Hugh Edward Mark Osmond & Anor v The Commissioners for HMRC

In summary Mr Afzal submitted as follows:

The appellants’ submission that because their main purpose was to secure EIS disposal relief (i.e. a CGT relief) they could not have a main purpose of obtaining an income tax advantage is a bold but flawed submission.

His primary position is that given the appellants’ stated purpose of securing EIS disposal relief, it necessarily follows as a matter of law that they had a main purpose of obtaining an income tax advantage. This arises from the definition of income tax advantage in section 687.

EIS disposal relief generates an income tax advantage. This is because obtaining EIS disposal relief means the person is within the definition of income tax advantage as the CGT payable on the relevant consideration is less than the income tax payable if that relevant consideration had been paid to that person by way of a qualifying distribution.

And so, if that persons intended purpose was to obtain EIS disposal relief, it must follow, as a matter of law, that the purpose was to obtain an income tax advantage as defined.

The crucial point about this submission is that once the appellants’ submitted primary purpose of obtaining EIS disposal relief is accepted, there is no need for any further analysis of the evidence. The purpose of obtaining an income tax advantage is established as a matter of law.

It is also to be noted that the appellants accept that an income tax advantage was actually obtained.

His secondary position was that, on the facts, the evidence shows that the appellants did have a main purpose of obtaining an income tax advantage.

Firstly, it is clear from that evidence that the appellants did not want to pay any tax on the consideration for the share buyback. Secondly, the appellants knew that the amount of consideration had been calculated to ensure that it reflected a return of capital and thus, in their view, did not comprise an income distribution on which income tax would be payable. The whole purpose of the transactions was to ensure that it was treated as a capital transaction (thus banking EIS disposal relief) and this could only be achieved if the consideration was treated as a return of capital. There was no point in returning more than that since it would be subject to income tax. And this reflects the appellant’s purpose to obtain an income tax advantage.

Thirdly, it is unsurprising that the appellants did not consider taking dividends given that they knew at the time of the share buyback, and had known for years, that this would crystallise an income tax charge which was something which they had been at pains to avoid. Whilst the appellants had structured their corporate affairs so as to retain EIS disposal relief on a disposal of the shares in the company, one of the reasons was not just to obtain a favourable CGT relief, but to avoid paying income tax on any sums received in respect of their shares.

In Allam v HMRC [2022] STC 37, (“Allam”) the Upper Tribunal approved dicta in the first instance decision that, for a person to have a purpose of avoiding income tax, there must be an alternative transaction that would incur an income tax cost. But it was not saying that as a matter of law you cannot have a main purpose if you have not thought about that alternative transaction. Whilst in Allam it was important to identify the main purpose, there is no principle that the existence of an alternative transaction will always be a strong factor in identifying a main purpose. The significance of an alternative transaction will depend on the facts of the case.

In this case the fact that the appellants might not have consciously considered taking a dividend as an alternative to the share buyback misses the point. The appellants already knew, and had known for years, that such a dividend would crystallise an income tax charge. And had structured the share buybacks to avoid this. The fact that they had not even thought about taking a dividend was because of that historical knowledge.

Similarly, it is irrelevant whether the appellants needed to extract value from the company by way of the share buyback. He accepts that they did not need the money. But the fact is that the transaction was structured as a capital transaction in order to avoid an income tax charge.

The decision in Lloyd v HMRC [2008] STC 681 (“Lloyd”) also supports HMRC’s position. In that case the appellant’s desire to obtain the benefit of retirement relief (which if obtained would mean that the CGT paid on the sale of shares was lower than the income tax that would have been paid on a dividend) was a tax advantage motive. This is very similar to the appellants’ position in this appeal.

In summary Mr Gordon submitted as follows:

The question of whether the appellants had a main purpose of obtaining an income tax advantage is a pure question of subjective fact (IRC v Brebner [1967] 2 AC 18 at 30 (“Brebner”) as approved and applied in Allam.

It is a purpose and not a benefit test, and concerns only an income tax advantage and not a more general tax advantage. So, if a transaction has another purpose so that the obtaining of an income tax advantage is not a main purpose then the motive test is not satisfied even if an income tax advantage actually arises.

Nor is it satisfied if the obtaining of an income tax advantage is not a main purpose but merely a subsidiary purpose.

As emphasised in Allam at [170] a taxpayer’s purpose is determined by conscious motives and not subconscious motives “it was common ground that the test as to Dr Allam’s purposes in being a party to the transaction is a subjective test. We cannot see that sub-conscious motives have to be taken into account, although we accept that inferences can be drawn from the primary facts as to a party’s true motives”.

It is clear from the facts in this case that the main purpose of the share buyback was to crystallise the EIS disposal relief in case that relief was suddenly withdrawn. Preservation of this relief was the driving factor behind the way in which the appellants had structured their commercial arrangements over the last 20 years.

No conscious thought was given to an alternative transaction (namely an extraction of value by way of an income distribution) at the time of the share buybacks. It simply never crossed the minds of the appellants that they should undertake such a distribution. This would have been wholly contrary to the way in which the value is extracted from EIS companies, and something which they would not have countenanced. Obtaining an income tax advantage was simply never an issue that arose in the minds of the appellants.

In Lloyd, it was found in the facts that obtaining a tax advantage was a main object. In this case, the purpose of the share buyback was to crystallise the CGT exemption, and an income tax advantage was an effect and not a purpose.

In this case HMRC appear to be saying that the appellants had a primary purpose of crystallising EIS disposal relief and were thus consciously taking advantage of that. But they also had a secondary or subconscious purpose of obtaining an income tax advantage by virtue of that claim. This secondary or subconscious purpose cannot be taken into account.

The appellants had nurtured EIS disposal relief for 20 years The purpose of crystallising that relief in 2015 was to ensure that it was not lost on a change of government. Had it been so lost, then the CGT advantage which it provided to the appellants would have been reduced or negated when compared to CGT payable on a capital transaction at a later date. It was never either commercially or legally intended that there should be an extraction of value from the company, or on a disposal of shares, which would comprise an income taxable distribution. So, the comparator when considering the crystallising of EIS disposal relief is not with an income taxable transaction, but with a capital transaction at a later date. It is a CGT and not on income tax play.

The effect of the share buybacks was a wholly unnecessary and unwanted extraction of funds. And if the crystallisation of the EIS disposal relief could have been achieved without such an extraction, the appellants would have acted accordingly. But it wasn’t. The only way they could achieve their purpose was to undertake the share buybacks. The purpose of that was to trigger the disposal rather than extract funds.

If HMRC’s primary argument is correct, then any person who desires CGT treatment for a transaction is potentially within the TIS regime. That would include any sophisticated EIS investor That simply cannot be right. There must be more. There must be some conscious thought given to an alternative transaction. It cannot equally be right that someone who stumbles into a capital transaction passes the motive test because that person has given no conscious thought to an alternative, yet someone who has given such conscious thought to it, fails it.

Furthermore, it is clear from Brebner that a taxpayer can make a conscious choice between two ways of carrying out a commercial transaction, and simply because carrying it out in a way which pays more tax than the other does not bring with it an inference that the main object for carrying out the transaction on which less tax is payable is to obtain a tax advantage.

So, there must be a comparative transaction, both theoretically and realistically. In this case, whilst there was a theoretical alternative (the payment of a dividend) there was no conscious thought given to this. And realistically, on the evidence, the appellants would never have extracted value from a company by way of a dividend. So, on the evidence (and indeed as a matter of common sense) no conscious thought was given to an alternative way of crystallising the EIS disposal relief other than undertaking the share buybacks.

We start by setting out the relevant legislation which, although set out in the appendix, is set out below for ease of reference.

There is no dispute in these appeals that the share buybacks comprised a transaction in securities, nor that the effect of the buybacks was that the appellant secured an income tax advantage. In other words, the appellant satisfied the section 684(1)(a) and (d) conditions. The condition in subsection 1(b) is not relevant, so the focus is on the application of subsection 1(c), i.e. the main purpose issue.

We remind ourselves of some general principles culled from the legislation and case law:

The question of whether the appellants had a main purpose of obtaining an income tax advantage is a pure question of subjective fact (Brebner at 30, as approved and applied in Allam).

It is a purpose and not a benefit test, and concerns only an income tax advantage and not a more general tax advantage. So, if a transaction has another purpose so that the obtaining of an income tax advantage is not a main purpose then the motive test is not satisfied even if an income tax advantage actually arises.

Nor is it satisfied if the obtaining of an income tax advantage is not a main purpose but merely a subsidiary purpose.

The test is however satisfied if one of the main purposes of the transactions is to obtain an income tax advantage. In other words, there can be more than one main purpose.

For a person to have a purpose of avoiding income tax there must be an alternative transaction that would incur an income tax cost.

However when the question of carrying out a genuine commercial transaction is reviewed, the fact that there are two ways of carrying it out - one by paying the maximum amount of tax, the other by paying no, or much less tax - it would be wrong, as a necessary consequence, to draw the inference that in adopting the latter course, one of the main objects is to obtain an income tax advantage. “No commercial man in his senses is going to carry out a commercial transaction except upon the footing of paying the smallest amount of tax that he can” (Lord Upjohn in Brebner).

We now consider Mr Afzal’s primary submission which is one which we have not seen made by HMRC in any previous case in which, at first blush, caused the judge to raise a quizzical eyebrow. However, for the reasons given below, we think he is correct when he says that, as a matter of law, the appellant’s main purpose of being a party to the share buybacks was to crystallise or bank their EIS relief was also a main purpose of obtaining an income tax advantage.

The appellants have accepted that the effect of the share buyback is that less income tax has been paid on the consideration would have been the case had it been paid to them as a qualifying distribution. This is because they have obtained EIS disposal relief on the consideration. And so paid no CGT. So, the amount of income tax that would have been paid on the consideration was less in the CGT actually payable.

He accepts that this is not the test. The effect of the transaction was to generate an income tax advantage, but he needs to go further than that. He needs to show that it was a main purpose.

He says that there was. His logic runs as follows. The definition of income tax advantage is, essentially, that the actual amount of income tax payable (in these appeals zero as the consideration is allegedly subject to CGT) in respect of the consideration is less than the income tax payable if that consideration had been paid by way of a qualifying distribution.

But if you obtain EIS disposal relief, you must be within the definition of an income tax advantage as the CGT payable is necessarily less than the income tax which would have been paid had the consideration been paid as a qualifying distribution.

So, it must necessarily follow that if you have, as a main purpose, the obtaining of EIS relief, you must necessarily have, as a main purpose, the obtaining of an income tax advantage. A claim for EIS disposal relief is necessarily an income tax advantage and so the main purpose of obtaining that relief must also necessarily be a main purpose of obtaining that income tax advantage.

And we need go no further than that.

Mr Gordon’s view is that this cannot be right that because if someone has a main purpose of obtaining a CGT “benefit” (our words) that automatically means they have a main purpose of obtaining an income tax relief. There must be more. Conscious thought must be given to the alternative transaction which would have generated the higher income tax charge. And, as Brebner shows, simply because someone carries out a transaction in a tax efficient way does not mean that one can infer that they had, as a main purpose, the obtaining of an income tax advantage.

We have to apply the legislation to a specific transaction, namely the share buybacks. That is a real life transaction. We have found as a fact that a main purpose of the parties for the share buybacks was to enable the appellants to enable them to crystallise or bank the EIS disposal relief which they had preserved and nurtured for many years.

It therefore follows that, as a matter of remorseless statutory logic, that a main purpose was also to obtain an income tax advantage as, as that phrase is defined. The amount of income tax which would have been paid had the consideration been paid by way of a qualifying distribution was always going to exceed the CGT payable on the consideration in light of the benefit of EIS disposal relief.

In response to Mr Gordon’s assertion that there needs to be a consciously considered comparable transaction (something with which we deal in the discussion regarding HMRC’s second submission on the main purpose issue) our view is that the alternative transaction is already built into the definition of income tax advantage. The alternative transaction is the qualifying distribution identified in that definition. In essence it is a deeming provision limited only by the availability of distributable reserves Whether or not the parties have any intention of carrying out a transaction in an alternative way, and in particular whether they consciously or subconsciously considered paying the consideration by way of a qualifying distribution, is, when considering the statutory provisions, neither here nor there. The legislation itself identifies the alternative transaction which would incur an income tax cost. It is the qualifying distribution.

We can only reach this conclusion because the appellants reason for undertaking the share buybacks was so clearly to obtain the benefit of EIS disposal relief. As soon as that is found to be a main purpose, it is necessarily, and as a matter of law, a main purpose of obtaining an income tax advantage.

Mr Gordon observes that this would then bring taxpayers into the ambit of the TIS regime which were never intended to be so affected by it. But the TIS regime is intended to be a freestanding anti-avoidance provision and, in our view, deliberately casts its net very widely. And indeed, when it was introduced in 1960, it was, in shorthand, designed to ensure that individuals who sought to structure a transaction in a way to avoid paying income tax which might otherwise have been justifiably payable on sums extracted by an income taxable distribution should be brought back into the income tax net. Effectively schemes which “converted” income to capital were to be subject to the TIS regime. And so, it is unsurprising to us that where someone has, as a main purpose of entering into a transaction, the obtaining of a CGT benefit, that person is potentially within the ambit of the TIS regime. As a matter of statutory construction, when considering the legislation in its context and in a purposive way, we do not think that this interpretation leads to injustice.

But this does not mean, as Mr Gordon seems to imply, that one can simply sleepwalk into the TIS regime. We accept that conscious thought must be given to the entering into of the transaction. But if that conscious thought includes a main purpose of obtaining a CGT benefit or advantage, we cannot see anything absurd about the legislation applying. Indeed, as Mr Afzal accepts, it is only because the appellants have been so frank about their motives that he can run this primary argument.

We can see no principled reason why a main purpose of obtaining a CGT benefit or advantage cannot also be a main purpose of obtaining an income tax advantage. On both a literal and purposive interpretation of the legislation it can be. And it is our view that Mr Afzal’s submission regarding the law and its application to these appellants is correct.

For these reasons we conclude that, as a matter of law, the appellants did have, as a main purpose of entering into the share buyback, the obtaining of an income tax advantage.

We now consider Mr Afzal’s secondary submission. He says that if we are against him on his primary submission, then the evidence shows that the appellants did have a main purpose of obtaining an income tax advantage when entering into the share buybacks.

He submits that the appellants knew that if they had taken the consideration by way of qualifying distribution, that it would have been subject to income tax save to the extent that it represented a return of capital. The consideration therefore was calculated to ensure that there was no such excess, and the £20 million or so paid for the share buybacks was all a return of capital. So conscious thought was given to the way in which the transaction was structured so as to ensure that no income tax was paid on the consideration. This is clear from both the oral evidence, and the way in which the buybacks were structured (as a capital transaction with payment made out of share premium).

Mr Gordon says that the evidence shows that no conscious thought was given to taking the consideration by way of distribution. Indeed, the evidence clearly shows that the appellants did not want to extract any value from the company at all. They were sitting on piles of cash and the last they wanted was to extract more. If the crystallisation of EIS disposal relief could have been obtained without undertaking any form of transaction, then that is what the appellants would have done. But their advice was that it was not possible to achieve that crystallisation without a real-life transaction. If the appellants did have any purpose of obtaining an income tax advantage by effecting the share buyback, that was a subconscious motive and should be discounted.

Shortly stated, we are with Mr Gordon on this point. We have accepted, and found as a fact, that the appellants did not wish to extract funds from the company. It was not as though they wished to and consciously chose between two alternative ways of achieving this at the lowest tax cost. Nothing could be further from the truth.

It was inevitable that to achieve their stated purpose, the transaction needed to be a capital transaction to which CGT would be prime facie applicable. We do not see this as evidence from which we can infer that the appellants had as a main purpose the obtaining of an income tax advantage.

We do not believe that the legislation applies where someone, having undertaken a transaction which has a certain tax consequence, is required to look around to see whether there are other, detrimental, tax consequences of that transaction and then compare the tax consequences of the actual transaction with those detrimental tax consequences to decide whether there was a main purpose of obtaining an income tax advantage. This is what HMRC appear to be doing in this case. Having taken the real-life transaction, namely a share buyback consideration for which was paid by a return of share premium, they have looked around to see what alternative transactions might be (a qualifying distribution) and said that that is evidence that the appellants had a tax avoidance motive.

But we do not accept this. As Mr Allen said in his evidence, another alternative might have been to extract value from the company by way of a bonus. But that would have been bonkers (we have paraphrased his evidence). Why on earth, he asked rhetorically, would he take money which he didn’t need from the company in a tax inefficient way. And we are with him on this. We cannot infer from the appellants structuring of the transactions as a capital transaction that they had a main purpose of obtaining an income tax advantage when entering into that transaction.

The same is true of Mr Afzal’s submission that further evidence of a main purpose of obtaining an income tax advantage is the fact that the consideration was tailored specifically to ensure that there was no income tax payable on the share buyback as it represented a return of capital on which no income tax was payable.

Mr Afzal submits that the fact that the appellants did not require the consideration is an irrelevance. We think it is highly significant. The transaction did not proceed on the basis that the appellants needed a certain amount of money from the company and then decided how best to extract it paying as little tax as possible. The share buyback was undertaken with some reluctance as it was the only way to crystallise the EIS disposal relief. It is inevitable, therefore, that to obtain the benefit of that relief, the transactions would be structured in a tax efficient way.

The purpose was achieved only if there was to be no income tax payable and this meant that the maximum to be extracted was limited to a return of share capital. It would have been bonkers to have extracted more than the amount of share capital, as the excess would have been subject to income tax. And, paraphrasing Lord Upjohn, no commercial man in their right mind is going to structure a transaction so that he pays the maximum amount of tax.

Mr Afzal cites Lloyd to support his position where it was found that the taxpayer’s motive in that case was to obtain retirement relief. “The tax treatment of the transaction was important to the existence and timing of the transactions”. Mr Gordon’s view is that in the case of these appellants, that tax advantage was an effect rather than an object of the transactions.

The difference between Lloyd and these appeals, is that in Lloyd it was possible for the judge to posit a reasonable alternative transaction namely “if the only object was for Holdings to acquire the appellant’s shares there could have been a share for share exchange”. On the facts of these appeals, there could not realistically have been an alternative transaction which would have achieved the same object. And it is clear from the evidence, and from our findings of fact, that it would be both unusual, generally, to extract funds from EIS companies by way of distribution, and that, in the specific circumstances of these appellants, there was no realistic possibility that they would wish to extract funds either at all, or by way of such a distribution. We do not think that a dividend from the company is a relevant alternative transaction against which an income tax advantage should be tested.

So, the appellants sensibly structured the share buyback so that they could bank their maximum EIS disposal relief. This meant that the maximum amount that could be paid as consideration for the shares was reflected by the value of their share premium accounts.

It is true that at the time of the share buyback, the distributable reserves of the company were about £36 million. And so, theoretically, the shares could have been repurchased for that amount. But we reject any suggestion that this is an alternative transaction against which we should test whether the appellants had an income tax advantage motive. This might have been the case had they wanted to extract money from the company but had restricted it to the £20 million reflected by the share premium account. But the facts do not show this.

We accept that the appellants knew the time of the share buyback, and had known for years, that value extracted from a company by way of a dividend would bear income tax. And they consciously structured the share buyback to ensure that no such income tax was paid by distributing, to themselves, an amount equal to share premium account as consideration for the share buyback. And so, no CGT was payable because of the application of EIS disposal relief.

The appellants accept that, as a matter of fact, they obtained an income tax advantage.

But in our view, as submitted by Mr Gordon and at the risk of labouring the point, this was not a main purpose of entering into the share buyback. It was a consequence of so doing. The appellants main purpose was to crystallise EIS disposal. They were concerned that a change of government would affect its availability in their circumstances. They therefore structured the transaction (the share buybacks) to crystallise that relief and did so in a tax efficient way. They did not need the money. There was no point in extracting more than their share premium. Whilst this meant that they paid no tax on the consideration something which they knew would have been the case had they extracted those sums by way of dividend, this was not a main purpose. It was a CGT play. It was designed to ensure that they obtained the benefit of CGT relief now. They did not have, as a main purpose, the obtaining of an income tax advantage.

If, therefore, we had not found for HMRC on their primary submission, we would have found against them on their secondary submission.

In summary Mr Afzal submitted as follows:

It appears to be common ground that the consideration falls within the definition of relevant consideration in section 685(4)(a). However, the issue is whether it is then taken out of that definition by virtue of the application of section 685(6).

It is equally common ground that in section 685(6) the text is incorrect, and references to “subsection (2)(a) and (b)” should be references to subsection (4)(a) and (b).

Section 685(4)(a) states the relevant consideration is consideration which “is or represents the value of assets which are available for distribution by way of dividend by the company or assets which would have been so available apart from anything done by the company”. It therefore operates as a cap. It identifies the maximum amount which could comprise relevant consideration.

If, as a matter of law, assets could not be distributed by way of dividend, then they fall outside the provisions of section 685(4)(a). This is clear from the judgment of Goff J (as he was then) in the decision in Addy v IRC 51 TC 71 (“Addy”). In that decision (construing an earlier but virtually identical piece of legislation) the judge said “the section applies where, in connection with the distribution or transfer or realisation of profits or income or reserves or other assets, a person receives a consideration which either is or represents the value of assets available for distribution by way of dividend; and in my judgment those words are in no way limited to distributions out of revenue funds”. The crucial words are “available for distribution by way of dividend”, and that means legally so applicable…”.

The share capital of a UK limited company cannot, under English law, be distributed by way of dividend. And thus, that share capital falls outside the provisions of subsection (4).

Since it is excluded from being relevant consideration within section 685(4)(a) it cannot then be taken out of that definition by the application of section 685(6).

So, in this case, the consideration was £20 million. That is considerably less than the distributable reserves at the time of the share buybacks of approximately £36 million. Thus, this entire £20 million was or represented the value of assets available for distribution by way of dividend.

The phrase in section 685(4)(a) that relevant consideration is “or represents the value of assets which are available for distribution by way of dividend” makes clear that the consideration does not need to actually be a sum of money or an asset that could actually have been distributed by way of dividend. Instead, it is sufficient, to be relevant consideration, that the consideration in question is equal to the amount that could be distributed by way of dividend. For example, if a company has distributable reserves and the shareholders sell the shares, then the consideration for the sale satisfies the words “is or represents the value of assets which are available for distribution by way of dividend” as was the case in IRC v Garvin 55 TC 24 (“Garvin”).

The proper interpretation of section 685(6) is then to reduce that £36 million by an amount which, under English law, could not be distributed by way of a dividend. However, since no part of the share capital of the company could have been so distributed, it had never been included so as to increase the distributable reserves above those which were so available. It is possible that those reserves could have been increased above what are effectively distributable profits, if, for example, under the law of a foreign jurisdiction, or in respect of a UK limited company, company law permitted share capital to be lawfully distributed. In those circumstances the cap in section 685(4)(a) would be increased by the amount of such share capital but would then be taken out of the definition of relevant consideration by the application of section 685(6).

But this does not apply to a UK limited companies since, as the law stands at present, there is no such possibility of share capital comprising assets available for distribution by way of a lawful dividend.

This interpretation is consistent with the tribunal’s decision in Bamberg v HMRC [2010] SFTD 1231 (“Bamberg”) at [14-15]. There is no merit in the appellants’ submission that dicta in that case suggest that share capital of a UK limited company is within the exclusion in section 685(6). The tribunal was suggesting that it could apply to UK unlimited companies and not stating that it could apply to UK limited companies.

The second half of subsection (6) which follows the words “despite the fact” is consistent with this interpretation. It makes clear that subsection (6) is directed at the situation where a particular law allows assets of the type mentioned (i.e. share capital or premium) to be distributed by way of a lawful dividend (in contrast to the position applying to UK limited companies).

There is no merit in the appellants’ submission that the example of Sebastian in a 2015 consultation document which resulted in a change in the legislation, supports their position.

It is clear that a share buyback comprises a transaction in securities. It would therefore be wholly illogical if the consideration for such a buyback was given by way of a return of capital, for that return of capital not to comprise relevant consideration. It would emasculate the legislation.

In summary Mr Gordon submitted as follows:

We should read the legislation as it was in force at the time of the share buybacks. And this was before the change to the legislation which took place in 2016. But the example given by HMRC in the consultation document is instructive. It is clear that in the example about Sebastian, HMRC accepted that Sebastian who was repaid £250,000 by way of share capital, benefited from the lower CGT rate as it was subject to CGT. And so, it was unsurprising that HMRC change the rules. At the time of the share buybacks, therefore, these appellants benefited from the same safe harbour that HMRC considered applied to Sebastian.

We must consider section 685 as a whole. HMRC are asking us to stop at section 685(4) and not consider section 685(6). HMRC’s interpretation renders subsection (6) redundant. But it must have a purpose. That purpose is to take away from the definition of relevant consideration in subsection (4), assets which fall within the provisions of subsection (6).

Section 685(6) contains the words “do not include assets which are shown to represent a return of sums paid by subscribers on the issue of securities…”.

The words “do not include” shows that subsection (6) operates as a reducer rather than a limiter to the cap. It is clear from the words of the legislation that a return of share capital to subscribers is excluded from being relevant consideration. In the case of these appellants, that is precisely what has happened. It is clear from the evidence that the consideration for the share buybacks comprised a return of capital.

Furthermore, the words “despite the fact” limit the words which precede it, to the words that appear after it. And it is clear therefore that the limitation is not limited simply to non-UK companies. It can apply to UK limited companies.

Bamberg demonstrates that the exclusion in subsection (6) is not limited to foreign companies. It can clearly apply to UK unlimited companies, and it leaves the door open for it to apply to UK limited companies. If consideration represents a return of capital paid by subscribers, it is within the exclusion in subsection (6) and so falls outside the definition of relevant consideration.

The purpose behind subsection (6) is to limit relevant consideration. If the consideration falls within subsection (6) it cannot be within subsection (4). That is the case for these appellants.

Although section 685 is set out in the appendix, for ease of reference we set out below the provisions of it which are relevant to these appeals.

The parties agree that the reference in subsection 6 to subsections (2)(a) and (b) should be references to subsections 4 (a) and (b). We agreed that this is how the legislation should be read and that that cross-reference is an error.

The essential difference between the parties is this. Mr Afzal suggests that in these appeals, the maximum amount which could be distributed by way of dividend is capped at the approximately £36 million of reserves which were available for distribution at the date of the share buybacks. Since the company could not distribute its share premium, that is not added to that £36 million. The £20 million consideration falls within that £36 million cap. It is not then taken out of account by subsection (6) which is intended to remove from the cap any increase to it in situations where share capital is legally distributable. There was no such increase in the circumstances of these appeals since UK limited companies cannot legally distribute their share capital.

Mr Gordon, we think, accepts that the amount available for distribution is the approximately £36 million, but it is his view that subsection (6) acts to exclude any return of capital to subscribers (as is the case in these appeals where the consideration for the share buyback was paid by way of a return of share premium) by dint of the words “do not include assets which are shown to represent a return of sums paid by subscribers on the issue of securities…” So, the £20 million consideration paid in this way is simply excluded from the subsection (4) definition of relevant consideration.

Whilst we recognise the force of Mr Gordon’s robust and commonsensical interpretation of the legislation, we prefer the interpretation submitted by Mr Afzal. We say that for the following reasons.

To comprise relevant consideration within subsection (4)(a) the consideration must be or represent the value of assets which are available for distribution by way of dividend by the company. It is clear from Addy that available for distribution by way of dividend means lawfully distributable and is not limited to distributions out of revenue reserves. So, share premium or capital of a UK limited company is not so legally distributable.

However, share premium or capital may be distributable if the company is located in a foreign jurisdiction or indeed by a UK unlimited company. For such companies, share premium does fall within the definition of relevant consideration in subsection (4)(a).

The purpose, therefore, of subsection (6) is to take out of this definition, share capital which is so distributable. This is clear from Addy in which Goff J stated, in respect of predecessor legislation to s 685:

“The Appellant's submission was that the Commissioners simply failed to discharge their statutory duty to exclude assets shown to represent a return of sums paid by subscribers on the issue of securities as provided in s. 28(2) in the passage immediately following the definition clause in para. (d) and that there is nothing in the Act about including or not excluding such assets if there be some quid pro quo. In my judgment, however, the passage in the section relied on has no application because it relates solely to foreign companies: see per Cross J. in Hague's case [1969] 1 Ch., at page 405 and 44 TC, at page 631, and per Megarry J. in Commissioners of Inland Revenue v Brown 47 TC 217, at page 234 - and his reasoning, it will be remembered, was adopted by Russell L.J. at page 236. Where the company is an English one assets representing share capital are excluded, but not because of this provision. It is because they are manifestly not available for distribution as dividend. In such circumstances, however, no problem arises unless, as in Hague's case, but not the instant case, the amount distributed exceeds the reserves, whether of a revenue or a capital nature, which are available for distribution as dividend. In my judgment, therefore, the Commissioners reached the right conclusion on this question, not precisely for the reasons which they state but for the reasons which I have just given.” (Emphasis added).

This was reinforced in Bamberg, which although not binding on us, is a decision with which we agree, and whose principles we adopt in this decision.

The tribunal started by noting as follows:

“[12] On the first issue, C(2) states that the assets in question 'do not include assets which (while of a description which under the law of the country in which the company is incorporated is available for distribution by way of dividend) are shown to represent a return of sums paid by subscribers on the issue of securities.' The courts have certainly assumed that the words in brackets refer solely to foreign law so that even if foreign law permits dividends out of share capital (or share premium account, as permitted for a Cayman Islands company in First Nationwide v Revenue and Customs Comrs [2010] UKFTT 24 (TC), [2010] SFTD 408) that is to be ignored with the effect that foreign companies are treated no worse than UK incorporated companies”.

Then, after referring to cases including Hague and Addy, the Tribunal stated:

“[14] In none of these cases did the possible application of C (2) to a UK company affect the decision since in all of them it was accepted that the capital of the particular UK companies was not available for distribution by way of dividend. In Addy Goff J said that for an English company assets representing share capital were manifestly not available for distribution as dividend. However, if the company had been an UK unlimited company there would be nothing to prevent distribution of share capital as dividend. If the point had been put to him, we are sure he would not have considered that the provision was restricted to foreign companies. While therefore there are clear statements that C(2) is restricted to foreign incorporated companies we consider that we are not bound by them as they are obiter.12 We see no reason why C(2) should not apply to a UK incorporated unlimited company. There may also be circumstances in which a purchase of own shares brings it into effect but this was not fully argued and the position is complicated by the fact that sometimes such a purchase is taxable as income.

[15] Having decided that C(2) can be applicable to a UK company, we do not consider that this assists Mr Thornhill's case. The effect of C(2) is that assets representing a return of sums paid by subscribers on the issue of securities are not available for distribution by way of dividend; it is not that the return of assets subscribed is never caught by Circumstance D. Suppose that WCL's loan stock had been TTEL's and that the appellant had bought it at the same discount from the original subscriber, we do not consider that this would prevent Circumstance D from applying to the repayment of the loan stock in circumstances where there were distributable reserves. C(2) determines the maximum that can be paid as dividend as the amount of distributable reserves (not including a return of share capital if that is otherwise distributable). Even if the appellant had received a repayment of the amount subscribed for the loan stock that would not prevent its being said that this was consideration in tax-free form that represented TTEL's distributable reserves. While this may look like overkill it is no different in principle from the sale of the shares of a D company for cash being potentially within Circumstance D. In both cases the Circumstance potentially applies and the taxpayer has to rely on the escape clause”. (Emphasis added)

These judgments clearly support Mr Afzal’s interpretation of the legislation, namely that the purpose of subsection (6) is to level the playing field, and to take out of account as relevant consideration, share premium or capital which is lawfully distributable. That is not the situation in these appeals where the share premium is not so lawfully distributable. We do not consider that Bamberg suggests, let alone is authority for the proposition advanced by Mr Gordon, that subsection (6) could, as UK company law currently stands, extend to UK limited companies. We reject his suggestion that the words “despite the fact” in subsection (6) can be so construed either literally or purposively.

Bamberg also demonstrates that, contrary to Mr Gordon submission, subsection (6) does not operate to exclude a return of amounts subscribed for a loan stock from the definition of relevant consideration in subsection (4). We see no principled distinction between amounts subscribed for loan stock in that case and amounts subscribed for shares in these appeals. Although subsection (6) refers to “securities” and the definition of “security” in subsection (9) does not appear to extend to shares, this was not a point that was taken by Mr Afzal who implicitly accepted that the exclusion in subsection (6) can, in principle, apply to a return of share capital.

But whilst it might be capable of applying in principle, it does not so apply in the manner suggested by Mr Gordon. The fact that the consideration comprised a return of share capital does not mean that it is excluded from comprising relevant consideration within subsection (4). That would only be the case if it had been included within that “pot” of assets available for distribution, which it was not. Subsection (6) does not operate as a free-standing exclusion clause. It must be read in the context of the legislation and the purpose for which it was enacted.

The TIS regime is an anti-avoidance regime which, in essence, permits HMRC to rewrite a transaction which has been treated by the parties as a capital transaction to one to which income tax treatment should apply. It was introduced in 1960 at a time when CGT was not on the statute books.

And the consideration which is subject to this treatment is, in our minds, a very broad concept. It cannot be limited to consideration to which the conventional tax provisions apply. So, in the circumstances of the share buybacks, the application of conventional tax provisions is that the return of share capital would not be treated as subject to income tax as it is simply repaying to a subscriber an amount which the subscriber had originally contributed to the company.

However, the TIS regime is not limited by such conventional treatment. It applies to recategorise a transaction and enables HMRC to treat consideration which would be outside the scope of income tax if conventional tax treatment applied, as being subject to income tax under this regime.

That recategorisation would be emasculated if the exclusion in subsection (6) acted to exclude consideration which, on conventional tax treatment, would be outside the scope of income tax (such as the repayment of share capital).

This interpretation is supported by the principles in Bamberg.

The TIS regime applies to transactions in securities, and the share buybacks fall within this definition. It would therefore, as submitted by Mr Afzal, be “illogical” if having brought such buybacks into the net of the TIS regime, they cannot be impugned if the consideration is provided by way of a return of share capital. The net of the TIS regime is intended to be very wide and applies where consideration is paid in capital form in circumstances where it would not otherwise be subject to income tax.

The provisions of subsection (6) need to be construed in light of the purpose for which the TIS regime as a whole was introduced. It is consistent with that purpose that it is interpreted as taking out of the definition of relevant consideration in subsection (4) only any additional amount which is been added to that pot by virtue of share premium being lawfully distributable. It is not intended to take out of account share premium which has not been originally added in to that pot. Such an interpretation would, in our view, be more than illogical, it would result in an absurdity. It would drive a coach and horses through the application of the TIS regime.

For completeness, we would add that we do not believe that the example of Sebastian in the 2015 Consultation Document and any changes to the legislation which Mr Gordon submits was to give effect to a lacuna in the legislation, demonstrate that there was such a lacuna. To our mind the legislation in force in 2015 is clear on its face, and when interpreted in a purposive way permits us to come to the conclusion that we have reached.

In summary Mr Afzal submitted as follows:

Section 698(2) provides that two matters need to be specified in the counteraction notice. Firstly, the adjustments required to be made to counteract the income tax advantage, and secondly the basis on which those adjustments are to be made. It is arguable that the counteraction notices contain that information (in the second schedule thereof), the only error being the amount of the assessment and the year of assessment. It is clear from the second schedule that the adjustments will be made by way of an assessment, and the basis was specified as being that the consideration for the share buyback was to be treated as a qualifying distribution so subject to income tax.

However, if it is accepted that they did not contain the prescribed information, they are nonetheless valid for three reasons.

Firstly, the counteraction notices cannot be read in blinkers but must be read in context. This is clear from the cases of Bristol & West v HMRC [2016] STC 1491 (“Bristol & West”) and R (on the application of Archer) v HMRC [2018] STC 38 (“Archer”).

In Bristol & West at [26], the Court of Appeal, when considering whether a closure notice was valid stated that “in our view the answer to the question identified in para [25] above depends upon the correct interpretation of the October Notice, as it would be understood by a reasonable person in the position of its intended recipient, namely B&W, having B&W’s knowledge of any relevant context….” (we describe this as the “objective reader” test).

The counteraction notices, therefore, must be read in their immediate context (the mini bundle) and the broader context (the background to the enquiry and the correspondence relating to that enquiry).

It is clear from the documents comprised in the mini bundle which included HMRC’s view of the matter letter, the assessment, and the detailed tax calculation, that the objective reader could have been in no doubt that the adjustments required to be made to counteract the income tax advantage were to be made by an assessment. Nor in any doubt as to the basis on which they were to be made (namely that the assessment includes an adjustment necessary to counteract the income tax advantage that HMRC believed would not have arisen had the appellant received consideration as a distribution rather than as a share buyback).

Indeed, the documents in the mini bundle contained considerably more information about the basis on which the assessments were being raised than is statutorily required.

There is no merit in Mr Gordon’s submission that the objective reader would have been confused by the fact that certain items of correspondence and notices were authored by Officer Rounding, and others by Officer Agnew.

Secondly, the counteraction notices expressly referred to various documents including the preliminary notifications of 5 January 2021 (in the case of the second appellant, this was incorrectly dated 5 January 2020) and the statutory declarations and counterstatements of HMRC. Archer is authority for the proposition that such documents can be incorporated by reference into the counteraction notices (at [28] “I accept that in principle a closure notice could incorporate another document by reference...” And at [29] “I accept that the closure notice incorporated Mr Archer’s original self-assessment by reference…”). It is clear from these documents that the adjustment was to be made by way of an assessment and the basis on which the assessment had been made.

Thirdly, any invalidity in the counteraction notices can be cured by the application of section 114 TMA. The counteraction notices are clearly within the class of documents covered by that section. It is Mr Gordon’s submission that the defects in the counteraction notices are so fundamental that they cannot be cured by section 114. But in Archer, and its analysis of the Court of Appeal decision in HMRC v Donaldson [2016] STC 2511 (“Donaldson”), it is clear that the correct approach is not to consider whether there is some a priori categorisation of defects which are fundamental or gross, but one must instead concentrate on the nature and effect of the defect in the particular circumstances of the case (see [35]). One must consider the impact on the recipient. It is clear that the test of whether the recipient was misled under section 114 TMA must be considered from the perspective of the objective reader equipped with the knowledge that, in that case, Mr Archer and KPMG had, including knowledge of what had led to the enquiry and what HMRC’s conclusions were.

In any event the flaws in the counteraction notices were not so fundamental that they oust the application of section 114 TMA. The assessments were correctly made out. It would be odd if failure to set out the basis for the assessments meant that section 114 TMA could not apply when it clearly could apply in Archer and Donaldson.

In this case the defect in the counteraction notices are ones of form not substance. It is clear that the objective recipient of the counteraction notices would not have been in any doubt as to the method of counteraction (the assessment) and its basis. And indeed, the appellants’ oral evidence shows that they understood that the counteraction notices and assessments were assessing them to the income tax which they would have paid had the consideration been paid by way of dividend.

Furthermore, the counteraction notices issue was not raised until after the appellants had brought their appeal. They and their advisers were perfectly able to understand the case which HMRC was making against them and to tailor their grounds of appeal accordingly. Neither they (subjectively), nor the objective recipient of the counteraction notice could have been confused and misled. Section 114 TMA therefore validates the defects.

Norton v HMRC [2023] STC 526 (“Norton”), an authority relied on by Mr Gordon is in fact consistent with the foregoing principles. It does not lay down, as a matter of principle, that a failure to comply with the statutory requirement will necessarily be an error of substance rather than form. It analysed the precise facts in that case. One can see why Norton was decided on its facts. The facts here are very different. On the facts of this appeal, the objective reader in the position of the appellants would have been then under no doubt as to what the adjustments were nor the basis on which they were to be made.

In summary Mr Gordon submitted:

The counteraction notices were defective on their face in that they did not specify either the adjustments to be made, or their basis. The schedules in the counteraction notices, which should have contained this information, do not contain taxpayer specific information. They are uncompleted templates.

The facts in Bristol & West are different from those in this appeal. In that case, when Bristol & West received closure notices, they had already been told they were not intended to be a closure notice. That is not the same as in this appeal.

However, if one adopts the objective reader test, the objective reader in the circumstances of this appeal would not have understood the closure notices as making the adjustment by way of an assessment, nor the basis on which the assessment was made.

It is clear from the evidence that correspondence and documents have been sent to the appellants by two officers, namely Officer Rounding and Officer Agnew. The preliminary notifications and the counteraction notices themselves were authored by Officer Rounding. The covering letter of the mini bundle, the view of the matter letter, and the notice of assessment were authored by Officer Agnew. Officer Rounding, therefore, is the officer who ostensibly had authority to issue the counteraction notices, and the objective reader would not have understood that any defects in that could be cured by information provided by another officer, Officer Agnew. That reader had no idea whether Officer Rounding had provided any form of authority to Officer Agnew to “fill in the gaps”.

There is nothing to suggest that Officer Rounding endorsed any of the information provided by Officer Agnew. But since Officer Rounding was the only officer permitted to issue the counteraction notices (“I as the officer duly authorised in that behalf…”), the objective reader would have considered that he had not complied with the provisions of section 698(2). It is not enough, as submitted by Mr Afzal, that the information is all there in the mini bundle. The issue is that certain statutory steps which had to be taken had not been so taken by the relevant officer. The objective reader would think that Officer Rounding was the sole authorised officer and had not done what only he could do. That objective reader would think that Officer Agnew had done what only Officer Rounding could have done. And critically, the objective reader would not have known whether what Officer Agnew did was what Officer Rounding, wanted.

Any additional information which might be incorporated by reference refers only to the transaction and not to the adjustments.

The Commissioners for Revenue and Customs Act 2005 might, at section 2(4), state that anything begun by or in relation to an officer of Revenue and Customs may be continued by or in relation to another, but that does not cure the confusion in the mind of the objective reader concerning the actions and authority of the two officers.

Bayliss v Gregory [1987] 3 W.L.R. 660 (“Bayliss”) is authority for the proposition that some defects are so fundamental that they simply cannot be cured by section 114 TMA. In that case a notice of assessment was served on a taxpayer marked 1974-1975. It was intended to refer to the tax years 1975-1976. The issue is whether the notice could take effect as an assessment for 1975-1976. The Court of Appeal held that it could not. The statutory and requirements of form required that the year of assessment was an essential element of the assessment itself. In that case, leaving section 114 TMA aside, the court thought that even though a taxpayer may have appreciated that a mistake had been made, that was irrelevant. An assessment for one specified fiscal year cannot take effect as an assessment for another fiscal year.

The court went on to hold the section 114 TMA could not cure the defect. The relevant fiscal year of assessment was an integral and fundamental part of the assessment itself and notwithstanding the width of section 114 TMA, it cannot be used to treat an assessment for one fiscal year, as an assessment for another.

A similar conclusion had been reached in the case of Norton. In that case a document purporting to be a closure notice did not state the essential items (that HMRC had completed their enquiry, HMRC’s conclusions, and any amendments to be made). The Upper Tribunal held that these were errors of substance rather than mere want of form and thus could not be cured by section 114 TMA.

In these appeals the errors in the counteraction notices are errors of substance. They cannot be cured by section 114 TMA as a matter of legal principle. And in any event, the omissions by Officer Rounding in the counteraction notices cannot be cured by information provided by Officer Agnew as there is no evidence that he had authority to do so.

Although set out in the appendix, for ease of reference we set out the relevant provisions of section 698 below:

There is no prescribed form for a counteraction notice but it must include two things. Firstly, the adjustments which are required to be made to counteract the income tax advantage (and thus must specify one of the adjustments set out in subparagraph (4)); and secondly the basis on which the adjustments are to be made.

Even though there is no prescribed form, it is clear that HMRC provide a pro forma, or template, to their officers which includes a first and a second schedule. The first schedule is intended to set out details of the transaction in question. The second schedule is designed to describe the adjustments that are required to counteract the income tax advantage.

In the case of these appellants and their counteraction notices, the issuing officer, Officer Rounding, had completed the text of the notice itself by including the dates of the preliminary notifications of 5 January 2021, and the date (26 March 2021) on which the tribunal had determined that there was a prime facie case for proceeding, but he had failed to complete the two schedules.

It is Mr Gordon’s submission that these omissions render the counteraction notices invalid. We agree. The first schedule does not contain details of the transactions in question, and this is required since the second schedule (which also fails to identify the amount of tax or the year of assessment to which it relates) indicates that the basis of the adjustment is that the appellant should be liable to income tax on an amount of qualifying distribution equivalent to the consideration which had been set out in the first schedule.

However, as Mr Afzal submits, and which Mr Gordon did not seriously challenge, that is not decisive. Mr Afzal has submitted that there are three saving provisions which validate the counteraction notices. Firstly, the objective reader test. Secondly by dint of the incorporation of documents specifically referred to in the notices themselves. Thirdly, by the application of section 114 TMA.

Bristol & West, the case from which the objective reader test derives, was a case involving closure notices. Like a counteraction notice there is no prescribed form for a closure notice, but it must state certain matters. The objective reader test asks how the reasonable person in the position of the appellants, having their knowledge of any relevant context, would have interpreted the counteraction notices.

The relevant context clearly includes the documents which accompanied the counteraction notices, and which were included in the mini bundles.

The objective reader in the position of the appellants would have been fully aware of the ongoing enquiry into the share buybacks. The reader would have been aware that HMRC were challenging the tax position reported on their tax returns. And that HMRC were looking to counteract the perceived tax advantage to such an extent that they had sent preliminary notifications and counter statements to the appellants and in response to which the appellants had compiled statutory declarations.

In those declarations the appellants were able to say they did not think that the share buyback was one to which section 684 could apply.

The reader would then have been presented with a mini bundle, which included the counteraction notices but also HMRC’s view of the matter letter dated 31 March 2021. That letter clearly identifies the share buybacks as being the transaction, and that HMRC, through the agency of Officer Agnew, was enclosing a counteraction notice and “tax assessment for the tax year ended 5 April 2015. The assessment includes an adjustment necessary to counteract the income tax advantage that HMRC believes would not have arisen have you received consideration as a distribution rather than a “buyback” of shares you held in [the company]”. The consideration is easily identified under the definition of “the Transaction”

This information clearly complies with the statutory requirement in section 698(2). Indeed, as Mr Afzal points out, it goes way beyond that statutory requirement and includes a synopsis of HMRC’s pleaded case.

The mini bundle also includes the assessment which is expressed to be for the tax year ended 5 April 2015, and specifies the amount charged. It also includes a schedule which sets out the original computations and the revised tax computations on the basis of the adjustments which HMRC have made.

We are in absolutely no doubt that the objectively reasonable person, in the position of the appellants, would have understood the documents in the mini bundle, read in light of their ongoing involvement in the enquiry, and the specific procedure relating to the preliminary notifications, statutory declarations, and tribunal hearings declaring a prime facie case, to have interpreted those documents as identifying the transaction which was the subject matter of the counteraction notices. And that the adjustments required to be made to counteract the tax advantage were the assessments and the associated numerical schedule which were included in the mini bundle. The basis on which those adjustments were made were clearly identified both qualitatively in the view of the matter letter and quantitatively in the numerical schedule.

Although this is not the test, we also find it telling that, as we have found as a fact, that the appellants fully understood that the effect of the counteraction notices and the accompanying documents was to assess them to income tax on the consideration rather than to CGT as originally submitted in their tax returns. We also find it telling that neither the appellants, nor those advising them, appear to have had any difficulty in understanding what HMRC were doing to, in their view, rectify the position.

It is Mr Gordon’s submission that the objective reader would have been confused by the fact that different officers had authored the various letters/notices. And would not have understood what authority the various officers had to issue those letters/notices. We reject the submission. In our view this would simply not have occurred to the objective reader let alone have been a point of concern or uncertainty regarding their content and effect

That reader would have received the documents and even if they had noticed that they were authored by different officers, would not, in our view, have given that a minute’s thought. This is true, too, of the reader having an insight into the provisions of the Commissioners for Revenue and Customs Act 2005 which was submitted, by Mr Afzal as providing some form of authority for the fact that Officer Rounding and Officer Agnew could be treated as acting seamlessly for HMRC. The objective reader would have no knowledge of this ostensible saving provision.

No, the objective reader would not have been confused by the fact that the letters/notices might have been authored by different HMRC officers. That reader, having received the mini bundle, with the background knowledge of these appellants, would have understood the counteraction notices, in context, as adjusting their respective tax positions, through the assessments, so that they were liable to income tax on the consideration rather than CGT.

We therefore find that the objective reader principle validated the counteraction notices and so we find for HMRC on the counteraction notice issue.

However, because the points were fully argued, we have also considered Mr Afzal’s alternative submissions. The first is based on Archer and is that the preliminary notifications of 5 January 2021 (in the case of the second appellant, this was incorrectly dated 5 January 2020), and the statutory declarations and counter statements can be incorporated by reference into the counteraction notices as they were referred to in them.

We agree that, as a matter of principle, this is one ratio of Archer.

We are not sure that the preliminary notification letter of 5 January 2021 (5 January 2020) takes Mr Afzal much further. Whilst that identifies the transaction and consideration, it does not state, categorically, that an adjustment will be made by way of an assessment, nor the basis of that assessment.

Nor are we satisfied that the contents of the statutory declarations fill the gap even if they are deemed to be incorporated within the notices themselves. Whilst it is abundantly clear that, (by dint of the process which culminated in the compilation of the statutory declarations and the directions of a prima facie case by the tribunal) HMRC were seeking to use the TIS regime to charge the appellants to income tax rather than CGT on the consideration, the statutory declarations contain no reference to the adjustments being made by way of assessments, nor (although this might be implied) the basis of the adjustments. They are largely concerned with motive.

So, on the second point, we do not agree that the incorporation of the documents referred to in the counteraction notices validates them.

However, we do agree with Mr Afzal’s third submission, namely that section 114 TMA validates the counteraction notices.

Section 114 TMA is set out below:

We are no doubt that the counteraction notices are within the ambit of section 114 TMA.

The fundamental issue between the parties is this. Mr Gordon submits, on the authority of Bayliss and Norton, that some defects or errors are so fundamental (they are errors of substance not form) that section 114 TMA is not capable of curing them. Mr Afzal submits that there is no principle that requires us to discern whether errors are fundamental, or of substance not form. It is simply a question of looking at the facts of any particular case, and then deciding whether or not the objectively reasonable reader of the document would have been misled. In other words, it is a test very similar to that of the objective reader.

We agree with Mr Gordon that both of the cases cited above do approach the application of section 114 TMA on the basis that it can apply to errors of form but not ones of substance. The million-dollar question of course is into which category an error falls.

However, the Court of Appeal in Archer adopted a somewhat different approach. As Mr Afzal submitted, in Archer, the court applied the test in Donaldson, namely one does not approach the application of section 114 TMA from the direction that some mistakes might be too fundamental or gross to fall within it. Instead, one needs to concentrate on the nature and effect of the omission in the particular circumstances of the case.

In Donaldson, a case which concerned the validity of penalty notices in which it was alleged that the notices were defective because they did not specify the date from which the penalty was payable, the Court of Appeal held that Mr Donaldson could have easily worked out that period, even though wasn’t stated, and he could have been in no doubt as to the period over which he had incurred the liability for a daily penalty. Furthermore, he was not misled or confused by the omission.

In Archer the Court of Appeal adopted the Donaldson test. It said that Mr Archer’s liability could have been easily worked out and he could have been in no doubt what he owed to HMRC. He could not be confused or misled. They went on to say that this meant that HMRC’s failure to amend Mr Archer’s return was a matter of form rather than substance on the particular facts of this case.

We have approached the section 114 TMA issue therefore, from the same direction as the Court of Appeal did in Archer. We have considered the particular circumstances of these appellants. We already found that they understood the counteraction notices to assess them to income tax under the TIS regime on the consideration they received for the share buybacks. So subjectively, there is no doubt that they were not misled by the omissions in the closure notices, nor confused by them, and they were in no doubt as to what the counteraction notices were intended to achieve.

This is what they thought subjectively. And we have already found that, if we apply an objective test (as Archer says one must) the objective reader would have also considered that the counteraction notices adjusted their tax returns by virtue of the assessments, and the basis for those assessments was that income tax should properly be payable on the consideration.

The test for section 114 TMA, therefore, is very similar to the objective reader test. “However, in applying an objective test, the reader of the closure notice must, I think, be taken to be equipped with the knowledge that Mr Archer and KPMG had, including knowledge of what had led to the enquiry and what HMRC’s conclusions were. This is consistent with Bristol & West at [26] and [38]” (Archer at [36]).

In our view, the knowledge of these appellants would include the information set out in the mini bundle.

It is our view, therefore, that on the facts of this case, the application of section 114 TMA does cure the failings in the counteraction notices.

Bayliss was a case which concerned the validity of assessments. HMRC had issued an assessment for the wrong tax year. It was held in that case that the assessments were invalid because the error was so fundamental it could not be cured by section 114 TMA. But the statutory and formal requirements in respect of the assessment were considerably more prescriptive in that case than under section 698. And furthermore, there is, in these appeals, no challenge to the validity of the assessments which effected the adjustments. As far as we understand it, Mr Gordon accepts that the defects are in the counteraction notices and not in the assessments.

In Norton the issue concerned the validity of a closure notice. The Upper Tribunal identified the matters which needed to be identified in a valid closure notice, one of which was that it needed to tell the taxpayer that enquiries had been concluded, and also needed to state a concluded view. On the facts of that case (it was clear that HMRC were continuing with their enquiries and had not come to a concluded view) the tribunal held that the notice was not in substance and effect a closure notice.

This is very different from the facts of the present case. The counteraction notices do not misrepresent the position. They are simply defective by omission. That omission is not, in our view, fundamental in view of the knowledge of the appellants regarding the enquiries, and the information set out in the mini bundle. Whilst the appellant in Norton might have been misled by the information in the closure notice, the same cannot be said of the appellants in these appeals. The objective reader of the counteraction notices, equipped with their knowledge, could have been in no doubt that the adjustments were being made by way of assessments, and the basis on which those assessments were made.

If, therefore, we had not already found for HMRC on this point we would have found for them on the ground that section 114 TMA applies to cure the defects in the notices.

In summary Mr Winter submitted as follows:

This issue focuses on whether the assessments, which were issued more than four years after the end of the tax year in question, but within 6 years of the end of that tax year, are out of time.

HMRC’s position is that a 6-year limitation period applies and so the assessments were in time. In support of this proposition, they make three submissions, two of which approach the position from the starting point of the ITA, and the third approaches it from the starting point of the TMA.

Firstly, section 698 itself gives HMRC the power to make an assessment within 6 years. Section 698(1) and (2) requires HMRC to make adjustments needed to counteract an income tax advantage and these adjustments allow HMRC to specify, inter alia, an assessment. That is set out in section 698(4)(a). That power is then limited by section 698(5) “Nothing in this section authorises the making of an assessment later than 6 years after the tax year to which the income tax advantage relates”. Thus, an assessment may be made within, but not after, the 6-year period. This 6-year period is designed to oust the 20 year time limit in section 36 TMA (a loss of tax under the TIS regime cannot be brought about by the deliberate non-compliant behaviour of a taxpayer) which does not extend the 6-year limitation period in TIS cases.

The 6-year time limit is consistent with the TIS regime being a self-contained regime and is therefore independent of the TMA.

Secondly, section 698(7) provides that “no other provision in the Income Tax Acts is to be read as limiting the powers conferred by this section”. This makes clear that the 4-year time limit in section 34 TMA does not limit the power to assess under section 698. This is because section 34 TMA is a provision in the Income Tax Acts.

The TMA is an Income Tax Act. This is clear as a matter of statutory interpretation, precedent, and the way in which the TIS regime operates.

As regards statutory interpretation, the definition in Schedule 1 to the Interpretation Act 1978 (“Interpretation Act”) is broad enough to encompass the TMA since it concerns assessments to income tax. HMRC v Inverclyde Property Renovation LLP [2020] STC 1348 (“Inverclyde”) a decision of the Upper Tribunal sitting in Scotland is the most relevant authority and we should follow it.

As regards precedent, there are four decisions at Upper Tribunal/High Court level regarding whether the provisions of the TMA can be part of the Income Tax Acts. There are two Scottish decisions one of which says yes and one of which says no, and two English decisions, one of which says yes and one of which says no. We should adopt the principles set out in the more recent decisions which say that the provisions of the TMA could be part of the Income Tax Acts.

In Inverclyde, having reviewed the authorities, and having considered the provisions of section 1 of the Interpretation Act (which states that every section 11 Act takes effect as a substantive enactment without introductory words) the court said at [38] “in our view these authorities provide ample support for the proposition that the word “enactment” is at the very least capable of referring to a section of an Act and not solely to a whole Act”.

At [37] it posed the question whether the expression “the Income Tax Acts” was capable of including provisions of the TMA concerned with income tax”, and in the courts view “that question falls to be answered in the affirmative”.

The court also recognised that it was differing from the view taken by Lady Smith in R (Spring Salmon and Seafood) v IRC 2004 STC 444, (“Spring Salmon”) and pointed out that it did not appear that reference was made in that case to Schedule 1 Interpretation Act.

In terms of being bound by Inverclyde, this is perfectly permissible although it was a decision of the Upper Tribunal sitting in Scotland. The tribunal system encompassing England, Ireland, Scotland and Wales is a unified system. The tribunal may sit in each of those four locations. However, a decision of the Upper Tribunal sitting in Scotland is just as binding on the FTT as a decision of the Upper Tribunal sitting in England.

Inverclyde also explains that all enactments relating to income tax should be construed as relating to any section of the TMA relating to income tax. And any reference in section 118 (1) TMA to “this Act” is of no practical significance. Nor is it otiose. The TMA contains provisions capable of applying to other taxes and there is no compelling reason to treat the expressions “this Act” and “the Tax Acts” in section 118(1) TMA as mutually exclusive.

Finally, if the appellants’ interpretation (section 34 TMA applies the extended time limits in section 36 are excluded by section 698(5)) applies, this would give odd results. In HMRC’s view the extended time limits in section 36 do not apply. If the time limits in section 34 TMA only applied, then that would render the six-year period in section 698(5) otiose.

Thirdly, approaching matters from the starting point of section 34 TMA, it is clear that section 34 TMA itself gives priority to the time limit in section 698. Section 34 expresses itself to be “Subject to the following provisions of this Act, and to any other provisions of the Taxes Acts allowing a longer period …an assessment to income tax may be made at any time not more than 4 years after the end of the year of assessment to which it relates”. The extended time period of 6 years set out in section 698(5) is just such an “other provision” given that the ITA is a Tax Act. It therefore overrides the 4-year period in s34 TMA and allows HMRC to make an assessment within 6 years from the end of the relevant tax year.

In summary Mr Gordon submitted as follows:

The adjustments which have been made under the counteraction notices are assessments. HMRC acknowledged that these were not made by way of discovery assessments under section 29 TMA. But the assessments are governed by the provisions of the TMA.

When section 698 refers to an assessment, even if the assessment is not made under the TMA, the ITA must still inevitably have had in mind the provisions in the TMA that govern the procedures for assessments.

In this case the 4-year time limit in section 34 TMA applies since there is nothing in section 698 which overrides that time limit.

The ostensible time limit in section 698(5) does not give HMRC the power to make an assessment within 6 years from the end of the relevant tax year. It is simply an additional limitation to HMRC’s powers to make an assessment over and above those set out in the TMA. This is consistent with the reading of the legislation and with the legislation considered in its historical context.

Section 698(7), which provides that “no other provision in the Income Tax Acts is to be read as limiting the powers conferred by this section” merely ensures that the substantive charging provisions elsewhere within the Income Tax Acts cannot exclude the application of the TIS regime. In other words, it gives that regime priority over other charging provisions. And section 698(5) provides that the application of the TIS regime is subject to an overriding time limit of 6 years.

In Tooth v HMRC [2021] STC 17, the Supreme Court stated that the provisions of the TMA set out a series of limitation periods for the making of assessments to tax each of them expressed in positive terms that an assessment “may be made at any time” up to the stated time limit. The language of section 34 TMA is different from the language in section 698(5). The former provides a clear limit on the power to issue an assessment. The latter simply imposes a further non-stop date and gives no power to assess in positive terms. It is not a permissive but a restrictive provision.

The language of section 698(7) (“is to be read”) is not the language used to disapply a provision, but it tells the reader how to construe a provision. All subsection (7) does is to ensure that section 698 can be invoked to ensure that income tax is charged on capital transactions despite what might be said elsewhere in the Income Tax Acts.

Section 698(7) deals with the interaction between the TIS regime and the rest of the Income Tax Acts; Section 698(5) deals with the interaction between the TIS regime, and the TMA.

So, the only modification of the time limits in the TMA by the TIS regime is to the absolute six-year time limit that is imposed by section 698(6). Save where one or more of the conditions outlined in section 36 TMA is met, there remains the four-year time limit on any assessment made in conjunction with a counteraction notice.

When Parliament changed the time limits in 2010 to permit assessments to be made within the 20-year time limit (for, for example, fraudulent conduct) under the TMA, that was still subject to the six-year cap which applied to the TIS regime.

The appeal rights in section 705 ITA are rights of appeal against the counteraction notice. They are not appeal rights against the assessments.

If, contrary to his primary argument, section 698 does confer on HMRC a power to issue an assessment within 6 years of the relevant tax year, then that time is cut down by the 4-year time limit in section 34 TMA. And, as the TMA is not an Income Tax Act the 6-year time limit in subsection (5) does not override the 4-year time limit in section 34 TMA.

Although the Interpretation Act defines “The Income Tax Acts” As “…all enactments relating to income tax, including any provisions of the Corporation Tax Acts which relates to income tax…” the TMA is not an enactment which relates to income tax.

The TMA was enacted in 1970 at the same time, and part of the same legislative architecture, as the Income and Corporation Tax Act 1970 (“ICTA 1970”). The TMA was intended to provide the administrative and procedural machinery, whilst ICTA 1970 provided the charging machinery for income tax. The TMA therefore did not relate to income tax. That function was carried out by ICTA 1970. Indeed, there is only one charging provision in the TMA and that relates to stamp duty.

In R (on the application of Derry) v HMRC [2019] UKSC 19 (“Derry”) Lord Carnwath said, at [20] “the TMA, as its title implies, is concerned principally with the management of the tax rather than fixing liability…”.

Furthermore, the TMA itself, in s 118(1), defines “the Taxes Acts” as meaning “…This Act and the Tax Acts…”. The Interpretation Act defines “the Tax Acts” as meaning the “… Income Tax Acts and the Corporation Tax Acts”. This strongly suggests that the TMA is not a Tax Act, and thus is not an Income Tax Act.

This was the conclusion reached by Lady Smith in Spring Salmon where she said, “it seems clear that TMA is separate and distinct from the group of statutes referred to as “the Tax Acts” in [Section 132 (1) ICTA 1988]”. Given that the tax law rewrite project was in full swing at that time, and Parliament did nothing to legislatively contradict this decision, we should presume that Parliament was content with Lady Smith’s conclusion.

We should prefer Spring Salmon to Inverclyde as the latter was wrongly decided. It failed to properly construe “enactments” as that phrase is used in the Interpretation Act.

The TIS regime is a freestanding anti-avoidance regime which contains the essential elements relating to the criteria for liability, the charging mechanics, and an appeals mechanism. We are here concerned with the charging mechanics, as we were in relation to the counteraction notices issue in section 698.

It is HMRC’s submissions that section 698 contains the essential ingredients for assessing the appellants and for the time period within which that assessment may be made. It is the appellant’s view that the six-year time period does not represent the only constraint on HMRC’s powers under the TIS regime. The constraints within the TMA also apply.

As regards the application and relevance of the TMA, we would make two preliminary points. Firstly, we were not shown any piece of legislation, nor binding case law, which conclusively states that any document that purports to be, or takes effect as, or describes itself as, an “assessment” is subject to the exclusive domain of the TMA. In the absence of any such legislation or case law, we fall back on the general rules of statutory interpretation, and we must consider the relevant statutory provisions in context. The context here is that the assessment is part of the charging regime set out in section 698 which itself is part of the freestanding TIS regime.

Secondly, as Lord Carnwath said in Derry, “turning to the TMA, it is true that words of Sch 1B taken on their own would be apt to apply to a claim under ss132-133. However, I do not regard that as enough to displace the clear provisions of the ITA in respect of liability. I do not see this turning so much on whether one set of provisions are more specific than the other, but rather on the fact that the ITA is in principle the governing statute in respect of tax liability, and as such should take precedence in the absence of any indication to the contrary”.

The TIS regime includes the provisions relating to liability, and thus must be determinative of that. The charging regime in section 698 as part of that regime must be read in that context. Unless, therefore, there is a clear qualification in the TIS legislation which demonstrates that an assessment made thereunder is subject to the provisions of the TMA, we consider that the ITA legislation should take precedence.

Mr Gordon suggests that section 698(5) is couched in different terms from the provisions of, for example, section 34 TMA, which provides for the time period within which an assessment must be made. The provisions of section 34 TMA are permissive. The six-year time limit in section 698 (5) is restrictive. So, section 698(5) does not give HMRC power to assess and that gap is filled in by the TMA.

But it is clear to us that HMRC not only have power to assess, but have a duty to do so, by dint of the language in section 698(1) and (2). In subsection (1)(b) HMRC are told that where there is a prime facie case “the income tax advantage in question is to be counteracted by adjustments.” (Emphasis added). This is prescriptive.

The same is true of the language in section 698 (2) “the adjustments...which… To be made are to be specified in the notice…” (Again, emphasis added).

This, too, is prescriptive, and far more so than the provisions in, for example, section 29 TMA, or in section 34 TMA where the language is “may”.

So HMRC have not just a power, as a duty, to issue a counteraction notice in the circumstances set out in section 698(1).

As we have already seen, the counteraction notice must specify the adjustment required to be made to counteract the income tax advantage, and that adjustment may be specified by an assessment (as in the case of these appellants). The issue of the assessment, therefore, is an integral part of the duty and power that HMRC have to counteract an income tax advantage.

It is unsurprising, therefore, that the time limit is placed on the exercise of that power. That is the 6-year period in subsection (5) which provides that “Nothing in this section authorises the making of an assessment later than 6 years after the tax year to which the income tax advantage relates.”

Without that limitation there would be no limit on the period within which HMRC could assess, something recognised by those debating the issue when the legislation was originally introduced in 1960.

So, HMRC have a duty and power to issue a counteraction notice and to issue an assessment to counteract the income tax advantage within 6 years from the date of the tax year in question. The restriction in subsection (5) must be construed in the context of the powers in section 698(1) and (2). The former obliges HMRC to issue a counter notice and an assessment. The latter provides that they cannot do so if more than 6 years elapsed from the end of the tax year to which they relate. This is a coherent and self-contained regime.

And we are afraid, for the appellants, that in these circumstances we cannot see any room for the application of the TMA, and in particular the more restrictive time limits in relation to assessments set out in section 34 TMA. There is no need for them. The regime in section 698, when considered as part of the TIS regime, does not permit HMRC to issue an assessment simply because they think that a taxpayer has obtained an income tax advantage (as would be the case if HMRC make a discovery and assess under the TMA). It includes a somewhat cumbersome process which is designed to safeguard a taxpayer, and requires; HMRC to issue a preliminary notification; following which a taxpayer may supply a statutory declaration in opposition to that notification; a process whereby an opposed notification needs to be brought before the tribunal which must certify whether there is a prime facie case for HMRC to take further action; the issue of a counteraction notice.

This suggests two things to us. The first is that the TIS regime is self-contained and highly specific. The second that is it is unsurprising that an assessment issued pursuant to a counteraction notice is qualified by the regime itself and not by the TMA.

There is no suggestion in section 698 that it is specifically subject to the provisions of the TMA. And section 698(7) makes it clear that apart from the limitations in subsection (6) no limitations in the Income Tax Acts are to be read as limiting the powers to issue a counteraction notice (and an amending assessment under section 698). It seems to us that by including this provision, the draftsman is attempting to reinforce the message that the TIS regime is self-contained not just as regards imposing liability, but also in respect of its charging mechanism.

Whether the draughtsman has succeeded in this is something that we consider below. But it fortifies our view that there is no need to look beyond section 698 when it comes to consider the regime which applies to assessments made thereunder.

Furthermore, the TIS regime has its own section (section 705) dealing with appeals. This provides, notably, that on appeal, the taxpayer may challenge not just the counteraction notice but also the adjustments directed to be made (in other words an assessment). And the tribunal has power not just to vary affirm or cancel counteraction notice, but to affirm vary or quash the assessment. These powers are identical to those in section 50 TMA which again suggests to us that there is a parallel regime to the TMA appeal regime under the TIS regime and so the provisions of the TMA which deals with appeals are not required.

It would seem strange to us that if the assessments were governed by the provisions of the TMA, there could be two routes of appeal against the assessments. One under the TIS regime and the other under section 31(1)(d) TMA. We do not believe that Parliament intended this. It is our view that Parliament intended the TIS regime to deal with liability, the charging of any liability, and the appeals process, in isolation, and without the need for the involvement of the TMA.

It is our conclusion, therefore, that the TIS regime, and in particular section 698, provides HMRC with a power to issue a counteraction notice and a corresponding assessment within 6 years from the end of the relevant tax year. There is no further limitation imposed on HMRC by, for example, section 34 TMA. In the circumstances of this appeal, where the assessments were made within 6 years of the relevant tax year, it is our further conclusion that they were valid in time assessments.

We have also considered this issue from the “TMA end”. In other words, if we are wrong, and the TMA does have some residual impact (and in particular if section 34 TMA takes precedence over the provisions of section 698) what the consequences might be.

As submitted by Mr Winter, section 34 TMA is “Subject… to any other provisions of the Taxes Acts allowing a longer period in any particular class of case…”. It is his submission therefore that it must be subject to the TIS regime (as the ITA is a Taxes Act) and in particular the six-year limitation period in section 698(5).

As we understand Mr Gordon’s argument, his position is that that subsection does not “allow” a longer period. It is not permissive, but restrictive. It simply says that any assessment must be made within 6 years but does not permit HMRC to make an assessment at any time up to the end of that 6-year period.

Our view, as set out above, is that the six-year period in section 698(5) must be seen as part of TIS charging regime, and in particular the duties and powers imposed on HMRC by section 698(1) and (2). Section 34 TMA is subject not just to the six-year period in subsection (5) taken in isolation, but in the context of subsections (1) and (2). Those subsections are permissive (indeed, as mentioned above, they are more so than the “may” obligation imposed, by for example, section 29 TMA). A combination of these subsections does “allow” a longer period for an assessment made under section 698.

And so, we are with Mr Winter on his secondary position. If there was a role for section 34 TMA in a consideration of the time limits within which an assessment under section 698 must be made, then section 34 TMA itself would be subject to the extended six-year time period set out in section 698(5).

When considering the foregoing, we are also conscious of two further principles. The first is that set out in the Rangers case (2017 UKSC 45) in which Lord Hodge said :

It seems to us that something very similar applies to the interaction between the TIS regime and the TMA. When one looks at the relevant provisions in historical context, as Mr Gordon has asked us to do, it seems clear that the legislative code governing assessments made under the TIS regime may not be a seamless garment, and there may indeed be loose ends, some of which he has identified in his submissions.

The second principle is that we must give a purposive interpretation to the legislation. And it is our firm view that the TIS regime was intended to be a self-contained regime, which provides a power to assess in section 698(1) and (2), and for these powers to be exercisable within the 6-year window set out in section 698(5).

Where, therefore, there may be a perceived patchwork of provisions across the TIS regime and in the TMA, we have resolved those by adopting this purposive approach in favour of the supremacy of the provisions of the TIS regime.

We now turn to the final point on this issue, namely whether the TMA is an Income Tax Act and thus any limitation under section 34 TMA is ousted by the provisions of section 698 (7). Mr Gordon says the TMA is not such an Act. Mr Winter says that it is.

We have already found that there is no scope for the limited 4-year time period in section 34 TMA when considering the time period within which an assessment can be issued under the TIS regime. And we have also found that if section 34 TMA were to apply, the extended time period in section 698(5) of 6 years would override the 4-year limitation period in section 34 TMA.

And so, we find ourselves in the happy position that we do not have to decide on whether the TMA is an Income Tax Act in order to come to a conclusion on this issue.

We have been treated to some extremely erudite written and oral submissions on this point, which we must confess to have found an extremely difficult one. And in coming to no conclusion, we mean no disservice to the intellectual capital which has been expended by counsel. This is a matter which will no doubt be determinative in another appeal. So notwithstanding that it was fully argued in this appeal, we leave it to those dealing with any such subsequent appeal to come to the relevant conclusion at the relevant time.

We therefore determine the limitation period issue in favour of HMRC.

This document contains full findings of fact and reasons for the decision. Any party dissatisfied with this decision has a right to apply for permission to appeal against it pursuant to Rule 39 of the Tribunal Procedure (First-tier Tribunal) (Tax Chamber) Rules 2009. The application must be received by this Tribunal not later than 56 days after this decision is sent to that party. The parties are referred to “Guidance to accompany a Decision from the First-tier Tribunal (Tax Chamber)” which accompanies and forms part of this decision notice.