpayers and advisers are already struggling to meet burdensome DAC6 compliance obligations covering similar ground so minimising the compliance burden should be a key consideration.

What to look out for:

  • The consultation on the draft Finance Bill 2021 legislation closes on 15 September.
  • The Upper Tribunal is due to hear HMRC’s appeal in the Embiricos case on 6, 7 or 8 October. The FTT had decided that a partial closure notice could be issued under TMA 1970, s28A without specifying amount of tax due. Since the FTT heard Embiricos, there have been two further decisions in this area, Henkes and Levy respectively, which reached opposite conclusions. It is hoped that the Upper Tribunal in Embiricos will provide clarity on the correct approach to full and partial closure notices and the jurisdiction of the FTT to determine a mixed question of law and fact when considering a closure notice application.
  • Comments on the Economic Crime Levy consultation are requested by 13 October.

This article was first published in the 11 September 2020 edition of Tax Journal.

 

This material is provided for general information only. It does not constitute legal or other professional advice.

Practices Tax
Contact Information
Mike Lane
Partner at Slaughter and May
Zoe Andrews
PSL Counsel & Head of Tax Knowledge at Slaughter and May