Burlington Loan Management DAC v The Commissioners for HMRC

The Double Taxation Relief (Taxes on Income) (Republic of Ireland) Order 1976 (SI 1976/2151); and

The Double Taxation Relief (Taxes on Income) (Ireland) Order 1998 (SI 1998/3151) (the “1998 Order”).

absent the application of the Treaty, BLM would have had no entitlement to recover the UK income tax which was withheld at source by the LBIE Administrators when they discharged the interest in respect of the SAAD Claim;

Article 12(1), were it to have applied, would have operated to exempt the payment made in the discharge of that interest from UK tax and therefore entitled BLM to recover all of the UK income tax so withheld; and

the only reason why Article 12(1) would not have applied to exempt the payment made in the discharge of that interest from UK tax was if Article 12(5) applied to that payment.

questions of law – i.e. what is the meaning of Article 12(5)?; and

questions of fact – i.e. how does the law as so construed apply to the agreed facts and the facts which we find in the light of the evidence with which we have been provided?

Mr Brian Bonneson – an associate in DKCM’s distressed investing team in New York, responsible for executing trades;

Ms Jennifer Donovan – a principal in DKCM’s distressed investing team in New York, responsible for closing trades;

Ms Suzanne Gibbons – a managing director in DKCM’s distressed investing team in New York, responsible for analysing and recommending trades;

Mr Christopher McGrath – a managing director in DKCM’s distressed investing team in New York, responsible for executing trades;

Mr Gregory Miesner – Jefferies;

Ms Julie Nettleton - Grant Thornton UK LLP, one of the SICL Liquidators;

Mr Naveen Sabharwal – a member of the team in the European affiliate of DKCM in London, responsible for claims against SICL;

Mr Gabriel Schwartz – a partner in, and managing member of, DKCM’s distressed investing team in New York and the most senior member of the team implementing the transaction;

Mr Travis Troyer – a principal lawyer in DKCM’s legal team in New York; and

Ms Aileen Watson – a managing director in DKCM’s distressed investing team in New York, responsible for executing trades.

a form UCC3 relating to the SAAD Claim dated 29 September 2017, which recorded that:

an email of 12 December 2017 from Ms Nettleton to Mr Sabharwal, in which Ms Nettleton asked “have you been able to consult your NY colleagues on the thinking around a “reserve” price for SICL’s LBIE interest claim?”;

an email of the same date from Mr Sabharwal to Ms Nettleton in which Mr Sabharwal replied:

“We are in consultation with NY and also the LC.

Will revert as soon as we can. In the meantime…please pause this work-stream until we have more clarity on how we would like to proceed”;

an announcement made by the LBIE Administrators on 22 December 2017 (the “Progress Report”), in which the LBIE Administrators informed creditors of LBIE that:

an email from Mr Sabharwal to Ms Nettleton of 2 January 2018, to which Mr Sabharwal had attached the Progress Report and in which Mr Sabharwal said as follows:

“…I assume you saw the below update from the LBIE JOL’s?

My reading of the proposal is that it should result in a full payment of all statutory interest claims (including SICL’s) from the date of the administration … This is obviously good news for the SICL estate.

Could you please confirm you have seen this and that my interpretation is correct?”;

an email of the same date from Ms Nettleton to Mr Sabharwal, in which Ms Nettleton replied:

“yes, I have seen this…I think we all interpret this in the same way as regards the calculation of interest and I agree that it looks like good news for the SICL estate.

Of course, it doesn’t deal with the withholding tax issue which, as things stand following the Court of Appeal decision of 19 December, could see SICL losing 20% of its interest in withholding tax, with no ability to reclaim that sum from [the Respondents]. As you know, this is one of the factors which motivates the discussion around the sale of the claim.

In light of the announcement re the scheme proposal, we are updating our assessment on the sale/hold strategy and will be in touch with the LC about that in due course”;

an email of 5 February 2018 from Mr McGrath to Mr Troyer, copied to Mr Schwartz, Ms Gibbons and Ms Watson, attaching a non-disclosure agreement (an “NDA”) which had been sent to Mr McGrath by Jefferies in relation to “the LBIE poss sale from SAAD estate”;

an email of the same date from Mr Troyer to Mr McGrath and the other recipients of Mr McGrath’s email, in which Mr Troyer replied that he would get the document signed up that day and asked Mr McGrath if, inter alia, Mr McGrath was happy with the three year term of the NDA;

an email of the same date from Ms Gibbons to the same group, in which she replied, inter alia, that she was prepared to accept the term of the NDA if it was non-negotiable;

an email of the same date from Ms Watson to Ms Gibbons and Mr McGrath, in which Ms Watson asked whether London needed to “to okay this re SAAD”;

an email of the same date from Ms Gibbons to Mr Schwartz, Mr McGrath and Ms Watson, in which she replied:

“Looping in Gabe, I believe he discussed with them last week. This is being sold for withholding tax reasons”;

an email of the same date from Mr Schwartz to the same group, and copying in Mr Sabharwal and Mr Troyer, in which he confirmed that he had discussed the proposed transaction with Mr Sabharwal and that he saw no issue with the proposed transaction because the details of the SAAD Claim had already been disclosed in the SICL creditor reports;

an email of the same date from Mr Sabharwal to the same group in which he confirmed that he agreed with Mr Schwartz;

a financial model of 6 February 2018 prepared by Ms Gibbons in which:

email exchanges of 6 and 7 February 2018 between Ms Gibbons and a number of her colleagues, including Mr Schwartz, Mr McGrath, Ms Donovan and Ms Watson, in which:

an email of 8 February 2018 from Ms Watson to Mr Schwartz, Ms Gibbons and Ms Donovan, in which Ms Watson confirmed that:

an email of 9 February 2018 from Ms Watson to Mr Sabharwal and others in which she said:

“I was trying to avoid this but the Seller is insisting they know the identity of the end-buyer for Monday’s close. They don’t seem to understand their buyer is Jefferies.

Gabe and Suzanne were fine with this, but we want to make sure you had no objection to our name give-up.

This is going to be a problem for them to say ‘done’ unless they know. They have assured Greg that the buyer’s identity would change nothing”;

an email of 10 February 2018 from Mr Sabharwal to Ms Watson and the same group in which Mr Sabharwal confirmed that he was content for Ms Watson to inform the SICL Liquidators that BLM was the end-purchaser of the SAAD Claim and said that the SICL Liquidators would know who BLM was because BLM was the lender of record to SICL as well;

an email of 12 February 2018 from Mr Miesner to Ms Watson which confirmed that the trade between Jefferies and DKCM had been executed “subject to successful buy in of the underlying claim” and “subject to all due diligence” and setting out the main terms of the assignment as to completion and price;

an email of the same date from Ms Watson to Mr Miesner confirming that DKCM was in agreement with the terms set out in Mr Miesner’s earlier email but pointing out that the purchaser was in fact BLM and not DKCM;

an email of 22 February 2018 from Ms Gibbons to Mr Bonneson, copying Mr Schwartz, in which Ms Gibbons said, in relation to LBIE, that the timing for the Scheme of Arrangement had slipped from April to May because of a German tax problem and then adding:

“We purchased SICL’s LBIE claim at an attractive price since they had a withholding tax issue”;

an email of the same date from Mr Schwartz to Ms Gibbons and Mr Bonneson, in which Mr Schwartz amended Ms Gibbons’ comment in relation to LBIE by truncating the sentence referred to in paragraph 79(21) above to:

“We purchased a large LBIE claim”;

an email of 26 March 2018 from Mr Prashan Patel of Grant Thornton UK LLP, on behalf of the SICL Liquidators, containing an update on the SAAD Claim, in which he reminded readers that the interest payable in respect of the SAAD Claim was subject to the Waterfall II proceedings and other live matters which had arisen since the last report of the SICL Liquidators relating to UK withholding tax and the “liquidation lacuna” and went on:

“As a result of those matters, the range of recovery for SICL moved to between £nil and £90,736,521.36.

The JOLs are pleased to report that they have successfully sold SICL’s claim in the LBIE administration to a third party for an amount of £82,400,000. This represents the sale of the potential statutory interest distribution on SICL’s claim in the LBIE administration, and has no effect on the principal distribution that SICL has already received.

The sale eliminates the risk to SICL of statutory interest not being received at all due to the “liquidation lacuna”. It also protects the SICL estate from the deduction by the LBIE administrators of withholding tax on SICL’s statutory interest. The administrators would have been required to make that deduction as SICL is not resident in the UK for tax purposes. Had the tax been deducted, SICL would not have had any ability to reclaim that tax”; and

two assignment of claim agreements in relation to the SAAD Claim, each dated 9 March 2018 – one between SICL and Jefferies and the other between Jefferies and BLM – in similar form. Each assignment of claim agreement specified a trade date of 12 February and provided for the assignment of the entire SAAD Claim excluding what it referred to as the “Retained Distribution”, being the principal amount of the SAAD Claim already repaid. Clause 9 in each assignment of claim agreement provided that the purchaser was entitled to transfer the SAAD Claim, together with all rights, title and interests of the purchaser under the relevant assignment of claim agreement without the consent of the seller.

although she was now a partner in, and managing member of, DKCM, at the time of the transaction which is the subject of the appeal she was a managing director in the distressed investing team. She was responsible for analysing and recommending the transaction to the team. The role of the team was to identify value in distressed or highly-leveraged assets;

tax was one of the issues which was taken into account by the team in its analysis of the risks associated with its investments. Although she herself was not a tax expert, she had access to tax advice both from external counsel and from the DKCM internal tax team;

in making its investments, the team had various metrics. One of those was the expected IRR, which was the return which the investment was expected to generate, taking into account time. In addition, the team took into account the perceived downside risk and the perceived upside (the gross return). The team would not invest in an asset unless it had an expected IRR of at least 10%;

she had been involved in acquiring assets in the LBIE administration since 2008. In total, DKCM had purchased 443 claims in the LBIE estate on behalf of BLM, of which 332 were Direct Claims and the remainder were Participation Claims. BLM was the natural purchaser of the claims in the LBIE estate as it was DKCM’s principal European investment platform. She personally had been involved in the purchase of more than 100 of those claims and many of them made since 2014 had effectively been limited to the statutory interest payable in respect of them as the principal amount in respect of them had already been repaid;

at the time when the team were considering its acquisition of the SAAD Claim, the two outstanding downside risks associated with the SAAD Claim from BLM’s perspective were that:

she had valued the SAAD Claim by reference to a financial model which she had produced on 6 February 2018. In preparing that financial model, she had accorded a value of 100% to the interest which was expected to be paid in respect of the SAAD Claim. That was because she knew that any UK withholding tax which might arise on the payment of that interest could be recovered by BLM. The only significance of the UK withholding tax was timing – i.e. the time that it would take for the Respondents to repay the UK withholding tax to BLM. From her own personal perspective, although she knew that the UK withholding tax was not expected to be a permanent cost for BLM because of its tax residence in the ROI, she wasn’t sure that she knew exactly why that was the case. It was just her implicit understanding of the situation. Her awareness of the specifics of the Treaty arose only later, after the SAAD Claim had been acquired;

in preparing the financial model, she had taken no account of the “liquidation lacuna” risk as, in her view, it was remote and, in any event, if it came to pass, it would mean that the SAAD Claim produced nothing and there would be no IRR or gross return to calculate in respect of it. Moreover, as BLM had a position in the debt of LBIE’s parent company, which would give rise to a greater return on that position in the event that the “liquidation lacuna” risk materialised, BLM was hedged to some extent in relation to that risk. In preparing the financial model, she had, however, taken the “late termination” risk into account (in the event that the Scheme of Arrangement did not materialise and the Supreme Court were to overrule the Court of Appeal in relation to the Waterfall II litigation). The financial model showed that, in that event, BLM would receive only 97% of the price which it was prepared to offer – a loss of 3%;

the financial model showed that, as long as the “late termination” risk did not materialise, a gross return of 9% (8.6% rounded up) would arise. In that case, the financial model assumed alternative dates for both the payment of the interest – an upside case of 30 April 2018 and a downside case of 30 June 2018 - and the repayment of the UK withholding tax – an upside case of 31 December 2018 and a downside case of 30 June 2019. The upside case produced an IRR of 28.93%, whilst the downside case produced an IRR of 16.11%;

the higher the purchase price:

For example, a 95% purchase price would not have been viable because the downside risk in that event would have been too great. It would have produced a downside of 6% instead of 3%. In addition, a 95% purchase price would have been incapable of producing an IRR of 10% or more. It was vital that, in addition to not being too exposed in circumstances where the “late termination” risk materialised, BLM could realise an IRR of at least 10% in the event that the “late termination” risk did not materialise;

at one stage in her testimony, she said that, at the time of calculating her offer price for the SAAD Claim, she was aware that SICL was expected to suffer, as a permanent cost, the UK withholding on the interest payable in respect of the SAAD Claim, but she had not factored that fact into her calculation of the offer price for the SAAD Claim. As far as she was concerned, it was simply a matter for SICL and didn’t affect the value of the SAAD Claim for BLM, which was her sole concern. She said that, as a general proposition, she “did not consider the seller’s position because it isn’t relevant” and that “the seller’s position or motivations are not relevant to me as a purchaser”;

at another, she accepted that, as a general proposition, she did take the seller’s position into account in terms of considering what price the seller would be inclined to accept. Her aim always was to offer the lowest price which might be acceptable to the relevant seller but bearing in mind that there were other purchasers in the market who might be prepared to bid more;

in this case, Jefferies had not marketed the SAAD Claim widely to prospective purchasers. Instead, it had given BLM the opportunity to make an offer. This was because BLM had for many years been an aggressive buyer of claims in the LBIE administration. However, if she had made a “low-ball” offer for the SAAD Claim, then Jefferies would have started to look elsewhere. That meant that she wanted to submit a bid that was realistic and fair, which would both result in SICL’s receiving a price that was acceptable to SICL and produce a good return for BLM;

she did not know why:

She could only surmise that those changes must have resulted from discussions at the time as the team members all sat in close proximity to each other in New York;

she also did not know why the SICL Liquidators had asked for the name of the end-purchaser before agreeing to finalise the trade although she surmised that this was for regulatory reasons and not tax-related reasons, given the assurance in the email from Ms Watson to Mr Sabharwal that the SICL Liquidators’ knowing the identity of the end-purchaser would not change anything. She rejected the proposition put to her by Mr Brinsmead-Stockham that, in saying that their knowledge of the purchaser’s identity would not change anything, the SICL Liquidators were leaving open the possibility of negotiating for a greater purchase price if the end-purchaser was not exempt from UK withholding tax under an applicable treaty. As far as she was concerned, the terms of the deal (including the purchase price) had all been agreed and the SICL Liquidators simply wanted to know the identity of the end-purchaser for reasons unrelated to tax. There never was any intention on the part of the SICL Liquidators to re-open the deal once they became aware of the identity of the end-purchaser. All the SICL Liquidators cared about was that they were getting a price with which they were content and they had already determined by that stage that they were;

as regards the statement by Ms Nettleton in her email of 2 January 2018 to the effect that UK withholding tax was “one of the factors which motivates the discussion around the sale of the claim”, she postulated that another factor might have been the disappointment felt by SICL at the unfortunate outcome to the Waterfall I litigation which had led to SICL’s making a lower return from the SAAD Claim than it might at one stage have expected. This might have made SICL less willing to retain exposure to the SAAD Claim, with its “liquidation lacuna” and “late termination” risks, and more willing to get rid of the asset – so-called “seller fatigue”. In addition, as SICL was in liquidation, that too might have made the SICL Liquidators keen to sell the asset;

since DKCM-managed funds were creditors in the SICL liquidation, DKCM was a member of the SICL creditors’ committee. However, responsibility for SICL was located in DKCM’s London office and not in her team in New York. As such, her only involvement with SICL had been in briefing her London colleagues on how they should approach value in relation to the SAAD Claim so that they could factor those thoughts into their recovery model in relation to SICL;

she had no specific recollection of speaking to Mr Sabharwal about the request for a “reserve price” made to him by Ms Nettleton on 12 December 2017 (because he could have consulted on that question with a variety of different people in New York at that time, including Mr Schwartz) but she had definitely spoken to him about the value of the SAAD Claim at some point in the past. Her reading of that exchange was that Ms Nettleton was enquiring about an appropriate market price for the SAAD Claim, bearing in mind DKCM’s involvement in the market; and, since New York was where the DKCM group responsible for marking to market the assets of the LBIE estate resided, it would have been logical for Mr Sabharwal to consult with New York about the market price;

she wasn’t sure how many LACA there had been but certainly more than two. The aim of the LACA was to enable creditors with smaller claims to dispose of their claims and the LACA operated by enabling potential purchasers to acquire a batch of different claims for a single aggregate price; and

it was normal for sales through a broker to be structured in the way this one was – on a “back-to-back” basis – but she accepted that the turn made by Jefferies in this case, which amounted to 1.39% of the sale price received by SICL, was surprisingly high. She said that commissions on the sales of plain vanilla debts tended to be in the region of 0.25%. Having said that, she noted that:

there is no direct case law authority as to the correct construction and application of Article 12(5) (or an equivalent provision in any other double tax treaty between the United Kingdom and any other jurisdiction); and

in the absence of any such direct authority, our task is:

Smallwood v The Commissioners for Her Majesty’s Revenue and Customs [2010] STC 2045 (“Smallwood”) per Patten LJ, at paragraphs [26]-[29];

The Commissioners for Her Majesty’s Revenue and Customs v Anson [2015] STC 1777 (“Anson”) per Lord Reed, at paragraphs [54]-[56] and [110]-[111]; and

Fowler v The Commissioners for Her Majesty’s Revenue and Customs [2020] STC 1476 (“Fowler”) at paragraphs [16]-[19].

double tax treaties had to be interpreted in accordance with Articles 31 and 32 of the Vienna Convention on the Law of Treaties (the “VCLT”); and

consequently, a double tax treaty had to be interpreted “in good faith in accordance with the ordinary meaning to be given to the terms of the treaty in their context and in the light of its object and purpose” (see Article 31(1) of the VCLT).

“the main purpose or one of the main purposes”;

“any person concerned with the creation or assignment of the debt-claim in respect of which the interest is paid”; and

“take advantage of this Article”.

the phrase “main purpose or one of the main purposes” was not defined in the Treaty;

however, it was legitimate to interpret the phrase in accordance with relevant UK case law pursuant to Article 3(2) of the Treaty, which specified that terms which were not defined in the Treaty were to be interpreted in accordance with the provisions of the tax law of the relevant contracting state (in this case, the UK);

determining a person’s main purpose was a question of fact – see Inland Revenue Commissioners v Brebner [1967] 2 AC 18 (“Brebner”) at paragraphs 26B-C and 30C-E;

the conclusion of fact was to be made “upon a consideration of all the relevant evidence…and the proper inferences to be drawn from that evidence” – see Brebner at paragraph 30G;

the test required consideration of the subjective intentions of the relevant person – see Brebner at 27D-E and 30B;

the word “main” did not mean “more than trivial”. Instead, it had the connotation of importance – Travel Document Services v The Commissioners for Her Majesty’s Revenue and Customs [2018] STC 723 (“TDS”) at paragraph [48]; and

in order for this test to be satisfied, it was not necessary for the relevant matter – in this case, taking advantage of Article 12(5) – to be the sole purpose or the only main purpose. It was sufficient for the relevant matter to be one of several main purposes.

the word “person” was defined in Article 3(1)(g) of the Treaty as comprising “an individual, a company and any other body of persons”;

that definition was not subject to any form of restriction as to residence, nationality, or citizenship;

the language cited by Mr Grodzinski from Article 1 of the Treaty merely meant that the Treaty applied only to persons who were resident in one of the contracting states so that, for example, in this context, Article 12(1) could apply to exempt the interest income only of persons resident in the ROI such as BLM, from UK income tax. However, it did not mean that, in applying a provision in the Treaty such as Article 12(5) to BLM or any other resident of the ROI, the word “person” in Article 12(5) of the Treaty should be limited to persons resident in a contracting state. In other words, the Respondents were not seeking to apply the Treaty to SICL. Instead, they were simply saying that, in applying Article 12(1) to BLM, SICL fell within the category of “persons” whose purposes were relevant for the purposes of Article 12(5); and

Mr Grodzinski’s construction would undermine a large number of provisions in the Treaty. For example:

the case was not concerned with the application of Article 12(5) (or a similar provision) at all;

the phrase “to take advantage of” had no particular significance in the context of the decision. It was not a feature of the legislation which was in issue and the court was not considering the meaning of the phrase;

the references in the decision to “artificial arrangements” were simply the way in which the facts of the particular tax avoidance scheme that was relevant in that case were described. The court was not seeking to limit the meaning of the phrase “take advantage” solely to artificial steps or arrangements; and

in any event, the Respondents accepted that Article 12(5) could apply in cases involving artificial steps or arrangements. That was not the point. The Respondents’ position was merely that the existence of artificial steps or arrangements was not a necessary condition in order for a person to be regarded as “taking advantage” of Article 12(1) for the purposes of Article 12(5).

the phrase “fiscal evasion” was not defined in the Treaty and could not, of itself, be said to import into any of the anti-avoidance provisions in the Treaty a requirement for there to be artificial steps or arrangements before the provision could apply;

the phrase was not actually used in the operative provisions of the Treaty at all. The preamble was simply the standard in many of the treaties to which the UK was party. It did not mean that operative provisions in those treaties were confined to cases of fiscal evasion;

Article 12(5) was, in any event, an anti-avoidance provision and not an anti-evasion provision. It was the information-sharing provisions in Article 25 of the Treaty which were anti-evasion provisions in the context of the Treaty; and

in Bayfine UK v The Commissioners for Her Majesty’s Revenue and Customs [2011] STC 717 (“Bayfine”), Arden LJ had said that the reference in the preamble to the treaty that was relevant in that case to “fiscal evasion” “would include the avoidance of taxation”. On that basis, Article 12(5), as an anti-avoidance provision, would be within the scope of the reference to “fiscal evasion” in the preamble to the Treaty in any event.

Article 12(5) was being introduced to counter arrangements intended to take advantage of Article 12(1) and that such arrangements “could consist of routing interest from the UK through Ireland to a third country in order to take advantage of the relief from UK tax under the [Treaty]”; and

Article 12(5) was “an anti-abuse measure designed to prevent artificial arrangements which exist mainly to obtain the benefits of the [Treaty]”.

did nothing more than identify one type of transaction that could fall within Article 12(5). It was not exhaustive;

made no reference to artificial steps or arrangement; and

in any event, in its terms, described the very economic effect of what had actually happened in this case when SICL, a company which could not benefit from the Treaty or any other treaty to which the UK was party with an equivalent provision to Article 12(1), effectively routed its receipt of the interest payable in respect of the SAAD Claim through the ROI in a manner which avoided the charge to UK income tax.

first, it made no reference to artificial steps or arrangements. It merely referred to the abuse of articles such as Article 12(1). Exactly what constituted abuse in that context was to be determined by reference to the language used in Article 12(5) itself and there was nothing in that language to support the contention that the article was confined to artificial steps or arrangements; and

secondly, there was no reference to artificial steps or arrangements in the explanatory note to the equivalent order which implemented the 1998 protocol to the Treaty into Irish law (SI1998/494 of the ROI) (the “ROI Order”). The ROI Order simply stated that Article 12(5) meant that “[the] provisions of the Article will not apply where the main purpose of the transaction is to take advantage of the Article”.

as regards the statements, as a matter of UK domestic law, the present circumstances clearly met the requirements which Lord Browne-Wilkinson in Pepper v Hart AC [1993] 593 (“Hart”) had laid down as being necessary in order for reliance to be placed on Parliamentary statements in construing legislation. Those were that:

the reason that the above tests were met in the present case was that:

if the statements would have been admissible in a UK domestic law context, then it must follow that they were admissible too in construing the terms of Article 12(5). This was because Article 3(2) of the Treaty provided that:

“As regards the application of this convention by a contracting state any term not otherwise defined shall, unless the context otherwise requires, have the meaning which it has under the laws of that contracting state relating to the taxes which are the subject to this convention.”

Since it was possible to apply the Hart principle to interpret the meaning of the phrase “take advantage” as a matter of UK law, it must follow from Article 3(2) of the Treaty that the Hart principle could be used to interpret the use of the phrase “take advantage” in Article 12(5);

in any event, the statements and the explanatory note fell within the ambit of Article 32 of the VCLT and were therefore a permissible aid to the interpretation of the Treaty. This was because:

in Fowler, at first instance (Fowler v The Commissioners for Her Majesty’s Revenue and Customs [2016] SFTD 535 (“Fowler FTT”), the First-tier Tribunal had used statements made in Parliament in relation to a section of the UK legislation as an aid in interpreting the treaty in that case (see Fowler FTT at paragraphs [63] to [65]) and the Supreme Court in Fowler had not criticised this approach; and

it was wrong to approach the interpretation of a treaty as if it were a private contract. If that were appropriate, then OECD and UN commentaries would not be admissible aids to interpretation of treaties whereas it was common ground that they were. If one were to approach the interpretation of a treaty as if it were a private contract, then those commentaries would not be admissible in interpreting a treaty because a third party’s commentary on a bilateral private contract was not admissible in interpreting that private contract. It would be very odd if it were permissible to take into account those commentaries on the meaning of a treaty and not a clear explanation of the treaty’s purpose given by one of the two contracting states.

the statements and the explanatory note were irrelevant to the construction of the Treaty because:

it could not be correct that the UK, having entered into a bilateral treaty provision with another jurisdiction, could then rely on a subsequent unilateral statement made in Parliament or an explanatory note in the enactment bringing the treaty provision into UK domestic law as to the meaning of that treaty provision in order to affect or alter the meaning of the treaty provision;

there was clear authority in the UK that a unilateral statement made by one party to a treaty could not inform the interpretation of that treaty. In Irish Bank Resolution Corporation Ltd and another v The Commissioners for Her Majesty’s Revenue and Customs [2020] STC 1946 (“Irish Bank”), at paragraphs [22] and [23], Patten LJ had endorsed the statement of the Upper Tribunal in the same case (at paragraph [31]) to the effect that:

“We do not consider that the unilateral practice of a contracting party - even if that practice shows a careful attempt by that party to abide by a treaty - can affect the meaning of that treaty or constitute material going to its construction”.

Although that extract was dealing specifically with the practice of the Respondents, the reasoning of the Court of Appeal was not so limited. The same reasoning was apt to apply to any unilateral act of one of the parties to a treaty after the treaty was signed;

as regards the point made by Mr Grodzinski in paragraph 108(3) above, there was a distinction between:

The statements in this case fell within the category described in paragraph 109(4)(b) above, whereas only statements falling within the category described in paragraph 109(4)(a) above were potentially relevant (pursuant to Article 3(2) of the Treaty) in interpreting the Treaty;

by way of example, the second of the two statements, upon which Mr Grodzinski was placing greater reliance, did not go to the meaning of any term of UK domestic law. In the first place, it was merely a statement in relation to the meaning of a treaty provision. As such, it did not fall to be taken into account in interpreting the Treaty pursuant to Article 3(2) of the Treaty. And, even if it might be alleged that it related not to the meaning of a treaty provision but rather to the meaning of the term “take advantage”, Mr Grodzinski had not pointed to any context in which the phrase “take advantage” was a term which had any particular meaning as a matter of UK domestic law;

Mr Grodzinski’s reliance on the decision in Hart was misplaced. Hart laid down a principle for the interpretation of UK domestic law. The principle was not applicable in the context of construing a treaty;

in any event, the principle laid down in Hart had no application in this case for a further reason and that was that the first limb of the principle – which required the relevant legislation to be ambiguous or obscure or to lead to an absurdity – was not satisfied here. In his view, when viewed objectively, the meaning of Article 12(5) was plain and it did not become ambiguous merely because BLM disputed that plain meaning; and

as regards the point made by Mr Grodzinski in paragraph 108(5) above, the First-tier Tribunal in Fowler FTT had looked at statements in Hansard not in order to interpret a treaty provision but rather in order to interpret a provision of UK domestic law, which it thought would be helpful to its interpretation of the treaty – see Fowler FTT at paragraphs [63] and [94].

in paragraphs 7 to 26 of the Commentary, under the heading “Improper use of the Convention”, the Commentary stated that:

and then set out the form of treaty article corresponding to Article 12(5) under the heading “Anti-abuse rules dealing with source taxation of specific types of income”;

in the Conduit Report, the OECD referred to the use of conduit companies as “treaty shopping” and noted, in paragraph 9, that treaty benefits should not be denied in the case of bona fide transactions because they clearly involved no improper use of the treaty; and

in the UN Report, it was noted in paragraph 23 that some members of the former group of experts established by the UN to consider the model form of treaty had pointed out that “there are many artificial devices entered into by persons to take advantage of the provisions of Article 11 through, inter alia, creation or assignment of debt claims in respect of which interest is charged” and that member states might want to address the issue specifically by including an article in the form of Article 12(5) in their treaties.

in commenting on the article in the form of Article 12(5), paragraph 21.4 of the Commentary merely referred to the application of the provision to deny treaty benefits “where transactions have been entered into for the main purpose of obtaining these benefits”. There was no reference to a need for artificial steps or arrangements;

similarly, there was no mention of artificial steps or arrangements in the reference in paragraph 12 of the Commentary to the need to ensure that bona fide economic activities were not unintentionally disqualified by provisions such as Article 12(5) or in the paragraphs which followed that relating to conduit companies;

conduit companies were simply one example of abusing treaties - and thus one way of engaging Article 12(5) - but that did not mean that Article 12(5) could not apply where there was no conduit company;

the earlier version of the anti-abuse provision in Article 12 of the Treaty before its replacement by Article 12(5) pursuant to the 1998 Order had contained an exclusion for bona fide commercial transactions. Thus, the UK and the ROI must be taken to have agreed that, even if a transaction was bona fide and commercial, it could still fall within Article 12(5) if it had a main purpose of taking advantage of Article 12(1). Having said that, Mr Brinsmead-Stockham did not consider this transaction to be bona fide and commercial in any event. The mere fact that SICL wished to realise the best price it could for the SAAD Claim was of no avail if the basis for its obtaining that price was to take advantage of Article 12(1); and

although paragraph 23 in the UN Report had referred to “artificial devices”, that was just one example of the types of transactions which would fall within Article 12(5).

the obtaining of a tax advantage in general; or even

“taking advantage” of the exemption from UK withholding tax in some unknown and unidentified treaty.

by the time that the trades were executed on 12 February 2018, SICL did in fact know that it was “taking advantage” of Article 12(1) specifically because it knew that the end-purchaser was BLM; and

in any event, it was sufficient to satisfy this test for SICL to know in general terms that its end-purchaser was benefiting from an exemption from UK withholding tax of some sort – whether under an applicable treaty or because it was UK resident - and for the end-purchaser in fact to be relying on Article 12(1) specifically. Otherwise, it would always be possible for a person deliberately to close its eyes as to the basis on which its purchaser was obtaining its exemption and thus prevent Article 12(5) from applying.

the appeal stemmed from a claim made by BLM under paragraph 2(3) of Schedule 1A to the TMA (“Schedule 1A”) and the closure notice had been issued under paragraph 7 of Schedule 1A. Since the claim was for a repayment of tax, the closure notice had been issued under paragraph 7(2) of Schedule 1A (and not paragraph 7(3) of Schedule 1A, which dealt with other claims);

the appeal had thus been brought under paragraph 9(3) of Schedule 1A, which provided as follows:

In the case of an appeal against an amendment made by a closure notice under paragraph 7(2) above, …the tribunal may vary the amendment appealed against whether or not the variation is to the advantage of the appellant”;

in contrast to the language used in paragraph 9(3) of Schedule 1A, paragraph 9(5) of Schedule 1A – which related to appeals against closure notices issued under paragraph 7(3) of Schedule 1A – provided as follows:

If, on an appeal notified to the tribunal, the tribunal decides that a claim which was the subject of a decision contained in a closure notice under paragraph 7(3) above should have been allowed or disallowed to an extent different from that specified in the notice, the claim shall be allowed or disallowed accordingly to the extent that appears … appropriate, but otherwise the decision in the notice shall stand good”;

it was significant that the language in paragraph 9(3) of Schedule 1A differed from the language in paragraph 9(5) of Schedule 1A in not providing for the closure notice to “stand good” in the event that the tribunal decided not to vary the closure notice. The formulation in paragraph 9(5) of Schedule 1A – which contained a statement to that effect– tracked the language in Section 50(6) of the TMA and the absence of equivalent language in paragraph 9(3) of Schedule 1A was therefore critical;

this could be seen in the decision of Henderson J (as he then was) in The Commissioners for Her Majesty’s Revenue and Customs v Household Estate Agents Ltd [2008] STC 2045 (“Household”) at paragraph [49]. In that paragraph, Henderson J had noted that “Section 50(6) in its present and earlier incarnations has always been the principal justification for holding that the burden lies on the taxpayer to displace an assessment made within normal time limits”. The same view had been expressed by Dillon LJ in Brady (Inspector of Taxes) v Group Lotus Car Companies plc [1987] STC 635 (“Lotus”) at 639 et seq.;

in the part of his decision in Household dealing with the burden of proof, at paragraphs [43] to [50], Henderson J had held that, once the Respondents had satisfied the court that they were prima facie entitled to make a discovery assessment under paragraph 42 of Schedule 18 to the Finance Act 1998 (“Schedule 18”) (because the conditions in one of paragraphs 43 or 44 of Schedule 18 were satisfied), the burden was on the taxpayer to show that the discovery assessment could not be made because the conditions in paragraph 45 of Schedule 18 were satisfied;

Henderson J had said that he was disinclined to base that conclusion on the fact that any objection to a discovery assessment would need to be by way of an appeal against the relevant assessment and that this therefore brought Section 50(6) of the TMA into play. Instead, he had said that the language in Section 50(6) of the TMA became significant only once it was shown that the relevant assessment had been validly made. If, however, an issue arose as to whether an assessment had been validly made in the first place, then Section 50(6) of the TMA provided no assistance in answering the question of where the burden of proof lay in relation to the alleged invalidity;

he had therefore preferred to base his conclusion on:

therefore, in the present case, the Respondents should bear the burden of proof for the following reasons:

the appeal related to main purpose and, in the context of a similar main purpose test in Oxford Instruments UK 2013 Limited v The Commissioners for Her Majesty’s Revenue and Customs [2019] UKFTT 254 (TC) (“Oxford Instruments”), the First-tier Tribunal had held that, in the absence of any guidance in relation to the burden of proof in the language of the relevant statutory provision in that case, the usual rule applied – see Oxford Instruments at paragraphs [91] to [97];

the fact that paragraph 9(3) of Schedule 1A did not contain the language used in paragraph 9(5) of Schedule 1A (or Section 50(6) of the TMA) to the effect that the relevant closure notice or assessment would stand good if the taxpayer did not establish that it was incorrect was irrelevant because those words were no more than a truism. Their absence therefore could not affect the location of the burden of proof;

in Household at paragraph [49], Henderson J had held that the “true proposition…is that if an assessment is validly made the onus is then on the taxpayer to displace it, in the absence of specific provision to the contrary”. Although Henderson J had said that the principal justification for this was the language in Section 50(6) of the TMA, he had not said that it was the only reason for it. Moreover, in Household at paragraph [47], Henderson J had also identified a policy reason for the taxpayer to bear the burden of proof in a case where the matter in issue depended on the exclusive knowledge of the taxpayer and/or his advisers. That was the case here as the matter in issue depended on the subjective purposes of BLM and SICL and those were not matters of which the Respondents had any direct knowledge;

as for the suggestion that, because the Treaty represented an agreement between two contracting states in relation to the allocation of taxing rights over a particular source of income, the burden of proving that Article 12(5) applied should fall on the relevant contracting state, that went too far. That reasoning would suggest that, in relation to Article 12(1), the Respondents would have the burden of proving that the relevant claimant was not resident in the ROI whereas it was plain that the claimant had the burden of proving residence in the ROI for the purposes of Article 12(1);

paragraph 9(5) of Schedule 1A had been inserted into Schedule 1A by the Finance Act 1996 and therefore two years after paragraph 9(3) of Schedule 1A had been introduced in the Finance 1994. As such, the language used in paragraph 9(5) of Schedule 1A shed no light on the meaning of the language used in paragraph 9(3) of Schedule 1A. It was not as if Parliament was seeking to make any point in omitting the relevant words from the earlier paragraph. Moreover, there was no reason to think that Parliament would have intended the burden of proof under Schedule 1A to be different depending on the type of closure notice which had been issued; and

in any case, the burden of proof was of significance only in cases where the relevant tribunal was left in doubt – see Mustill LJ in Lotus at 644h - and the evidence in this case pointed overwhelmingly toward the conclusion advanced by the Respondents.

the meaning of the phrase “main purpose or one of the main purposes” in the context of Article 12(5);

whether SICL was a “person” concerned with the assignment of the SAAD Claim for the purposes of Article 12(5);

the meaning of the phrase “take advantage” in the context of Article 12(5);

the purpose which it is necessary for a person concerned with an assignment to have before Article 12(5) can be engaged; and

the burden of proof.

reject the proposition that the inevitable and inextricable consequences of an action should be regarded as the sole benchmark for determining the subjective purposes of the person taking that action; and

nevertheless, treat such consequences as forming part of the overall factual matrix to be considered in reaching a conclusion on that question.

there was some discussion at the hearing in relation to whether paragraphs (a) and (b) in Article 32 of the VCLT qualified the phrase “in order to confirm the meaning resulting from the application of article 31” as well as the phrase “in order …to determine the meaning”. We think that it is clear that they do not, not least because it would make no sense to be permitted to confirm a meaning resulting from the application of Article 31 of the VCLT where that meaning leads to a manifestly absurd or unreasonable result. It therefore follows that, in our view, in order to rely on Article 32 of the VCLT, there is no need for BLM to be able to establish that the meaning resulting from the application of Article 31 of the VCLT is ambiguous or obscure or leads to a manifestly absurd or unreasonable result. It is sufficient for BLM merely to be seeking to confirm the meaning resulting from the application of Article 31 of the VCLT;

Article 32 of the VCLT refers in general to “supplementary means of interpretation” and then specifies that that includes “the preparatory work of the treaty and the circumstances of its conclusion”. This raises the question of whether, despite the fact that they are not specifically listed as such, unilateral statements made on behalf of one of the contracting states to a treaty may nevertheless be a permissible “supplementary means of interpretation”;

the decision of Patten LJ in Irish Bank suggests that they are not. We recognise that the question at issue in Irish Bank in this regard was solely the extent to which the practice of the Respondents could be taken into account in interpreting a treaty and the case is direct authority for the proposition that it cannot. However, we think that the observations made by Patten LJ in Irish Bank at paragraph [23] in answering that question suggest that unilateral statements made by one of the contracting states as to the meaning of the treaty should not be regarded as permissible aids for the same reason. In that paragraph, Patten LJ noted as follows:

“the Appellants have been unable to identify any established principle of international law which recognises the unilateral practice of a contracting state as an aid to the construction of a treaty. In that respect, there is no divergence between international law and the English private law system which has never received evidence of what a party to a contract believed the language of the agreement meant except in relation to a claim for rectification. The legal meaning of the words used is an abstract question of law to be determined on an objective basis”;

we infer from this statement that, by parity of reasoning with that applicable in the case of English private contract law, a unilateral statement made by one of the parties to a treaty is not a permissible aid to interpreting the treaty. It follows from that conclusion that such statements should not be regarded as falling within the phrase “supplementary means of interpretation” in paragraph 32 of the VCLT. Instead, that phrase should be limited in its scope to evidence which is indicative of bilateral agreement between the parties, of which preparatory work and the circumstances of the treaty’s conclusion are two non-exhaustive examples;

Mr Grodzinski suggested that this conclusion sits a little uneasily with the fact that the OECD and UN commentaries have long been accepted as permissible aids to the construction of treaties. He pointed out that, on the basis of the reasoning of Patten LJ in Irish Bank, that would be the equivalent of accepting that comments made by a third party as to the meaning of an English private law contract are a permissible aid to the interpretation of that contract. However, we think that Mr Grodzinski’s analogy is imperfect in that the parties to a bilateral treaty in the form of the OECD or UN model have signed up to, and thereby implicitly agreed, the form of the commentaries to the model. The commentaries are not wholly extraneous to the treaty in the same way as would be a third party’s commentary on an English private law contract;

in the absence of Article 32 of the VCLT’s applying, we do not see how the principle laid down in Hart can be engaged, even if it could be said that Article 12(5) was ambiguous or obscure (which we do not think that it is). The principle laid down in Hart applies in the context of interpreting UK domestic law. It is not applicable in the context of construing a treaty;

the decision in Fowler FTT is not binding on us but, in any event, we agree with Mr Brinsmead-Stockham that the First-tier Tribunal in that case had recourse to Hansard not as an aid to the interpretation of the relevant treaty as such but rather as an aid to the interpretation of a provision of UK domestic law which it thought might be helpful to its interpretation of the treaty. As such, it has no impact on the question we are considering; and

finally, we agree with Mr Brinsmead-Stockham that Article 3(2) of the Treaty is not an alternative gateway pursuant to which the statements and the explanatory note may be prayed in aid in interpreting Article 12(5). This is because Article 3(2) of the Treaty is confined in its application to specific terms which are not defined in the treaty and where the relevant specific term has a particular meaning under the domestic laws of one of the contracting states. In this case, even if the relevant material could properly be described as going to the meaning of a specific term – such as, for example, the phrase “take advantage” – as opposed to the meaning of Article 12(5) as a whole, to which we consider it does in fact relate, the phrase “take advantage” has no distinct and special meaning as a matter of UK domestic law.

in order for a person to fall within the ambit of the article, it is necessary for that person to have a main purpose of “taking advantage” of Article 12(1) specifically; and

a person cannot have a main purpose of “taking advantage” of a provision of UK domestic law or a treaty which it has not specifically identified.

Henderson J in Household noted that the principal justification for holding that the burden of proof in displacing an assessment falls on the taxpayer is the language used in Section 50(6) of the TMA to the effect that the assessment must stand good if the tribunal decides not to reduce it. That language is missing from paragraph 9(3) of Schedule 1A. Whilst Henderson J acknowledged that the language used in Section 50(6) of the TMA was of no assistance in a dispute over the validity of an assessment (as opposed to a dispute over the quantum of an assessment) (see Household at paragraph [49]), the present case does not relate to the validity of the closure notice but simply relates to whether or not the amendment made by the closure notice was correct. As such, applying the reasoning of Henderson J in Household, the omission from paragraph 9(3) of Schedule 1A of the language to the effect that the assessment must stand good if the tribunal decides not to reduce it is, in our view, significant;

that omission is all the more conspicuous by the appearance of that language in paragraph 9(5) of Schedule 1A, although the fact that paragraph 9(5) of Schedule 1A was introduced at a later time than paragraph 9(3) of Schedule 1A means that we would not wish to make too much of this distinction;

in Household, where the language in question was also missing from the provision of the legislation which was in issue in that case but the case related to the validity of the assessment, Henderson J based his conclusion on:

applying the reasoning set out in paragraph 151(3)(a) above, the structure and wording of Article 12(5) very strongly points to the conclusion that it is the Respondents who have the burden of proof in this matter. On the language of the article, it is they who are asserting that a person concerned with the assignment had the requisite bad main purpose or, as Henderson J put it, who are “substantially asserting the affirmative of the issue”. The article could have been phrased in such a way that it was for the person claiming the benefit of Article 12(1) to “substantially assert the affirmative of the issue”. In other words, relief under Article 12(1) could have been expressed to be subject to the absence of the requisite bad main purpose in each person concerned with the assignment. Had that been the case, then, notwithstanding the absence from paragraph 9(3) of Schedule 1A of the language which appears in Section 50(6) of the TMA, the position might have been more finely-balanced. However, as it stands, both the omission from paragraph 9(3) of Schedule 1A of the language which appears in Section 50(6) of the TMA and the manner in which Article 12(5) is worded point in the same direction, which is that the burden of proof in this case is on the Respondents;

the above is sufficient to dispose of this question in favour of BLM but we should note that the arguments in relation to how policy should affect the answer are quite finely-balanced. We see some force in Mr Grodzinski’s submission that the person who is the claimant under Article 12(1) has no ability to access the subjective purposes of other persons who are concerned with the assignment whereas the Respondents, as a tax authority, are able to use their information-gathering powers to do that. On the other hand, we think that, so far as the subjective purposes of the claimant itself are concerned, it is the claimant who has the more direct knowledge of its purposes than do the Respondents. We therefore think that the policy arguments go both ways and we decline to decide the question on that basis; and

we are also unwilling to place any reliance in deciding the question on Mr Grodzinski’s argument that, as the Treaty represents an agreement between two contracting states in relation to the allocation of taxing rights over a particular source of income, the burden of proving that Article 12(5) applies should fall on the relevant contracting state. We agree with Mr Brinsmead-Stockham that, on the basis of that reasoning, in a case where Article 12(5) does not apply, it would be up to the Respondents to show that a claimant was not entitled to the benefit of Article 12(1) and that is patently not the case. We therefore consider that this argument does not advance the position either way.

the evidence of Ms Gibbons had made it abundantly clear that, in deciding to purchase claims in the LBIE administration, BLM did not have in mind that it was taking advantage of Article 12(1) in any negative sense. Instead, BLM was simply benefiting from the tax attributes of its being resident in the ROI for tax purposes;

Ms Gibbons was aware, at some level, that UK withholding tax was the reason why SICL wished to dispose of the SAAD Claim but that was the limit of her thinking on that subject and had had no impact on the amount which she was willing to offer for the SAAD Claim;

it was important to view the acquisition of the SAAD Claim in context, which is that it was part of a huge number of acquisitions of claims in the LBIE administration dating back over a number of years. The acquisition of the SAAD Claim should not be viewed as if it had taken place in isolation. In respect of each of those acquisitions, BLM’s focus had been on how much it expected to receive by way of principal and/or interest, taking into account a variety of different risks including the “liquidation lacuna” risk, the “late termination” risk and, at an earlier stage in the process, the currency conversion risk. What it had not focused on in any of those transactions was the characteristics (including the tax residence) of its seller;

so far as this particular transaction was concerned, the price for the sale was dictated solely by the attributes of the SAAD Claim, as the relevant asset, and not the attributes of BLM, as the purchaser. Ms Gibbons had made it clear that the drivers for the transaction were the expected IRR (including the timing of the payment of interest and the UK withholding tax repayment) and the “late termination” risk and that her focus was on what she would need to offer for the SAAD Claim in order for BLM to achieve its desired commercial return whilst not leading Jefferies to offer the debt to other potential purchasers. Whilst she was clearly aware of the potential UK withholding tax risk for SICL, that did not factor into her thinking as to what to bid. The email from Ms Donovan referred to in paragraph 79(15)(c) above was indicative of the focus of BLM in deciding to make the acquisition – there was no mention of UK withholding tax and the email dealt instead with the interest accrual and the “late termination” risk. Likewise, the focus of the financial model was on the timing of receipts and, to some extent, on the “late termination” risk as these were matters which went to the IRR and the downside risk for BLM;

at the time when the price and the main commercial terms of the transaction had been agreed, on or before 8 February 2018, SICL was unaware of the identity of the end-purchaser. If SICL had wished to negotiate a price which was dependent on the tax-advantageous position of its end-purchaser, then it would have insisted on knowing the identity of the end-purchaser before it agreed the price;

the fact that clause 9 of the assignment between SICL and Jefferies allowed Jefferies to on-sell the SAAD Claim to whomever it wished was also inconsistent with the proposition that SICL was wishing to negotiate a price which depended on the tax-advantageous position of the end-purchaser; and

finally, the terms of the report by the SICL Liquidators following the transaction made no mention of the fact that SICL had managed to achieve a sale price for the SAAD Claim which reflected the UK withholding exemption available to its purchaser under an applicable treaty. The main thrust of the report was that the sale had enabled SICL to avoid the “liquidation lacuna” risk which would have meant that it received nothing in respect of the interest. Moreover, when the UK withholding tax risk was mentioned, it was in the context that SICL was potentially subject to UK withholding tax because it was not resident in the UK. It was not in the context that SICL was potentially subject to UK withholding tax because it was resident in the Cayman Islands which did not have a treaty with the UK conferring full exemption from UK withholding tax on UK source interest.

the documentary record was crucial, given that, on a number of points, Ms Gibbons’ oral evidence was that she was trying to reconstruct her state of mind at the time by reference to the documentary record;

Ms Gibbons had had access to tax advice both externally and internally within DKCM and she was a highly-sophisticated investor;

by the time of the transaction, many of the risks which had previously been associated with LBIE claims had been resolved. The only risks apart from the timing of the interest payment and the repayment of the UK withholding tax were the “liquidation lacuna” risk – which Ms Gibbons had effectively discounted and which was not factored into the financial model at all – and the “late termination” risk – which was considered to be minimal in the light of the imminent Scheme of Arrangement and was shown in the financial model only as a “bookend” possible downside; and

there was very little negotiation over the purchase price in the transaction. It appeared that DKCM offered what it thought to be a fair price and that was accepted by Jefferies without much back and forth.

the principal amount of the SAAD Claim made it the largest individual LBIE claim which BLM had purchased;

the SAAD Claim was the last of the LBIE claims which BLM had bought;

the SAAD Claim was the only LBIE claim which had been purchased by BLM following the date of the Court of Appeal decision in the Withholding Tax Litigation to the effect that the interest on LBIE claims was annual in nature and therefore subject to UK withholding tax in the absence of any applicable exception;

the SAAD Claim was the only Category 2 claim which had been acquired on a “back-to-back” basis from an end-seller which was not entitled to a full exemption from UK withholding tax under the terms of an applicable treaty; and

the 1.38% spread made by Jefferies between its buy and sale prices in relation to the SAAD Claim was considerably greater than the 0.25% which Ms Gibbons in her evidence had accepted would typically be the commission on the sale of a plain vanilla debt claim.

from the economic perspective, the principal amount of the SAAD Claim was irrelevant as the principal amount had already been discharged long before the purchase of the SAAD Claim by BLM. As such, the appropriate comparator in this case was not the principal amount but the interest amount and that meant that the SAAD Claim was not the largest individual LBIE claim which BLM had purchased;

in any event, the purchases of LBIE claims made by BLM in the LACA had been bulk purchases - where the aggregate amount paid by BLM in respect of certain purchases had exceeded the amount paid by BLM for the SAAD Claim - and the evidence of Ms Gibbons was that she saw no difference in economic terms between the purchase of a single LBIE claim and the purchase of multiple LBIE claims for a single purchase price;

Ms Gibbons had also testified that there were very few LBIE claims in the market at the time when the Appellant purchased the SAAD Claim and that this was because:

thus, the fact that the SAAD Claim was the last of the LBIE claims which had been purchased by the Appellant and the only LBIE claim which had been purchased by the Appellant following the date of the Court of Appeal decision in the Withholding Tax Litigation was entirely attributable to the state of the market in LBIE claims at the time when the SAAD Claim was purchased and had nothing to do with the fact that the interest payable under the SAAD Claim was going to be subject to UK withholding tax. Ms Gibbons had testified that she never thought about the tax-status of the seller when she was making purchases;

the reason why the SAAD Claim was the only Category 2 claim which had been acquired on a “back-to-back” basis from an end-seller which was not entitled to a full exemption from UK withholding tax under the terms of an applicable treaty was that most Category 2 claims had been purchased in the LACA, which involved direct acquisitions and not “back-to-back” sales. Moreover, it was not clear why the fact that the SAAD Claim had been acquired on a “back-to-back” basis as opposed to being acquired directly was relevant. BLM had made many direct acquisitions of Category 2 claims from sellers which were not entitled to a full exemption from UK withholding tax under the terms of an applicable treaty in the course of its substantial dealings in LBIE claims over the years and Ms Gibbons had testified that she viewed assets in the LBIE administration as cash flows and didn’t distinguish between Category 1 claims and Category 2 claims in that regard; and

the evidence of Ms Gibbons had been that, whilst the figure of 0.25% was generally indicative of commissions on the sales of plain vanilla debts, higher commissions were often paid in connection with such disposals, including one of 3% in relation to one of the transactions with which she had been concerned. Moreover, BLM had been unaware of the spread made by Jefferies in connection with the transaction until much later.

FACT - at the time when the opportunity to acquire the SAAD Claim arose, BLM was well aware of the fact that, in the absence of a sale, the SAAD Claim had a value to SICL of approximately 80% of the interest which was payable in respect of the SAAD Claim because of the likelihood of SICL’s suffering irrecoverable UK withholding tax when the interest in respect of the SAAD Claim was paid. We say “approximately 80%” in this context because, at the time when the opportunity arose, the SAAD Claim was subject to three significant uncertainties as follows:

BLM knew that each of the above risks would have an impact on the precise value to be accorded to the SAAD Claim in SICL’s hands should it choose to retain the SAAD Claim. BLM also knew that, whatever the precise value of the SAAD Claim to SICL were it to retain the SAAD Claim, SICL was likely to be willing to sell for the best price which it could get in excess of that value.

REASON - the fact that BLM was aware of SICL’s position vis-à-vis the UK withholding tax can be seen from the fact that:

FACT - at the same time, BLM was aware that the SAAD Claim had a value to BLM of approximately 100% of the interest which was payable in respect of the SAAD Claim because of BLM’s ability to reclaim any UK withholding tax which might arise in respect of that interest in the event that the Supreme Court were to uphold the Court of Appeal’s decision in the Withholding Tax Litigation. Again, we say “approximately 100%” in this context because BLM faced the same “liquidation lacuna” and “late termination” risks as SICL in relation to the SAAD Claim and it was also aware that there was likely to be a delay of some nature both in its receiving the interest from the LBIE Administrators and in its receiving repayment of the UK withholding tax from the Respondents if the Supreme Court did not reverse the Court of Appeal in the Withholding Tax Litigation.

REASON – the fact that the SAAD Claim had this value to BLM can be seen from:

FACT – at the time when BLM made its offer for the SAAD Claim, BLM was aware that it was not the only prospective purchaser of the SAAD Claim for whom the UK withholding tax on the interest payable in respect of the SAAD Claim would not be a permanent cost.

REASON – the evidence of Ms Gibbons was that, if her offer price was too low, then Jefferies would have started to look for other prospective purchasers – see paragraph 80(12) above. It is possible to infer from that, and the price that was actually offered by DKCM on behalf of BLM, that Ms Gibbons knew that there were other prospective purchasers in the market for whom the UK withholding tax would not be a permanent cost; and

FACT - when it came to setting the price that it was prepared to offer for the SAAD Claim, BLM took into account the following:

REASON - the fact that the primary driver of the purchase price paid by BLM for the SAAD Claim was the cash flows to which it was expected to give rise and that BLM was focused on achieving an appropriate IRR and avoiding too significant a downside in the event that the “late termination” risk materialised can be seen from:

As regards the price which BLM considered that SICL was likely to accept, although one of the factors which BLM took into account when setting its offer price was the fact that the SAAD Claim, were it to be retained by SICL, would have a much lower value to SICL than it had to BLM, BLM also knew that the value of the SAAD Claim to SICL was much greater than that retention value because of the number of people in the market for whom the UK withholding tax would not be a permanent cost. That fact would inform the price which those potential purchasers were prepared to pay and, hence, would inform the SICL Liquidators’ knowledge of the value of the SAAD Claim to SICL when it came to their agreeing a price for the SAAD Claim.

BLM knew that, although Jefferies had not initially marketed the SAAD Claim to other potential purchasers in the market, Jefferies would do so if BLM sought to “low-ball” it in this respect. This thinking can be seen in the evidence of Ms Gibbons to that effect – see paragraph 80(12) above – and, in any event, it can be reasonably inferred from the fact that both the SICL Liquidators and Jefferies would have been interested in maximising the value realised by SICL for the SAAD Claim (and, hence, maximising Jefferies’ commission) that this is what Jefferies would have done.

FACT - at the time when it was looking to realise the SAAD Claim, SICL was aware that, in the absence of a sale, the SAAD Claim had a value to SICL of approximately 80% of the interest which was payable in respect of the SAAD Claim because of the likelihood of its suffering irrecoverable UK withholding tax when the interest in respect of the SAAD Claim was paid. The factors which informed that value were the same as those set out in paragraph 164(1) above.

REASON - the fact that the SICL Liquidators were aware of the potential exposure to UK withholding tax on interest in respect of the SAAD Claim were they to retain the SAAD Claim can be seen in:

FACT - at the same time, SICL was aware that there were people in the market for whom UK withholding tax would not be a permanent cost. Those people included UK resident corporates and residents of jurisdictions with treaties that provided for a full exemption from UK withholding tax on interest. SICL knew that such people would be prepared to pay more for the SAAD Claim than the value which the SAAD Claim had to SICL in the absence of a sale.

REASON - the fact that, at the time of employing Jefferies to sell the SAAD Claim, the SICL Liquidators were aware that there were purchasers in the market to whom the SAAD Claim might have a greater value than it did to SICL can be seen in the emails from Ms Nettleton to Mr Sabharwal of 12 December 2017 and 2 January 2018 – see paragraphs 79(2) and 79(6) above;

FACT - when it came to agreeing the price for its sale of the SAAD Claim to Jefferies, on or before 8 February 2018, SICL wanted to realise the greatest value possible for the SAAD Claim bearing in mind the potential market.

REASON - the fact that the SICL Liquidators wanted to realise the greatest possible value from the SAAD Claim can be inferred from the fact that SICL was in liquidation and the SICL Liquidators had a duty to the creditors in, and the shareholders of, SICL to realise the assets of SICL at their greatest possible value in the liquidation;

FACT - at the time when it agreed in principle to the terms and price for its sale of the SAAD Claim, on or before 8 February 2018, SICL did not know, and did not care about, the identity of the end-purchaser. All that it knew and cared about was that the terms and price which had been offered to it were acceptable to it.

REASON - the fact that, at the time when it agreed in principle to the terms and price for its sale of the SAAD Claim, SICL did not know the identity of the end-purchaser can be inferred from the fact that the SICL Liquidators did not ask about the identity of the end-purchaser until after they had reached that agreement (see Ms Watson’s email to Mr Sabharwal and others of 9 February 2018 referred to in paragraph 79(17) above). The fact that, at the time when it agreed in principle to the terms and price for its sale of the SAAD Claim, SICL did not care about the identity of the end-purchaser can be inferred from the fact that the SICL Liquidators agreed to those terms on or before 8 February 2018, before it knew the identity of the end-purchaser – see the email from Ms Gibbons to her colleagues referred to in paragraph 79(16) above.

At that point, although the SICL Liquidators could not yet formally enter into the trade because they considered that they needed to be in the Cayman Islands to do that, they considered that they were morally bound to proceed at the agreed price. That was the nature of the assurance which the SICL Liquidators gave to Jefferies and which Jefferies, in turn, reported to Ms Watson, as recorded in Ms Watson’s email to Mr Sabharwal and others of 9 February 2018 – see paragraph 79(17) above;

FACT - the reason why SICL asked Jefferies, after it had agreed in principle to the terms and price for the sale of the SAAD Claim, for the identity of the end-purchaser had nothing whatsoever to do with checking on the tax characteristics of the end-purchaser with a view to re-negotiating the price of the transaction or walking away from the transaction.

REASON - the fact that the SICL Liquidators’ request to know the identity of the end-purchaser prior to executing the transaction had nothing to do with the tax characteristics of the end-purchaser and was not made with a view to re-negotiating the price of the transaction or walking away from the transaction can be inferred on the balance of probabilities from the reported statement and conduct of the SICL Liquidators.

It is clear from Ms Watson’s email to Mr Sabharwal and others of 9 February 2018 that the SICL Liquidators considered the price for the sale to have been finally agreed no matter the identity of the end-purchaser – see paragraph 79(17) above. As such, we consider that their insistence on knowing the name of the end-purchaser at that point must have been motivated by reasons unrelated to tax. In our view, the request was almost certainly motivated solely by a desire for good governance or to comply with regulatory requirements.

We can certainly see no basis for the allegation made by Mr Brinsmead-Stockham in his cross-examination of Ms Gibbons at the hearing to the effect that the enquiry was made with a view to confirming that the tax status of the end-purchaser had been adequately reflected in the price which the SICL Liquidators had already agreed, with a view to their re-opening price negotiations in the event that it hadn’t.

Mr Brinsmead-Stockham’s suggestion was based on the fact that the SICL Liquidators were reported by Ms Watson as saying merely that “the buyer’s identity would change nothing” and not that the “buyer’s identity and tax residence would change nothing”. He said that this meant that the SICL Liquidators were reserving the right to increase the price if they discovered that the end-purchaser’s tax residence had not been adequately factored into the offer price. Despite Mr Brinsmead-Stockham’s considerable forensic skill, we can see no justification for that reading of the relevant statement;

FACT - as a result of making that request and receiving the response to it, by the time that SICL formally entered into the trade with Jefferies on 12 February 2018, it was aware that the end-purchaser of the SAAD Claim was BLM and that BLM was resident in the ROI for tax purposes.

REASON - the fact that, by the time that they formally entered into the trade with Jefferies on 12 February 2018, the SICL Liquidators were aware that the end-purchaser of the SAAD Claim was BLM and that BLM was resident in the ROI for tax purposes can be inferred from the fact that:

– see paragraphs 79(17) and 79(18) above.

We do not think that the exchange between Ms Watson and Mr Miesner which we have recorded in paragraphs 79(19) and 79(20) above should be construed as evidence that BLM’s identity as the end-purchaser had not yet been conveyed to Jefferies (and, hence, the SICL Liquidators) prior to the formal entry into the trades. In our view, a more likely explanation for the terms of that exchange are that Mr Miesner had temporarily forgotten that DKCM were acting for BLM and not for itself as principal when he sent his email confirming the terms of the trade. That is certainly a much more plausible interpretation of the events which transpired than that the SICL Liquidators, after making it plain that they would not enter into the trade without knowing the identity of the end-purchaser, simply forgot about that request and proceeded to enter into the trade without knowing that identity; and

FACT – once SICL had been informed, immediately prior to its entering into the trade with Jefferies on 12 February 2018, that BLM was the end-purchaser, there was no prospect that Jefferies would on-sell the SAAD Claim to anyone other than BLM.

REASON – the disposal of the SAAD Claim was always intended to take the form of a “back-to-back” transaction. Once Jefferies had agreed in principle with BLM that it would on-sell the SAAD Claim to BLM and informed the SICL Liquidators that BLM was the end-purchaser, Jefferies was, for all practical purposes, limited in its choice of end-purchaser to BLM. In any event, Jefferies entered into the trade with BLM on 12 February 2018 immediately after it entered into the trade with SICL on the same day. Thus, by the time that SICL transferred the SAAD Claim to Jefferies on 9 March 2018 in completion of the trade between those parties, Jefferies was contractually bound to transfer the SAAD Claim to BLM in completion of the trade between Jefferies and BLM. As such, the freedom apparently conferred on Jefferies by clause 9 of the assignment of claim agreement to transfer the SAAD Claim to someone other than BLM was illusory.

the principal amount of the SAAD Claim had been discharged long before the assignments were made; and

the commercial substance of the assignments was that they simply involved the transfer of the interest which was payable in respect of the SAAD Claim and not the right to receive the repayment of principal.

the way in which each assignment was worded – as an assignment of the whole SAAD Claim but subject to the retention by the relevant seller of the prior payment in respect of the principal; and

the language used in each notice of assignment to the LBIE Administrators – which referred to the subject matter of the relevant assignment as being the whole SAAD Claim,

the burden of proof in this case is on the Respondents;

SICL is a relevant “person” for the purposes of Article 12(5);

accordingly, the Respondents need to show that either BLM or SICL (or both) had, as its main purpose or one of its main purposes, “taking advantage” of Article 12(1) by means of the assignment of the SAAD Claim;

the main purposes of each of BLM and SICL in relation to the assignment of the SAAD Claim are to be determined subjectively in the light of all of the evidence, including (but not exclusively) inevitable and inexorable consequences;

it is not necessary in order for Article 12(5) to apply for there to have been any artificial steps or arrangements. It is merely necessary for one or both of BLM and SICL to have had a main purpose of “taking advantage” of Article 12(1) by means of the assignment of the SAAD Claim; and

so far as the main purposes of SICL are concerned, it is insufficient in order for Article 12(5) to apply for SICL to have had a main purpose of “taking advantage” of a provision of UK domestic law or a treaty which was unknown to SICL. Instead, in order for Article 12(5) to be capable of applying, SICL needed to have had a main purpose of “taking advantage” of Article 12(1) specifically.

did BLM have, as a main purpose in the assignment of the SAAD Claim, “taking advantage” of Article 12(1) itself by means of the assignment?

did BLM have, as a main purpose in the assignment of the SAAD Claim, enabling SICL to “take advantage” of Article 12(1) by means of the assignment?

did SICL have, as a main purpose in the assignment of the SAAD Claim, “taking advantage” of Article 12(1) itself by means of the assignment? and finally

did SICL have, as a main purpose in the assignment of the SAAD Claim, enabling BLM to “take advantage” of Article 12(1) by means of the assignment?

both parties knew that SICL could not benefit from Article 12(1); and

SICL received a price for the SAAD Claim which reflected the significant spread between the amount which it would have received in respect of the SAAD Claim had it retained the SAAD Claim and the amount which BLM received in respect of the SAAD Claim following the acquisition.

it was one of BLM’s main purposes to enable SICL to “take advantage” of Article 12(1); or

it was one of SICL’s main purposes to “take advantage” of Article 12(1) itself.

it obtained the desired IRR;

it faced an acceptable downside exposure from the “late termination” risk should that risk materialise; and

it staved off other potential bidders by offering a sensible price.

the SICL Liquidators agreed on the price in principle before they thought to enquire about the identity of the end-purchaser; and

the SICL Liquidators enquired about the identity of the end-purchaser after that not because they had any interest in the tax attributes of the end-purchaser but solely out of a desire for good governance or to comply with regulatory requirements.

a UK resident who was exempt from UK tax altogether or who was able to receive the interest without UK withholding tax and who had significant losses against which to offset the interest; or

a person resident in a jurisdiction other than the ROI with which the UK had concluded a treaty conferring on residents within the relevant jurisdiction a full exemption from UK withholding tax.

the SICL Liquidators’ only purpose in entering into the transaction was to realise the SAAD Claim for the best possible price;

the SICL Liquidators knew at the time when they agreed in principle to the transaction that their end-purchaser must have an entitlement to an exemption of some sort from UK withholding tax; and

by the time that they formally entered into the transaction, the SICL Liquidators knew that, in fact, the end-purchaser was relying on Article 12(1) for its exemption from UK withholding tax,

one of the SICL Liquidators’ main purposes in entering into the transaction was to “take advantage” of Article 12(1) itself?

there is a sale of a debt, the interest on which is subject to UK withholding tax, from a seller resident in a non-treaty jurisdiction to a purchaser resident in a treaty jurisdiction;

the sale price is a market price reflecting the fact that there are parties in the market who are able to benefit from a domestic exemption from UK withholding tax or from being treaty resident and are thereby able to avoid suffering the UK withholding tax as a permanent cost; and

the purchaser is one of those parties and the seller happens to be aware of the identity and tax residence of the purchaser,

there were people in the market with different tax attributes;

those different tax attributes were reflected in the market price of the debt claim which was the subject of the transaction; and

the seller happened to know the reason why its purchaser had the necessary tax attributes to be able to pay the market price of the debt claim.

taking into account all of the evidence, and viewed subjectively, SICL did not have, as one of its main purposes, enabling BLM to “take advantage” of Article 12(1); and

in any event, as noted in paragraph 171 above, this is not a relevant question, based on the language of Article 12(5).

the Respondents’ refusal to repay UK income tax withheld on payments of interest made by the LBIE Administrators in respect of the English Law Participation Claims – see paragraphs 67(2) and 68(2) above; and

the Respondents’ decision to impose a Penalty in respect of the “Mackay Shields claim” – see paragraph 65 above.