Khan v Revenue & Customs (Rev 1)

This is a cautionary tale, which illustrates all too graphically the importance of seeking specialist tax advice before entering into commercial arrangements that might have adverse tax consequences, however remote that risk might appear.

It is impossible not to feel some sympathy for the appellant, Mr Khan, who was held liable to pay income tax of almost £600,000 on £1.95 million that was paid into his bank account and then paid out again almost immediately in respect of connected share sale and buy-back transactions. Yet he found himself in that situation because he relied upon an assumption that the consequence of his having to expend the £1.95 million as soon as it was received, was that the persons to whom he paid it would be liable to pay tax on it instead of him.

HMRC regarded the assumption as ill-founded, and both the First-Tier Tribunal (“FTT”) and Upper Tribunal (Tax and Chancery Chamber) (“UT”) supported HMRC’s interpretation of the relevant statutory provisions and their application to the facts. In a decision promulgated on 14 January 2020, [2020] UKUT 0168 (TCC), the UT decided that Mr Khan was liable to pay tax on the £1.95 million under s.385(1)(b) of the Income Tax (Trading and other Income) Act 2005 (“ITTOIA”). He appeals against that decision with the permission of Nugee LJ.

Despite the valiant efforts of Mr Laurent Sykes QC to persuade us that the UT erred in law, for the reasons set out below, I have concluded that it reached the correct conclusion. Unfortunately for Mr Khan, his assumption has proved costly.

Section 383(1) of ITTOIA provides that income tax is charged on dividends and other distributions of a UK resident company. “Distribution” in this context is defined by section 1000 of the Corporation Tax Act 2010 (“CTA”) in these terms:

Any dividend paid by the company, including a capital dividend.

Section 385(1) ITTOIA identifies who is liable to pay the income tax chargeable on a distribution under s.383(1). It provides that:

“(1)

The person liable for any tax charged under this Chapter is –

(a)

the person to whom the distribution is made or is treated as made (see Part 6 of ICTA and sections 386(3), 389(3) and 396A), or

(b)

the person receiving or entitled to the distribution.” The present case only concerns s.385(1)(b).

The phrase, “the person receiving or entitled to” is used in many other provisions of ITTOIA (e.g. sections 332, 338, 352, 371, 404, 481, 581, 611 and 616) to denote the person who is liable to pay the tax that is chargeable on the income, benefit, or other payment treated as income with which the corresponding charging provision is concerned. Most of these sections are derived from earlier tax legislation, most recently Schedule D to the Income and Corporation Taxes Act 1988 (“ICTA”).

However, that earlier legislation did not expressly state who was liable for a charge to tax on distributions of a UK resident company. This appears to be because the relevant charging provisions historically fell under Schedule F rather than Schedule D. Instead, as the UT put it: “the implication that the person liable was the person receiving or entitled to the distribution had to be found in a patchwork of other provisions that were being rewritten in ITTOIA”. Those included the provisions of ICTA which identified the persons who would be entitled to claim tax credits in respect of certain qualifying distributions (and thus, by necessary implication, would otherwise be liable to pay the tax). As the Explanatory Notes to s.385 state at paragraph 1558, section 231(4) of ICTA suggested that:

“where the distribution actually belongsto someone other than the recipient, or where under any provision of the Tax Acts it is treated as belonging to someone other than the recipient, that other person is liable for the tax charged.” (Emphasis supplied).

That indicates that the person “entitled to” the taxable income means the person to whom that income belongs. In the case of a dividend or distribution by a company out of its profits, that will usually be the owner of the shares.

Section 385(1) of ITTOIA was introduced to remedy the omission, and made it clear that the correct inference had been drawn as to the identity of the persons who were liable to pay the tax. The Explanatory Notes go on to explain, at paragraph 1564, that the new provision stating who is liable for any tax charged on distributions from UK resident companies needed to cover all bar one of the situations addressed specifically in ICTA (and replicated in ITTOIA) namely, (i) the person to whom a distribution is made or to whom it is treated as being made; (ii) the person receiving a distribution; and (iii) the person entitled to the distribution. It did not need to address the situation where a distribution is treated by statute as the income of a person other than the recipient, because it was to be expected that the legislation which did so would also make express provision for who is liable to pay the tax.

The person who receives the distribution which is treated as income for taxation purposes may also be the person entitled to it, and often will be. However, since s.385(1)(b) is couched in terms which allow the tax to be charged to the recipient of the distribution or the person entitled to it, it follows that either receipt or entitlement

Mr Charles Bradley, who appeared for HMRC in this appeal, rightly accepted that the phrase “the person(s) receiving or entitled to” must be given a consistent meaning wherever it appears in ITTOIA. This reading of s.385(1)(b) is consistent with the way in which the phrase has been interpreted when it appears in other provisions denoting the persons liable to be charged to income tax.

Thus for example in Aplin v White [1973] 1 WLR 1311 the relevant provision was

Timpson’s Executors v Yerbury [1936] 1 KB 645 is an example of the alternative scenario in which the person entitled to taxable income does not receive it. In that case, the relevant provision was rule 1 of the Miscellaneous Rules applicable to Schedule D to the Income Tax 1918, which defined the persons on whom tax under Schedule D was to be charged as being “the persons … receiving or entitled to the income” in respect of which the tax was to be charged. Mrs Timpson, the English domiciled beneficiary of a New York trust who was entitled to have the net income of the trust applied to her use, directed the Trustee to use that income to pay allowances to her children, which were remitted to the UK and paid to the children or their bankers directly.

It was held by the Court of Appeal that although the payments were not received by Mrs Timpson in the UK, and she had disposed of the money to her children by way of gift, it did not become their money until they received each payment. Therefore she was a person “entitled to the income” and chargeable to tax on it when it came into the UK. They rejected the argument by her executors that she had made an equitable assignment of the money to the children in New York when she instructed the Trustee to make the payments.

Although it is clear that either “receipt of” or “entitlement to” the distribution will suffice to make an individual liable to pay tax on it, one of the issues that arises in the present case is whether either of these concepts carries with it, by necessary implication, a requirement of an element of control over the money or real benefit from its receipt.

Mr Khan is an accountant. Since the mid-1990s he had prepared the management accounts of a company named Computer Aided Design Ltd (“the Company”) which also rented its premises from him. The Company carried on business as an employment bureau for consultants. In June 2013, when the Company had distributable reserves of £1.95 million, but its profitability was in sharp decline, the three shareholders decided that they wanted to extract all available funds from the Company and then go their separate ways. To that end, they entered into negotiations with Mr Khan with a view to his acquiring the Company.

It was envisaged that following that acquisition, Mr Khan would use his know-how to effect an efficient and orderly winding-up over two to three years, thereby avoiding the immediate redundancy of the Company’s four remaining employees. The attraction of this proposition from Mr Khan’s perspective was that he would benefit from the profits of the short period of continued trading, and any surplus realised on the winding-up, whilst the day-to-day management of the Company would be left to the employees, who had the necessary knowledge of the business.

The proposal was that most of the price sought by the shareholders for their shares would be funded by the Company’s resources; Mr Khan could not have afforded the acquisition otherwise. The shareholders sought to structure the deal in a tax-efficient manner. Negotiations took place at arms’ length, and the shareholders and Mr Khan were legally represented.

The original plan was that the Company would buy back 96 of its 99 issued shares from the shareholders for £1.8 million, and Mr Khan would then buy the remaining 3 shares from them for an amount equivalent to the remaining net asset value (agreed by the parties to be £18,771). However, late in the day the proposed structure was changed. Instead, Mr Khan bought the entire issued share capital of 99 shares from the shareholders for £1.95 million in cash, plus an amount equal to the net book value, i.e. £18,771; then, less than 40 minutes later, the Company bought back from Mr Khan 98 of those shares for cash consideration of £1.95 million, leaving him with one share.

The share sale and buy-back transactions were documented respectively in a share purchase agreement (“the SPA”) and an off-market purchase agreement, (“the OMPA”) each of which completed on 28 June 2013. The relevant sequence of events was as follows:

At a board meeting of the Company at 11.35am the transfer of shares from the shareholders to Mr Khan was approved. Mr Khan was appointed as director with immediate effect, and the shareholders all resigned as directors.

The SPA was executed.

At a board meeting of the Company at 12.05pm, attended by Mr Khan as sole director, the Company decided to purchase 98 of its own shares. The draft OMPA was approved, and Mr Khan was authorised to execute it on behalf of the Company. That meeting was then adjourned to enable a general meeting to take place.

At a general meeting of the Company at 12.10pm, attended by Mr Khan as sole shareholder, a special resolution was passed approving the terms of the OMPA and authorising the Company to enter into it, and an ordinary resolution was passed consenting to the failure of the Company to display the OMPA at its registered office for 15 days prior to the meeting.

The board meeting then resumed, noting that the resolution to approve the OMPA had been duly passed. Mr Khan then executed the OMPA on behalf of the Company.

Most of the cash that was needed to effect the transactions was provided by the Company drawing down an invoice discounting facility made available to it by NatWest. Mr Khan indemnified the vendor shareholders against any liability under that facility and provided NatWest with a personal guarantee. The Company’s bank statement showed the receipt of £1.216 million from NatWest and the payment of

The upshot was that Mr Khan acquired the Company and became its sole shareholder at a personal cost to him of £18,771 (though the price he paid was £1.95 million more than that), and the vendor shareholders between them received a sum equal to the value of the Company’s distributable reserves.

The UT found at [87] that the two transactions (the share sale and buyback) were agreed and implemented as one, having regard to their interdependency and the short time period over which they took place. There was no practical likelihood that the transactions would not have happened together. It further found that “as a matter of practical fact” Mr Khan had no control over the buyback proceeds, though this was only because of the terms of the agreement he had entered into.

The UT also found at [92] that the payment of the £1.95 million by the Company to Mr Khan was a loan to enable him to acquire the shares from the vendor shareholders under the SPA. Although the loan was not documented as such, it was the only way in which that payment could be characterised; it was plainly not a gift and Mr Khan did not hold the money on trust. His obligation to repay that loan was then set off against the Company’s obligation to pay Mr Khan £1.95 million as the price of the 98 shares it was purchasing under the OMPA [96].

It is common ground that the payment made by the Company to Mr Khan on the purchase of its own shares was a “distribution” chargeable to income tax under s.383 of ITTOIA. Unfortunately for Mr Khan, it is not permissible to deduct the (capital) acquisition cost of shares against dividend or distribution income in respect of those shares for the purposes of mitigating an exposure to tax on the latter.

Although the cashflow was all in one direction and there was only one cash transfer made by the Company, it is of importance to note that it was not the earlier advance but rather, the payment of the purchase price for the 98 shares by the Company, effected by the set-off, that constituted the taxable distribution by the Company under the buyback arrangement. The Company had no legal obligation to purchase the 98 shares until Mr Khan executed the OMPA on its behalf, the final step in the sequence of events outlined above.

At the time of the transfer of the £1.95 million into his bank account, and then out of that account to his solicitor, Mr Khan was neither a shareholder in the Company nor a trustee of shares, and therefore he was neither legally nor beneficially entitled to a distribution out of the Company’s assets in respect of those shares. In any event, the parties are bound by the UT’s finding that that transfer was not made under or pursuant to the OMPA. By the time of the implementation of the OMPA, less than 40 minutes later, by which time the money had already left Mr Khan’s bank account, that situation had changed, and he was the legal and beneficial owner of the entire issued share capital in the Company and the only person entitled to such a distribution.

Whilst an exception to the treatment of such payments as a “distribution” taxable as income exists for payments made by an unquoted trading company on the purchase of its own shares under s.1033 of the CTA, provided certain conditions are fulfilled, it is accepted that the conditions for the operation of that section were not met by Mr

Whether all those conditions would have been met by the vendor shareholders if the transactions had been structured in the manner originally envisaged by the parties is a matter of pure speculation. That would have depended, among other matters, on the purpose of the shareholders in selling their shares to the Company. There have been no fact-findings in that regard, and the answer to that question is not strictly relevant to the issue of Mr Khan’s liability that we have to determine on this appeal. For whatever reasons, the vendor shareholders were not prepared to take the risk that s.1033 of the CTA would not apply to a direct buyback of the shares by the Company from them. Mr Khan was unwilling to give them the indemnity they sought from him.

On 5 July 2017, following a tax inquiry, HMRC issued a closure notice in respect of Mr Khan’s self-assessment income tax return for the tax year 2013-2014. The closure notice increased the income tax due for that tax year by £594,814.57 on the basis that the buy-back of shares was a distribution taxable under s.383 of ITTOIA 2005.

Mr Khan appealed unsuccessfully against the closure notice to the FTT on two grounds (one of which is no longer pursued). He then appealed to the UT on the basis that the FTT erred in failing to recognise the true substance of the transaction, namely that it was a single composite transaction whose effect was to make Mr Khan the owner of one share in the Company devoid of its distributable reserves, at a small net cost. For the purposes of s.383 and s.385(1)(b) ITTOIA he contended that, viewed realistically, the persons who received and were entitled to the distribution were the selling shareholders, even though they no longer owned the shares.

Before the UT Mr Khan also contended that, so far as he was concerned, the transaction was taxable under s.687 ITTOIA on the reward for his service in participating in the transaction, which was the value of the single share he received. The UT held that s.687 was not engaged. That section imposes a charge to tax on income “from any source that is not charged to income tax under or as a result of any provision of this Act or any other Act.” Therefore it could not operate if tax was charged on a distribution under s.383. In the present case the distribution was taxable under s.383. There is no appeal against that finding.

The UT found that the FTT had erred in law in failing to construe s.385(1)(b) in the light of its statutory purpose, and that because of this it had failed to make relevant findings of fact. It therefore set aside and re-made the decision. However, after making the fact-findings referred to above, the UT rejected Mr Khan’s argument that he was not the person who was liable to pay the tax under s.383.

The UT began by considering the purpose of the statutory provisions. The judges considered the Explanatory Notes, as they were entitled to, and found that s.385 was merely making express what was implied from the existing law that was being rewritten [53]. They concluded at [56] that it was clear that Parliament has provided that the tax chargeable on any distribution is to be recovered from the actual recipient of the distribution or from someone other than the actual recipient in a case where the distribution is the income of that other person (on the basis that that other person is entitled to the distribution).

The UT noted at [45] and [47] that it was accepted that Mr Khan was the owner and seller of the shares at the time the purchase was made by the Company, and yet he submitted that he did not receive and was not entitled to receive the purchase price for them. It found that the buy-back funds were actually paid to and received by Mr Khan [59].

Having considered the chronology of events on 28 June 2013, the UT pointed out that at the time of the taxable distribution the selling shareholders were no longer shareholders, and Mr Khan did not receive the distribution as a trustee for them. They were not entitled to the distribution, because at the time when the distribution was made they were no longer entitled to the shares; all that they were entitled to was the purchase price for the shares they had sold to Mr Khan ([83]-[84]). Mr Khan’s obligation to repay the loan that the Company had advanced to enable him to buy those shares was “entirely separate” from the Company’s obligation to pay £1.95 million to him as the purchase price of the shares under the buy-back [92].

The UT decided that none of the factors relied on by Mr Khan and which it had accepted at [87] showed that he did not receive the share buyback proceeds or that he was not entitled to them [89]. It pointed out that the statutory question was focused on actual receipt and entitlement to the distribution, not with how the shares were acquired by Mr Khan in order to be able to effect the buy-out, nor with how that acquisition was funded, nor with the fact that the release of funds by the Company was essential to the SPA and the successful implementation of the transactions viewed as a whole. As to the money flows connected to the making of the distribution, the UT correctly observed that: “the monies flow as a result of the imposition of legal

The UT concluded that the proceeds of the share buyback were not to be regarded as belonging to the vendor shareholders; that proposition was unsustainable both legally and factually [95]. The vendor shareholders were entitled to £1.95 million but that was in respect of the sale of their shares to Mr Khan, and not in respect of a distribution which by definition had to (and did) rest with the actual shareholder at the time. The UT therefore dismissed Mr Khan’s appeal.

Mr Sykes submitted that the UT erred in law in refusing to consider the sale and buyback of the shares as a single composite transaction and consider its overall effect rather than concentrating on the machinery by which it was effected (i.e. the legal steps in the chain). In substance and in truth, Mr Khan was no more than a conduit for the selling shareholders to effect the buyback of the 98 shares themselves and his intermediate role in that aspect of the transaction should be ignored. As a matter of practical reality, the 98 shares were never Mr Khan’s to do with as he pleased, nor were the buyback proceeds. He never had the benefit of nor control over the £1.95 million and it was “absurd” to tax him on that sum, all the more so if the selling shareholders were liable to pay CGT on that sum (less their cost of acquiring the shares). Moreover, if the FTT and UT’s approach were correct, it would be all too easy to avoid payment of the tax on the distribution by interposing a person who was not liable to taxation in the UK in the role of Mr Khan.

Mr Sykes submitted that this analysis was unaffected by the fact that this was not a pre-planned artificial tax avoidance scheme, but rather a transaction comprising stages that had been structured in a manner designed to mitigate the vendor shareholders’ tax exposure (in that it ensured that they paid CGT on the gain they made by disposing of their shares to Mr Khan, which is chargeable at a lower rate than income tax). The practical and economic reality did not depend on whether the transaction or its various elements were the subject of arms’ length negotiations.

Nor did it matter that Mr Khan acquired and kept the 99^th share for himself and thereby obtained ownership of and control over the Company. It was important not to confuse the question whether he benefited from the transaction as a whole, with the question whether he benefited from the buyback proceeds; s.385(1)(b) was only concerned with the latter. The distribution of company assets did not cease to be a

I consider that we need not concern ourselves about what might have happened had the vendor shareholders decided to structure the deal in a way which involved interposing a person based offshore, who was not a genuine purchaser and re-seller, into the transactions instead of making the contractual arrangements that they did with Mr Khan. There is ample anti-tax avoidance legislation available upon which HMRC could rely to frustrate artificial arrangements of the type described by Mr Sykes. However, in this case Mr Khan was genuinely acquiring the Company. A bona fide structure was adopted which was designed to mitigate the tax exposure of the vendor shareholders. HMRC has never suggested otherwise.

At the heart of Mr Khan’s appeal was the submission that the concept of “entitlement” was to be given a “wide practical meaning” which not only allowed but required the Court to have regard to all transactions which were intended to have a commercial unity. In this regard Mr Sykes relied on the fourth of the well-known principles adumbrated by Lord Wilberforce in WT Ramsay v Inland Revenue Commissioners [1982] AC 300 at pp 323G-324D. In that passage, Lord Wilberforce said that:

“it is the task of the court to ascertain the legal nature of any transaction to which it is sought to attach a tax or a tax consequence and if that emerges from a series or combination of transactions, intended to operate as such, it is that series or combination which may be regarded.”

Mr Sykes also relied on the approach taken to the interpretation of “entitlement”, albeit in a different statutory context, in Inland Revenue Commissioners v Scottish Provident Institution [2004] 1 WLR 3172 at [19], and how that approach was described by Lord Reed JSC in UBS and others v Commissioners for HMRC [2016] 1 WLR 1005 at [71].

Mr Sykes submitted that on the true construction of s.385(1)(b) it is insufficient to establish that a person is entitled to the distribution in the sense that he is the legal and equitable owner of the money once it falls due (though a bare legal entitlement would never be enough); nor that the money was credited to their bank account or comprised a book entry in their favour. In order to meet the requirement of “receipt” for the purposes of s.385(1)(b), that person must also have an element of control over the money, and in order to meet the requirement of “entitlement” that person must benefit from their ownership. The UT fell into error in failing to recognise this and in the light of its findings of fact, Mr Khan did not satisfy those requirements.

Mr Sykes pointed out that in Aplin v White the interest was paid into a bank account which the estate agent controlled, in that irrespective of the fact that the interest was earned on funds he owed to his clients, he was in a position to direct whether the interest should be paid out and if so, to whom; and that in Timpson’s Executors v Yerbury Mrs Timpson had the power under the Trust to direct the Trustees to make payments to whomsoever she wished. Therefore, he submitted, both cases were consistent with his analysis.

Mr Bradley submitted that on any reading of s.385(1)(b), Mr Khan both received and was entitled to the distribution, which was the payment by the Company on the purchase of 98 of its own shares from Mr Khan under the OMPA, and that neither the statute nor the case-law put any further gloss on the words “received” and “entitled to” or required some further element of control or benefit. Mr Khan plainly received that payment (by way of set-off of his indebtedness to the Company under the loan) and was plainly entitled to it as counterparty to the OMPA. What the selling shareholders received and were entitled to was something different, namely, a payment from Mr Khan under the SPA. That was the analysis adopted by the UT and it was correct.

Mr Khan’s case that he was not the person receiving or entitled to the distribution essentially rested on the proposition that he was bound to use the £1.95 million distribution to extinguish his liability to the Company in the same amount, but that did not assist him. Where X gets a dividend of £100 in circumstances where he owes Y £100 and has no other funds from which to pay Y, then the fact that he is practically certain to use the dividend to discharge his liability to Y does not mean that he did not receive or was not entitled to the dividend in the first place. Mr Bradley submitted that there was no distinction in principle between that example and the present case.

In the UBS case, the Ramsay approach was explained by Lord Reed at [61]-[68]. As he said, prior to Ramsay, fiscal legislation had been interpreted predominantly on a linguistic analysis. Moreover, the courts had treated each element of a composite transaction which had an individual legal identity as having its own separate tax consequences. Ramsay did away with both those features, and required the same purposive approach to be applied to fiscal legislation as to any other legislation. It established that the factual analysis depended on the purposive construction of the statute. The ultimate question was:

“whether the relevant statutory provisions, construed purposively, were intended to apply to the transaction, viewed realistically.”

Lord Reed added the following clarification of how that question should be approached:

“the requirement to view the transactions realistically means no more than that the facts must be analysed in the light of the statutory provision being applied. If the legislation is concerned with the overall economic outcome of a series of commercially linked transactions, then that is where the focus should be; but if the legislation requires the court to focus on a specific transaction, then other transactions, although related, are unlikely to have any bearing on its application.” (Emphasis added.)

That passage is particularly pertinent when considering the approach to be taken in the present case. It serves as a salutary reminder of the dangers of trying to import interpretations of similar words or phrases used in the context of other statutes which may have entirely different purposes. It also stresses that not all fiscal legislation is concerned with the overall effect of a series of related transactions viewed as if they were one composite transaction.

It is unusual for a taxpayer to rely upon the Ramsay approach, which is generally invoked by HMRC when seeking to challenge artificial tax-avoidance schemes (which this undoubtedly was not). However, the principles in Ramsay are of general application, and our focus must be upon whether the UT erred in refusing to look at the sale and buyback transactions as a single composite whole and if so, whether that led to their reaching the wrong conclusion as to who “received or was entitled to” the distribution.

In my judgment, this is a case in which the legal nature of the transaction to which a tax consequence is attached does not emerge from looking at the connected transactions taken as a whole. On the contrary, the statutory provisions require the focus to be on the transaction under which the taxable distribution arose. However, even if one were to look at the transactions taken as a whole, they do not produce the end result contended for by Mr Sykes, namely, a distribution by the Company in respect of its shares to the vendor shareholders.

The UT undoubtedly started in the right place by seeking to ascertain the purpose of the relevant statutory provisions and to give them a purposive construction. Mr Sykes sought to challenge that construction on the basis that the UT had erred in finding at [54] that s.385 uses “entitlement” as a result of the way in which tax credits in respect of distributions previously operated, citing s.231(4) of ICTA. He said this ignored the fact that the expression “receiving or entitled to” appeared throughout ITTOIA and in the predecessor statutes. However, in that paragraph the UT was doing no more than referring to specific features of the earlier legislation that supported the inference that the owner of the taxable income was liable to pay the tax, if that person was not the actual recipient. As I have already mentioned, the point about the inference to be drawn from the provisions relating to tax credits was made in the Explanatory Notes.

The matters considered by the UT did support their interpretation of s.385(1)(b) that, irrespective of actual receipt, the person “entitled” to the distribution is the person to whom the distribution belongs. That interpretation is also consistent with the approach taken by the Court in Timpson’s Executors v Yerbury. The focus of the Court’s concern in that case was not on Mrs Timpson’s power to direct the Trustee what to do with the money, but rather, on whether she had exercised that power in a way which meant that at the time when the money came within the jurisdiction, and prior to actual payment, it belonged to one of her children instead of her. On the facts that was not so, because Mrs Timpson could have revoked her instructions at any time before the money went into the recipient’s bank account.

The Scottish Provident case relied upon by Mr Sykes related to a completely different set of statutory provisions. It concerned a series of linked transactions in which an option was granted to a financial institution to buy certain gilts at a price well below the market price on the date of sale, which would cause a loss to the seller. However that option would be cancelled out immediately by an option on the part of the seller to purchase the same quantity of gilts from the buyer at a price calculated to ensure that no loss was in fact suffered. There was a theoretical risk that the second option would not be exercised. The issue was whether, under the relevant statutory provisions, the buyer was “entitled to” the gilts in consequence of the existence of the first option. The argument that he was so entitled depended solely upon the existence of that theoretical risk. If one looked at the first option agreement in isolation, (ignoring the second option because of the theoretical risk that it would not be exercised) the answer to that question would be yes; but if one looked at the transactions holistically, as the Court decided was the correct approach, the answer would be no.

At [71] of the UBS case, Lord Reed explained Scottish Provident as relating to a statutory provision which, properly construed, concerned a real and practical entitlement to the gilts, rather than a legal entitlement which was expected and intended to be cancelled out by an equal and opposite obligation. Put another way, the gilts in the Scottish Provident case could not have been realistically described as “belonging” to the financial institution which had an option to buy them, because if it exercised the option it was obliged to sell them back immediately to the institution from which it had acquired them. However that decision, relating to a different statute, and very different circumstances, gave rise to no wider principle of interpretation of the concept of “entitlement” which can be carried across by analogy to ITTOIA.

I consider that the UT correctly identified the purpose of s.385 and correctly interpreted it. S.385(1)(b) is concerned with the person who actually received the distribution from the Company, and (if different) the person to whom that distribution belongs. In this case, they were one and the same.

In support of the proposition that an element of control is required to demonstrate receipt, Mr Sykes referred to Macpherson v Bond [1985] 1 WLR 1157, a case in which interest was credited to a deposit account in the name of the taxpayer which was subject to a pre-existing charge in favour of the bank for the indebtedness of a company. The terms of the charge entitled the bank to transfer money out of the account at any time in satisfaction of the debt owed by the company to the bank. It was held on the facts that the taxpayer had no immediate right to the money in the account or to the interest credited to it, at a time when it was obvious that the bank would take all the money out in satisfaction of the charge. The taxpayer would only be entitled to the interest to the extent that the liability of the company to the bank turned out to be less than £10,000. In the event, though it took a long time to establish the company’s indebtedness, it transpired that the bank had the right to all the money in the account, including the interest. Vinelott J held that the taxpayer was never entitled to the interest.

All that demonstrates is that on the facts of that case, someone other than the taxpayer (i.e. the bank) was entitled to the interest, by reason of the operation of a pre-existing charge over any sums credited to the deposit account. Because of this, the mere fact that interest was credited to the account in the bank’s books did not mean that the taxpayer had “received” it. The bank was never obliged to pay that interest to anyone other than itself.

Mr Sykes also referred to Williams v Singer [1921] 1 AC 65, but that was not a case about the construction of the phrase “the person(s) receiving or entitled to” taxable income. The issue in that case was whether UK based trustees were assessable to tax on income from a source outside the UK, paid outside the UK, to which non-UK domiciled beneficiaries were absolutely entitled, and which the trustees never received. Perhaps unsurprisingly, it was held that they were not.

Although there are references in Viscount Cave’s speech to “receipt and control” they must be read in their context, which was a rejection of the contention that, as between trustee and beneficiary, the burden of taxation must always fall on the beneficiary, because the beneficiary has the beneficial interest in the trust income. He explained that in the trust context, the basis on which a trustee or beneficiary will be liable to pay tax is not their status as trustee or beneficiary as such, but their receipt of the trust money. That is because “profits are to be taxed where they are found”. If the beneficiary, rather than the trustee, has both possession and control of the trust

In that case, the trustees had control over the trust income in the sense that they were able to direct to whom it should be paid, but they never received it, and their ability to direct payment to the beneficiaries outside the jurisdiction was not sufficient in itself to make them liable to pay tax. The case does not support Mr Sykes’ submission that

“control” is an implicit aspect of “receipt”. If anything, it suggests that receipt and control are different concepts, and practical control over the money is irrelevant if the money belongs to someone else. At most, the case supports the proposition that the legal title to income which falls to be distributed under a trust is insufficient to make the trustee liable to pay the tax instead of the beneficiary. The trustee is only liable to pay the tax if he receives the money.

Mr Sykes next referred to Anson v Revenue and Customs Commissioners [2015] UKSC 44, [2015] 4 All ER 288, which was a case about double taxation relief. At [109] Lord Reed JSC explained that the issue in that case was whether the income on which Mr Anson had paid tax in the USA was the same as the income on which he was liable to pay tax in the UK, which depended on whether certain profits which accrued to an LLC constituted the income of its members. That, in turn, depended on an analysis of the legal regime governing the rights of the entity and its members in relation to the profit. I could find nothing in that case which shed any light on the meaning of “received “ or “entitled to” in the context of s.385 or any other provision of ITTOIA, beyond perhaps making the obvious point that the answer to the question, as between two candidates, “whose money is this?” will usually depend on an analysis of the relevant legal arrangements between them.

In support of his submissions on the meaning of “entitlement” Mr Sykes placed considerable weight on a decision of the UT, BUPA Insurance v HMRC [2014] UKUT 262 (TCC) [2014] STC 2615. That was one of a line of cases, beginning with Wood Preservation v Prior [1969] 1 WLR 1077, concerning what is meant by

“beneficial ownership” for the purpose of establishing the requisite degree of connection between different companies that will allow one company to treat another company’s losses as its own for the purposes of group tax relief (or similar). Mr Sykes submitted that this case was the closest in terms of analogy to the present case, because “entitlement” under s.385(1)(b) plainly means more than bare legal title.

The question under the relevant statutory provision (s.403C(2)(b) of ICTA) was whether company A had a “beneficial entitlement” to a percentage of profits available for distribution by company C. The answer to that question would determine the

“relevant fraction” of C’s losses that could be set off against A’s profits. A had agreed with C’s parent company, B, to purchase shares in C. Part of the purchase price was

“earn out” under which A agreed, within 10 days of a distribution by C to A, to pay a certain percentage of the distribution to B. It was held that despite these contractual arrangements, A had a “beneficial entitlement” to the distribution, because A was free to use the money for its own commercial purposes. It was not legally obliged to use it to pay B, even though it was envisaged that this would happen in practice. In addition, during the 10 day window, A did not hold the money as trustee, it was entitled to keep any interest earned on it, it was exposed to foreign currency fluctuations and it was entitled to assign the money to someone else.

At [89] it was observed that this analysis of the facts “might have been very different” if A had no access to any benefits of ownership of the cash in the time between its receipt and its payment to B. That was the paragraph upon which Mr Sykes placed most emphasis. He submitted that if A had been obliged to pay the money out to B immediately, it would have been held that there was no “beneficial entitlement”. However, the focus on the recipient being able to enjoy the fruits of ownership only arises in that context because the legislation has been interpreted on the basis that Parliament intended to deny consortium relief to a company which has passed all the rights of ownership of the shares on to another company before ownership itself is transferred. In those circumstances, the connection between the profits and the losses to be set off against them becomes too remote to qualify. That is why equitable ownership under a trust or similar arrangement will not always suffice to constitute “beneficial ownership” or “beneficial entitlement” in that context, though it usually will.

The question that was being addressed in the BUPA case was not “to whom does this distribution belong?” because the distribution plainly did belong to company A as the legal and equitable owner of the shares. The issue was whether, as a result of its contractual arrangements with B, A retained sufficient benefit or enjoyment from the distribution to which it was entitled in law and in fact, to amount to a “beneficial entitlement” for the purposes of consortium relief. As the UT explained at [63] and [64] of the decision in the present case, it was HMRC in that case who argued that the contractual obligation between A and B meant that A did not enjoy the receipt of the distribution in any meaningful sense, but on the facts, that argument was rejected.

The UT considered that the BUPA case was generally unhelpful to Mr Khan, in that it illustrated that a person does not escape a tax charge (on a distribution or dividend) merely because of commitments made to pay the money elsewhere or to use it for other purposes. That point is also well illustrated by an earlier decision of the Court of Appeal in the same line of authorities, Sainsbury Plc v O’Connor [1991] 1 WLR 963. The Court in that case was strongly attracted by the proposition that in the context of statutory provisions relating to group tax relief, “beneficial ownership” should be equated with “equitable ownership.” However it felt constrained by the earlier decision of the Court of Appeal in Wood Preservation Ltd v Prior to find that the concept of “beneficial ownership” required a taxpayer company which had agreed to sell or assign the relevant shares, or granted an option over them, to retain the rights that would attach to ownership pending completion of that transaction. Ownership had to be something more than a “mere legal shell”.

The issue in the case was whether the taxpayer company, which had granted an option to another company over 5% of its shares in a subsidiary, which was not yet exercisable, was the “beneficial owner” of the shares. It was held that it was. At page 976 C-D Lloyd LJ said this:

“ The question is not whether the taxpayer company required the consent of [the option holder] before a dividend could be paid, or whether a payment of dividend was likely or not (it was clearly contemplated as a possibility). The question is rather whether the taxpayer company would have received the dividend if it had been paid. The answer is in the affirmative. The fact that the amount of any dividend would have been deducted from the option price … does not mean that the taxpayer company was not beneficially entitled to the dividends in the meantime. So I am not persuaded that the taxpayer company’s rights in relation to the shares were no more than a “mere legal shell”.” (Emphasis supplied).

That case establishes that even in a statutory context in which equitable ownership may not be enough to amount to “beneficial ownership” or “beneficial entitlement”, the focus is on identifying the person who in reality has the right to the money. Any legal obligation they may have to treat that money in a particular way after that right accrues or after payment is made is irrelevant.

None of the cases cited to us by Mr Sykes comes anywhere near establishing that in circumstances where the only person who is (or could be) legally and beneficially entitled to the distribution at the time when it is made, is contractually obliged to pay an equivalent amount to someone else and uses the distribution to discharge that obligation, (a) he is not “entitled to” the distribution, (b) he does not “receive” the distribution, or (c) his entitlement to or receipt of the distribution is to be disregarded, and the person who receives the payment from him is to be treated as chargeable under s.385(1)(b). That factual scenario is not even the same as in the present case, because the payment to the vendor shareholders was made before the taxable distribution took place, and the distribution was set off against Mr Khan’s obligation to repay the loan.

As Mr Bradley put it, once a tax liability arises under s.383 upon the making of a distribution, the net is cast wide in terms of the persons from whom the Revenue can seek payment. Either receipt or entitlement will suffice.

In some cases, the identification of the person to whom the distribution truly belongs could involve having to stand back and look at the matter realistically, ignoring any technical or artificial legal arrangements that might have been put in place to obscure their identity. However, the fact that the question is one of actual receipt or entitlement at the time of the distribution, means that the statute requires the focus to be upon the situation at that time, not on anything that happens to the money afterwards, still less on how the person from whom the Company is buying the shares came to be in the position to sell them in the first place.

On the face of it, therefore, s.385(1) is not a statutory provision that is concerned with the overall economic outcome of a series of commercially interlinked transactions, but only with the question of who was entitled to the distribution or who actually received it. In this case, the distribution was the money that was payable by the Company in respect of the 98 shares under the buyback agreement, the OMPA. The agreements in this case did not resemble the cross-cancelling option arrangements entered into in the Scottish Provident case. Mr Khan did not have a “bare legal entitlement” to the distribution. He had a contractual entitlement to the price for the shares he had sold to the Company under an agreement that was last in time to be executed. That price was to be paid by means of a taxable distribution. He had not created a charge or trust over

The fact that, at the end of the day, Mr Khan ended up as the owner of the sole remaining share at a modest personal outlay whilst the shareholders ended up with a sum equivalent to the Company’s previously distributable reserves tells one nothing about who received or was entitled to the distribution when it was made.

Mr Sykes submitted that the overall economic outcome was precisely the same as it would have been had the deal been structured in the manner originally envisaged and most of the shares had been the subject of a direct buyback by the Company from the vendor shareholders; but even if that structure might have achieved the tax efficiencies that this structure also sought to achieve, that is completely irrelevant. As Lord Greene MR said in response to a similar argument in Henriksen v Grafton Hotel Ltd [1942] 2 KB 184 at 193:

“This argument has a familiar ring. The answer to it is that this was not the contract which the parties chose to make. It frequently happens in income tax cases that the same result in a business sense can be secured by two different legal transactions, one of which may attract tax and the other not. This is no justification for saying that a taxpayer who has adopted the method which attracts tax is to be treated as though he had chosen the method which does not or vice versa.”

Moreover, as Mr Bradley submitted, even in a case in which there are a series of preordained steps designed for no reason other than to save tax, and the application of the principle in Ramsay requires consideration of the transaction taken as a whole, the characterisation of the composite transaction must be consistent with the result of the component transactions. In support of that proposition, he relied on Knox J’s judgment in Piggott v Staines Investments Ltd [1995] STC 114, and in particular the passage at pages 141a-142c, which signifies that the end result must be one that can be achieved lawfully and consistently with the individual steps taken towards it.

In that case, company X owned company Z and was liable to pay corporation tax on payment of dividends on Z’s profits. In order to minimise the group’s exposure to corporation tax, X purchased company Y, and then transferred all its shares in Z to Y. Z subsequently declared and paid a dividend to Y and on the following day, Y paid an identical dividend to X. Y then sought to carry back its exposure to advance corporation tax on the dividend it had received from Z, and set it off against the corporation tax it had paid in earlier years.

HMRC argued that the payment by Y to X should not be treated as a dividend but characterised instead as a contractual obligation to pay over to X the dividend that it had received from Z. That argument failed. It was held, among other matters, that it was impossible to characterise the end result of these composite arrangements as a direct payment of a dividend by Z to X in a way that ignored and overrode the rights of Y as shareholder in Z. At the time when payment of the dividend was made by Z to Y, X was no longer a shareholder in Z and was not entitled to payment of a dividend by Z. If Z had paid the money directly to X, X would have held the money as constructive trustee for either Z or Y. It was not factually possible to treat the dividend as having been paid earlier than it was (i.e. before X purchased Y) because

There are obvious parallels with the present case. It is not possible to ignore Mr Khan’s position, at the time of implementation of the buy-back, as the sole shareholder in the Company, nor the effect of the individual legal transactions, negotiated at arms’ length, by which he purchased all the shares and sold all but one of them to the Company. There were two sales, not one. Even if it had been possible to go behind the UT’s fact-findings, which Mr Sykes did not seek to do, the loan from the Company could not possibly be re-characterised as payment of a taxable distribution under the OMPA because of its timing. The transfer of the money occurred before the OMPA came into effect. That agreement was not operative or enforceable until Mr Khan became the Company’s director and shareholder and passed all the relevant resolutions.

Even viewed as a composite whole, these transactions cannot be characterised as one in which the Company made a distribution out of its profits to the vendor shareholders via Mr Khan. Indeed the deal was structured deliberately so as to avoid a direct buyback of the shares by the Company from the vendor shareholders. It was not legally possible for the Company to make such a distribution to the former shareholders. Only Mr Khan was a party to the OMPA and he was not acting as their agent or trustee but selling the shares to the Company in his own right. Indeed at the time of the buyback, the vendors had not only sold their shares to Mr Khan but had been paid £1.95 million of the purchase price for them.

However, even if matters had been different, and there had been no prior loan by the Company, and Mr Khan had paid the purchase price to the vendor shareholders out of the proceeds of the buyback after they were distributed, it would still have made no difference to the fact that he both received and was entitled to the money from the Company at the time of the distribution. The fact that payment of the distribution was made by way of set-off against the liability to repay the loan, does not mean that there was no receipt. Mr Khan derived a real benefit from the payment because it extinguished his corresponding liability to repay the loan. This was not a case in which Mr Khan’s interest in the money could be described as a “mere legal shell” with the vendor shareholders having all the rights of beneficial owners over that money.

A contractual obligation to expend the money in a particular way after it has been received does not mean that the recipient ceases to be entitled to the distribution; here, the contractual arrangements did not confer on the vendor shareholders any rights to the distribution, let alone any greater rights than those of the legal and beneficial owner of the shares that were sold under the buyback agreement.

Despite Mr Sykes’ attractive presentation of the arguments, I am not persuaded that the concept of “receipt” in s.385(1)(b) contains an implicit requirement that the person who receives the distribution must also have practical control over it. “Entitlement” means no more than having the right to the taxable income, in this case, the distribution, and there is no further implicit requirement of benefit in the sense

The vendor shareholders were entitled to a sum of £1.95 million in respect of the sale of their shares to Mr Khan, not to the distribution in respect of the shares which the Company bought from him; it is because their £1.95 million related to something other than the distribution that there will be no double taxation if they pay CGT on the sum that they received, after making any appropriate deductions.

However one views these transactions, they cannot be re-characterised as a buyback arrangement made directly between the vendor shareholders and the Company, ignoring the genuine role played by Mr Khan and disregarding his legal rights and obligations.

For those reasons, the UT was right for the reasons that it gave. I would dismiss this appeal.

I agree.

I also agree.