First-tier Tribunal (Tax)

 

 

TC07502

 

Appeal number:   TC/2016/06377

 

INCOME TAX – whether under the Income Tax (Pensions and Earnings) Act 2003 amounts paid under “spread betting” contracts were taxable as earnings from employment – or as employment related benefits – or under Part 7 A of that Act

 

FIRST-TIER TRIBUNAL

TAX CHAMBER

 

 

 

	ROOT 2 TAX LIMITED	Appellant
	- and -
	THE COMMISSIONERS FOR HER MAJESTY’S	Respondents
	REVENUE & CUSTOMS

 

 

 

TRIBUNAL:	JUDGE HARRIET MORGAN

 

 

 

Sitting in public at Taylor House, 88 Rosebery Avenue, London on 18 to 21 June 2018

 

Mr Jonathan Bremner QC and Ms Emma Pearce, counsel, instructed by Field Fisher Waterhouse LLP, for the Appellant

 

Mr Richard Vallat QC and Ms Laura Poots, instructed by the General Counsel and Solicitor to HM Revenue and Customs, for the Respondents (“HMRC”)

 

 

 

 

DECISION

 

1.              The appellant appealed against determinations and decisions issued by HMRC for income tax and primary and secondary Class 1 national insurance contributions (“NICs”) which HMRC asserted are due in respect of payments made under financial arrangements entered into by the appellant and its sole shareholders and directors, Mr Blair Forsyth, Ms Shelley Baker and Mr Garry Hughes (the “Individuals”) (made under Regulation 80 of the Income Tax (PAYE) Regulations 2003 (SI 2003/2682) (the “PAYE Regulations”) and s 8 of the Social Security Contributions (Transfer of Functions) Act 1999). 

2.              In short, HMRC’s main argument was that the relevant payments are liable to income tax as “earnings from an employment” within the meaning of the Income Tax (Employment and Pensions) Act 2003 (“ITEPA”) and to NICs under corresponding provisions.  The appellant’s stance was that the arrangements triggered only minimal tax liabilities on the basis that the majority of the Individuals’ returns from them were received as tax free “winnings” from gambling under “spread bets”. 

3.              Unless expressly stated to the contrary, statutory references in this decision are to the relevant section, chapter or part of ITEPA.

Part A Overview

Background

4.              The appellant was incorporated in 2011 to carry on a tax consultancy business.  At all relevant times, the Individuals were the sole directors and shareholders of the appellant. The appeal relates to three sets of “spread bets” and related option arrangements which the Individuals and the appellant entered into in the tax year 2012/13 (the “transactions”). It was accepted that these arrangements were undertaken under a structure known as “Alchemy” which many of the appellant’s clients also undertook. 

5.              The appellant did not disclose the Alchemy structure to HMRC under the provisions of Part 7 of the Finance Act 2004, which requires parties to notify HMRC of certain tax avoidance structures.  However, the tribunal has held that it was required to do so in the case of HMRC v Root2Tax Limited [2017] UKFTT 696. The appellant sought permission for judicial review of that decision, but at the time of the hearing permission had to date been denied (see R (oao Root2 Tax Limited) v HMRC [2018] EWHC 1254 (Admin)). 

Typical transaction

6.              Each Individual and a financial counterparty, Risk Profiles Limited, trading as Heronden (“Heronden”), entered into (a) a “spread bet” contract (a “Bet”) and (b) a  related “call spread option” contract (a “CSO”), the success of which depended on the growth in value of the same reference index of five hedge funds (the “Basket”).  Heronden is a financial bookmaker registered with the UK Gambling Commission which is regulated by the Financial Services Authority and is listed on the London Stock Exchange.  It acted in conjunction with a related financial adviser firm, Aston Collie.

7.              In outline, under each pair of Bets and CSOs which each Individual entered into:

(1)          If the value of Basket grew by a defined amount by the end of a stated period (of around three months) (the “term”), the Individual would “win” under the Bet and make a profit and would “lose” under the CSO and make a loss (and vice versa if the funds failed to grow by that amount): 

(a)          The size of the profit and loss could vary where the growth in value of the Basket fell between the specified minimum or maximum thresholds or barriers of 1.4% and 1.6% (the “lower threshold” and the “upper threshold”).

(b)           However, under each contract, there was, in effect, a cap on the maximum profit or maximum loss which the Individual may realise by reference to a specified sum where the lower or upper threshold was met or exceeded (the “maximum winnings” and the “maximum loss”).   

(c)          The amount of the maximum loss (which drove the level of the maximum winnings and other amounts due under the contracts) was set by the Individual as specified in a letter he or she sent to Aston Collie setting out the parameters of the transaction he or she wished to enter into. 

(2)          In order to participate in these contracts (a) under the Bet, the Individual was required to pay a “Stake” when the contract was made equal to the maximum loss he or she may have to pay under the Bet and (b) under the CSO, Heronden had to pay an “Initial Premium” when the contract was made and a “Final Premium” at the end of the term.  Mr Forsyth gave evidence that the Individuals used their own resources to fund the net amount it had to pay to enter into the contracts (the Stake less the Initial Premium due from Heronden).  By way of example, using the figures taken from the first transaction Mr Forsyth undertook in July 2012 (the “BF July transaction”):

(a)          Mr Forsyth was required to pay £96,502 as a Stake under a Bet under which the maximum winnings were set at £374,400; and

(b)          Heronden was required to pay £11,286 and £74,667, as the Initial Premium and Final Premium respectively, under a CSO under which the maximum loss was set at £468,000.

(3)          If the Individual retained both contracts, the overall financial effect, whether he or she “won” or “lost” on the Bet, was that inevitably he or she would make a net loss.  In that case, the Bet and the CSO were entirely self-cancelling with the only financial consequence being the payment of a fee to Heronden.  Continuing with the example as regards the BF July transaction:

(a)          If Mr Forsyth “won” the maximum winnings under the Bet he would correspondingly “lose” the maximum loss under the CSO and in net terms would make an overall loss of £7,647: (i) under the CSO he would have to pay Heronden the maximum loss of £468,0000 but (ii) he would receive £460,533 from Heronden comprising the maximum winnings of £374,000 under the Bet and £85,953 of the premiums due under the CSO. 

(b)          If Mr Forsyth “lost” the maximum loss under the Bet he would correspondingly realise the maximum winnings under the CSO and in net terms would make a loss of £10,549: (i) under the Bet, he would lose his Stake of £96,502 but (ii) under the CSO, he would receive total premiums of £85,953.

(c)          As noted, in each case the loss represented Heronden’s fee for entering into the transaction plus, where Heronden “won” under the Bet, an amount to cover the betting duty it was liable to pay on its profit. 

8.              In each case, the Individual did not retain both the Bet and the CSO to the expiry of the term.  In fact, very shortly after the parties had entered into the Bet and the CSO and the Stake and Initial Premium had been paid, the Individual’s liabilities and rights under the CSO were novated to the appellant.  In return for Heronden agreeing to the novation, the appellant was required to deposit with it a “Margin” equal to the maximum net amount the appellant may have to pay to Heronden under the CSO.  In the BF July transaction the Margin was £393,333, computed as a sum equal to the maximum loss of £468,000 under the CSO less the Final Premium of £74,667.   

9.              It was common ground that the legal effect of the novation of each CSO was that the rights and obligations of the Individual under the relevant CSO were extinguished and a new CSO was created between the appellant and Heronden on the same terms as applied under the original CSO.  For convenience, in this decision I use a shorthand way of describing this as the novation of the CSO to the appellant or as the appellant taking on the novated CSO and to the resulting new CSO, as the novated CSO.

10.           Prior to the novation, at the Individual’s request, Heronden provided the Individual with a valuation of the CSO as at the date of the novation according to the “bid price”, using the Black-Scholes method of valuation.  In each case, Heronden attributed a very small negative value to the CSO (as the price which the Individual would have to pay a third party to take the contract on) which, as regards the BF July transaction, was £213.  The appellant considered that in taking on an onerous CSO under the novation it provided the Individual with a benefit related to his/her employment with the appellant of an amount equal only to this low amount.  The Individuals accounted for tax on this benefit in accordance with the relevant charging provisions in ITEPA.    

11.           In each case, on the expiry of the term under the Bet and CSO, the Individual “won” the maximum winnings under the Bet and the appellant “lost” the maximum loss under the CSO.  The net result was that, leaving aside the amount representing Heronden’s fee, the appellant was required to pay Heronden a sum which exactly matched the sum which Heronden was required to pay to the Individual.  Using the figures from the BF July transaction:

(1)          The appellant paid £393,333 to Heronden comprising (a) the maximum loss of £468,000 due to Heronden under the CSO less (b) the Final Premium of £74,667 due from Heronden. 

(2)          Mr Forsyth received £385,686 from Heronden comprising (a) the Initial Premium of £11,286 and (b) the maximum winnings under the Bet of £374,400.

(3)          The difference between these two sums represented the fee due to Heronden of £7,647.

12.           In all the transactions under consideration in this appeal, the outcome was as set out in [11] above.  However, in later sets of transactions in 2013 and 2015, the Individuals realised the maximum loss under the Bets and the appellant gained the maximum winnings under the CSOs.  Using the figures from the BF July transaction for illustration only, in that scenario:

(1)          The Individual would make a loss of £85,216 comprising (a) the Stake of £96,502 which Heronden would retain under the Bet less (b) the Initial Premium of £11,286 due under the CSO.

(2)          The appellant would make a profit of £74,667, being the Final Premium due from Heronden under the CSO.

(3)          In that case, in effect the Individual would bear Heronden’s fee of £10,549 (being the difference between £85,216 and £74,667). 

Tax position

13.           The appellant’s view was that, for tax purposes, these arrangements generated only minimal tax charges for the Individuals: (a) an income tax charge on the very small employment benefit which it considered arose as a result of the novation of the CSOs to it (under ss 203 and 204 as set out at [18]) and (b) a capital gains tax charge by reference to the Initial Premium.  In its view, the “winnings” which the Individuals received from Heronden under the Bets were not taxable on the basis that they were receipts from a gambling transaction. 

14.           HMRC argued, in the alternative, that:

(1)          On a purposive approach to the construction of the relevant statutory provisions (as set out in full at [17] below), as regards each transaction (a) the net amounts which the appellant paid to Heronden under the CSOs, comprising in each case the maximum loss less the Final Premium (the “Heronden Payment”), or (b) the amounts which Heronden paid to the Individual, comprising in each case the Initial Premium and maximum winnings under the Bet (the “Bet Profits”), are “earnings from an employment” (within s 10(2)) in respect of which the appellant is obliged to account for income tax under the PAYE system and is liable for primary and secondary Class 1 NICs: 

(a)          As set out by Lord Nicholls in Barclays Mercantile Business Finance Ltd v Mawson [2005] STC 1 (“Barclays”) the correct approach to the construction of tax provisions is to apply them purposively viewing the facts realistically.  In this case, that involves determining the tax consequences of the transactions according to their overall effect on the basis that, on the evidence, they comprised a number of elements intended to operate together as a composite whole with commercial unity.  Viewed in that way, their overall effect was plainly to provide the Individuals with cash as a reward for their employment services in the form of the Heronden Payments or the Bet Profits.  

(b)          The decision in RFC 2012 Plc (in liquidation) (formerly The Rangers Football Club Plc) (Appellant) v Advocate General for Scotland (Respondent) (Scotland) [2017] UKSC 45) (“Rangers”) is authority that the Heronden Payments (i) are not prevented from being “earnings” because the Individuals did not receive those payments and (ii) constitute a relevant “payment” for PAYE purposes on the basis that the Individuals “agreed to, arranged and/or acquiesced” in the various arrangements which resulted in them being made (the “Rangers argument”).

I refer to this issue as the “Ramsay issue”.

(2)          The employment related benefit arising under ss 203 and 204 (see [18]) in respect of the novation of the CSOs to the appellant for no payment is not confined to the small amount produced by Heronden’s valuation.  Rather the Individuals received such a benefit of an amount equal to the Heronden Payments.  HMRC accepted that, if they succeeded on this argument, the resulting income tax would be due from the Individuals.  The relevance to these proceedings is that any such benefit in kind gives rise to a charge on the appellant to Class 1A NICs.  

I note that a tax charge arises under the these provisions only to the extent that the benefit has not already given rise to an amount of earnings taxable under the provisions referred to in (1) or where the benefit arises as a result of a relevant step within Part 7A (see (3) below) (under s 64 and s 554Z2(2) respectively).

(3)          The disguised remuneration provisions in chapter 2 of part 7A apply to the arrangements. 

The detailed arguments and decision on each of these issues are set out at Parts C, D and E (respectively). 

15.           The parties were agreed that the tribunal is not to consider the validity of the Black-Scholes method of valuation. The tribunal was requested to give a decision in principle leaving the parties, in the light of that decision, to seek to agree the figures (including, if relevant, in relation to that valuation).

16.           The Rangers argument was not included in HMRC’s original statement of case.  On 16 May 2018, HMRC applied to the tribunal for permission to amend their statement of case to include this argument.  As the appellant objected, the tribunal decided that this application should be considered at the start of the hearing.  Having heard submissions on this, I decided to approve HMRC’s application for the reasons set out at in Part F of this decision.   

Law

Charge to tax on employment earnings

17.           In outline, under the law in place at the time:

(1)          It was common ground that the relevant amounts are taxable as employment income (under s 6) if they are “earnings from an employment in a tax year” (s 10(2)) where “earnings” are defined as: “(a) any salary, wages or fee”, “(b) any gratuity or other profit or incidental benefit of any kind obtained by the employee if it is money or money’s worth”, or “(c) anything else that constitutes an emolument of the employment” (s 62(2)).  

I refer to these provisions as the “employment tax provisions” and amounts taxable under them as “employment earnings”.

(2)         Broadly, the PAYE system applies in respect of “PAYE income” for a tax year which includes “PAYE employment income for the year” (s 683(1)) as defined to include “any taxable earnings from an employment in the year determined in accordance with [ITEPA] section 10(2)” (s 683(2)(a)).  Under regulation 21 of the PAYE Regulations (made under s 684) an employer is required to deduct income tax from such amounts on making a “relevant payment” to an employee.

(3)         Liability to pay Class 1 NICs on earnings in respect of employment is governed by section 6 of the Social Security Contributions and Benefits Act 1992 (“SSCBA”). Schedule 1 to that Act requires the employer, who pays earnings to an employed earner, to pay both the employer’s and the earner’s Class 1 contributions to HMRC.  It appeared to be common ground that the determination of the appeal in relation to income tax will govern liability to NICs. 

18.           ITEPA contains a discrete benefits code intended to bring non-cash benefits provided by employers to their employees within the charge to income tax:

(1)          The “cash equivalent of an employment-related benefit is to be treated as earnings from the employment for the tax year in which it is provided” (s 203(1)).

(2)           The cash equivalent of such a benefit is the cost of the benefit less any part of that cost made good by the employee to those providing the benefit (s 203(2)). 

(3)          The general rule is that the cost of the benefit is “the expense incurred in or in connection with provision of the benefit (including a proper proportion of any expense relating partly to provision of the benefit and partly to other matters)” (s 204).   

(4)          Where the employee receives a benefit in kind in a given tax year, class 1A (and not primary or secondary class 1) NICs are payable for that tax year in respect of the amount of the benefit which is treated as general earnings (s 10 SSCBA).  

19.            Part 7A of Chapter 2 to ITEPA contains anti-avoidance provisions designed to bring certain payments made by third parties to or for the benefit of employees within the charge to income tax and NICs.   Further details of these provisions are set out in Part E.

Conclusion

20.           In summary, I have concluded that:

(1)           The Heronden Payments made by the appellant under the transactions are subject to income tax as “earnings from an employment” which the appellant is liable to account for under the PAYE system and the appellant is liable to primary and secondary Class 1 NICs in respect of those earnings. 

(2)          If that is not correct, my view is that the disguised remuneration provisions in Part 7A apply to tax the relevant payments made under the transactions as set out at Part E. 

(3)          Determining the “cost” of any employment related benefit which the appellant provided to the Individuals in taking on the novated CSOs raises a valuation issue which it is not appropriate to determine separately from considering the Black-Scholes valuation method adopted. 

Part B Evidence and Facts

21.           I have based the facts found in this decision on (a) the evidence of Mr Forsyth who attended the hearing and was cross-examined, (b) the documents contained in the bundles produced to the hearing and (c) to a very limited extent only the evidence of Mr Barrett, an officer of HMRC, who also gave evidence at the hearing and was cross-examined.  As regards the witness evidence:

(1)          Mr Barrett set out a summary of the transactions typically carried out in the Alchemy structure based on his review of the documents in many implementations of the Alchemy structure and discussions on it with the appellant, its advisers and a number of those who have used it.  This evidence demonstrated, as Mr Forsyth in any event accepted, that the Alchemy planning was widely used under what may be termed a standard process using standard documentation.  Otherwise Mr Barrett’s evidence has no relevance given he had no contemporaneous knowledge of the particular transactions under consideration. 

(2)          Overall, I considered that some aspects of Mr Forsyth’s evidence have to be viewed with caution.  It seemed to me that his answers to factual questions were coloured by his desire to present the Alchemy structure in a way he considered favourable from a legal perspective.  He was somewhat laboured in accepting propositions that were evidently correct according to the economics of the transactions and some of his views lacked credibility in the overall context of all the evidence.  I have addressed this where relevant. 

Background – appellant’s business

22.           The Individuals first worked together in a tax consultancy practice in 2008.  They left that business in late 2009 to establish a new business which evolved into the business carried on by the appellant.  Mr Forsyth said that the appellant’s aim is to develop and implement practical and commercial ways to manage its clients’ tax liabilities and that, in doing so, it seeks to develop long-term relationships with its clients and their advisors. 

23.           Mr Forsyth said he was familiar with transactions such as those under consideration in this appeal from advising on share options and in his personal capacity.  Previously, he spent a significant period of time advising financial services clients.  He noted that Ms Baker had worked at Kleinwort Benson and Mr Hughes had worked in the “Big Four” accountancy firms as well as for an investment bank in New York, Saloman Smith Barney. 

24.           Mr Forsyth said that the Individuals regarded themselves as equal partners in the business.  There were no formal written contracts of employment between them and the appellant but each of them had an understanding of their roles: Mr Hughes was responsible for financial affairs and provided general strategic input into key business decisions, Ms Baker was primarily responsible for liaising with clients, and Mr Forsyth was mainly responsible for liaising with the third parties involved in the Alchemy transactions, primarily Heronden and Aston Collie.  They were all engaged in the business on a full-time basis and in the early years worked 10 to 12 hours a day. 

25.           Mr Forsyth said that, typically, the Individuals received (a) salaries of around £5,000 to £10,000 per year in 2012 and 2013 and £10,000 to £15,000 in 2014 and 2015 which were paid in one instalment in March each year and (b) as shareholders, dividends on an annual basis of around £5,000 to £20,000 each.  They each held one ordinary share in the appellant which entitled them to dividends in equal amounts. 

26.           The appellant received services from around 10 to 15 other employees whose services were provided under an agreement with a related company.  There were three grades of employee who received salaries of around £22,000, £35,000 and £75,000 per year respectively and who could receive discretionary bonuses of up to 25% and 50% of their salaries.  The business development manager, Ms Donoghue, received a salary of around £75,000 and could receive a discretionary bonus of over 100% of her salary.  The Individuals met as directors to decide on any bonus award in December in each year.  They looked at the overall performance of the business, the performance of the individual as regards the business and the individual’s general participation (for example in mentoring).

27.           When setting up the business, the Individuals considered it was fundamental that the fixed costs of the appellant were kept to a minimum.  The employees were paid salaries which did not exceed the market rate which meant the appellant retained the ability to reward high performance with discretionary bonuses.

Development of the Alchemy planning

28.           Mr Forsyth explained in his witness statement that in 2011 he was approached by a number of clients for tax advice in relation to them undertaking what he described as “investment strategies”.  These involved the client entering into a pair of contracts over the same reference index which determined the outcome of the transactions (such as a currency index or a stock market) but which were to be paid or delivered in gilts.  He understood that the use of gilts was likely to be because profits derived from gilts are not subject to capital gains tax.  The individual would have (broadly) an equal and opposite position on the specified index and would then divest himself of the contract that was unlikely to be successful.  The relevant contracts contained cancellation terms that would enable the individuals to extricate themselves from the potentially onerous contracts. 

29.           To gain a full understanding of this proposal the Individuals discussed them with people they knew at a number of financial institutions (such as Schroders Private Bank, Investec Private Bank and Kleinwort Benson).  Mr Forsyth thought that part of what was being proposed could be undertaken as a financial “spread bet” and so he spoke to a contact at Heronden.  He thought that took place in 2011.

30.           He said that the Individuals were concerned that the “investment strategy” was not suitable for recommendation to clients as their view on the tax analysis was different to other advisers.  He thought, for example, that the cancellation terms reduced the market value of the contracts artificially.  However, the Individuals thought that it may be possible “to devise an investment strategy that we would then be able to provide tax advice on”. Mr Forsyth said that the most important consideration was that “any investment strategy was genuine and commercial” and that:

“it was fundamental that the outcome of the trade could not, in any way, be pre-ordained.  Nor could there be any features written into the contracts to depress the open market value.  Any individual who undertook the trades would be exposed to the risk of real economic gains and losses.  The individuals would have to be prepared to assume real (and potentially sometimes significant) risks when entering into transactions with the aim of generating a potential return.”  

31.           Mr Forsyth said that the Individuals could not identify counterparties who could devise suitable “trades” by which he meant financial transactions that “would consistently beat market expectations…..where its market value was low, because the market felt that its succeeding was an unlikely outcome”.  Eventually, however, through discussions with Heronden “an interesting solution was developed”.  Mr Gledhill at Heronden indicated that it was not necessary that the counterparty itself “beat the market”.  Rather it could rely on the ability of hedge fund managers to “beat the market” so enabling the individual “to leverage off their skill and expertise” and:

“It would be possible for clients to adopt a financial investment strategy that was based on the skills of hedge fund managers and profit from their performance”. 

32.           Mr Forsyth noted that (a) Heronden was established to provide investors with efficient access to the capital markets, primarily through the mechanism of spread betting and (b) its employees included a number of former hedge fund managers.  He thought that, on the basis of these factors and, from his discussions with individuals at the firm, Heronden was well placed to identify appropriate hedge funds to be used as the reference index for the Bets and CSOs. 

33.           He noted that he had started spread betting in 2000 and from 2010 had an account with Sporting Index that he used when he thought that there was an opportunity to make a profit.  He believed that “most of the clients of the [a]ppellant either participated directly in spread betting (as a leveraged way of gaining access to the capital markets) or, at least, were aware of the concept”.  Based on his experience of spread betting he considered “it could be used as an investment strategy, or as part of an investment strategy”.  He said that after discussions with Heronden, it was agreed also that CSOs should be used for the other side of the transactions.

34.           The structure or, as Mr Forsyth would put it, the “investment strategy” which resulted from these discussions is that known as Alchemy.  Mr Forsyth confirmed at the hearing that the transactions were undertaken under that “strategy”.  He also confirmed that, as a director/employee of the appellant, he was involved in the design and operation of Alchemy, such as by providing draft board minutes to users, and in giving tax advice on it, for which he was paid in that capacity.  He said that Alchemy was used by many of the appellant’s clients at “a best guess” he thought by “a couple of hundred clients”.  He agreed that what he undertook personally under the transactions was the same as what the appellant’s clients undertook under the Alchemy structure and that he and the clients received the same advice from Aston Collie (as set out below).  He said he thought that was a reason why the appellant agreed to the novation of the CSOs because it regarded “it as a helpful thing for its employees to be seen to be taking its own medicine, as it were”.  He confirmed that the appellant told clients that the appellant was doing this. 

35.           Mr Forsyth said that the Individuals did not formally give themselves tax advice in respect of the transactions.  He said that the Individuals “understood very well the tax consequences of…agreeing to the novation” as they had discussed the position with counsel and received written opinions on the issues so that they understood the process very well.  It was put to him that he understood the advice in both his capacities (as an individual and as a director of the appellant) and did not need advice in either capacity.  He said that the appellant did take tax advice, (although not in the respect of the specific transactions in issue) but “the company obviously ….in order to confirm its view of the transactions would have taken advice from leading counsel on more than one occasion”.

36.           I have commented on this further below, but I note at this stage that the presentation of the scheme as an “investment strategy”, when viewed in the context of the design and operation of the scheme as summarised at [7], is inherently implausible:

(1)          For the avoidance of doubt, I do not suggest (nor did HMRC argue) that the Bets and CSOs and related transactions involved in this structure were a sham.  It is accepted that the parties entered into real contracts which viewed individually, as a legal matter, produced the legal and commercial effects they provided for. 

(2)          However, I do not accept that the Individuals’ or the appellant’s purpose in entering into them was to speculate under financial spread bets as the description of the Alchemy structure as an “investment strategy” suggests.  For all the reasons set out below, they were specifically designed to enable the parties to extract cash from the appellant, in the amounts in which it wished to remunerate the Individuals for their employment services, in a form which was intended to be virtually tax free.  At this stage, I note that:

(a)          An individual can hardly be described as entering into a pair of Bets and CSOs of the type used in Alchemy as an investment given that if the contracts were held to term the individual would inevitably make a loss.

(b)          Moreover if, as must realistically be the case, the individual intended from the outset to divest himself of the CSO, as the contract he expected to “lose”, any investment or, more accurately, trading strategy would be dependent on him being able to sell the contract for no or minimal cost.  However, given the design of the CSO as an onerous contract, it is improbable that anyone other than a party who wished to benefit the individual, such as his employer, would be willing to take accept a novation of it on that basis. 

(3)          That the structure is devised for such tax avoidance purposes does not necessarily of itself mean that the resulting payments are subject to income tax tax and NICs.  As set out below, the question is whether, on a purposive construction of the relevant provisions, the relevant payments are “earnings from an employment” (and whether there is a “payment” of them for PAYE purposes).  The point I am flagging here is that, as is something of a theme in the documents recording the transactions and in Mr Forsyth’s evidence, the parties were at pains to present their purpose as being to speculate under spread bets but, in my view, that is not reflected in reality as is indicated at the outset by the very design of the contracts (and as is confirmed by the totality of the evidence). 

Advice from Ernst & Young and Grant Thornton to the appellant

37.           Once the planning was developed, as well as taking tax advice, the appellant took valuation and accounting advice on the “strategy” in order for it to advise its clients on using it.  The fact that the appellant took this advice and the scope and nature of the advice given clearly indicates that the scheme was devised on the basis that employees would enter into Bets and CSOs as a pair of transactions with a view to novating the CSO to the relevant employer on the basis that a low (negative) value could be ascribed to the CSO at that point in time (thereby, in the appellant’s view, minimising the employment related benefit triggered by the novation). 

38.           On 13 September 2011, Ernst & Young provided the appellant with a report on the suitability of the valuation method used by Heronden in respect of the CSO.  The scope of the report was stated to be as follows:

“[The appellant] provide[s] advice to clients regarding the tax treatment of an investment in an Option.  The owner of the Option will be required to estimate the amount the Option would fetch in the open market and may be required to provide evidence that the price paid is determined on this basis.

[The appellant] ha[s] asked us to …..Comment on the methodology used by clients to value the Option and confirm whether it would be appropriate to be used by participants in the open market to bid for the contract in question…..Confirm an example calculation of the valuation for a given set of inputs.

In undertaking the valuation we have relied on the information and explanations provided to us by you. We have not sought to verify the accuracy of this information for which you are solely responsible.”

39.           Having considered the position in detail Ernst & Young concluded as follows:

“The Black-Scholes model is an appropriate model for pricing options of this nature.  More sophisticated models may exist but due to uncertainties in the volatility of the underlying funds and the illiquid nature of the instruments, these are unlikely to lead to a more accurate answer and hence are unlikely to be used.

The price of the Option obtained by using [the appellant’s] client’s valuation methodology is based upon a historic volatility estimate.  Whilst the true volatility of funds of this nature may be higher, we believe it unlikely that a market participant would systematically pay more than the price implied by historic volatility for options of this nature on a consistent basis without a specific offsetting exposure or additional information regarding the funds in question.” 

40.           On 10 March 2012 Grant Thornton wrote to the board of directors of the appellant as regards the accounting implications of the structure for an employer which took on a novated CSO from an employee (the “Grant Thornton letter”).  They set out the scope of their advice as follows:

“to highlight the key accounting implications…for the proposed transaction to purchase a derivative instrument (ordinarily in a loss position) by the company from an employee.  You have specifically asked us to provide advice on how the transaction will be accounted for and disclosed under UK GAAP….

We understand that an individual will be taking out two derivative instruments (predominantly option contracts) that initially offset the risk of each other…..The company would not be party to or have any knowledge of the transaction with the bank….Early in the contract duration (c. 10 days) the employee will determine which option they wish to hold and will novate the other option to a UK Limited company.  For the employee the option they wish to hold will give them the potential for a small loss and for a significant fixed profit. 

For the company owning the option it will potentially be out of the money and lead to a substantial loss if the option reaches maturity in line with that anticipated by the employee, or a small profit if the market moves against the employee’s expectations.

In taking on the instrument, the company will be taking it on as if it were onerous with the anticipation that it will be making a loss (the company will seek financial advice on the scope of the loss).  The cost of taking this contract will be deemed to be remuneration by the company to the employee….

The commercial substance behind this scheme is to provide an alternative form of remuneration to the employee by acquiring the onerous contract…” (emphasis added)

41.           It is plain from the above that Grant Thornton advised on the basis that, as the appellant must have told them, under the Alchemy structure the employee would novate the CSO to the employer in order to enable the employer to remunerate him.  This is also apparent from their conclusions on the accounting position:

“The company is taking on a contract that in all likelihood will result in a loss on maturity.  As such it should be accounted for as an onerous contract…and a provision included being the lower of the cost to exit or breach the contract and the cost of fulfilling it.  Our understanding is that the company taking on the contract will seek financial advice as to the scope of the potential loss and this is likely to be a reasonable measurement for the cost of fulfilling the contract.

With the company entering into the scheme to provide an alternative form of remuneration, we deem it appropriate to show the cost of fulfilling the contract as remuneration…

With the company knowingly taking on a financial liability in the form of a probable onerous contract, the directors of the company will need to consider if they are adhering to their fiduciary duties under CA 2006.  We recommend that legal advice is sought on this matter prior to proceeding with any transaction.” (emphasis added)

42.           The bundle also contained an example of the advice letter which the appellant sent to its clients regarding Alchemy (the “tax letter”).  Mr Forsyth accepted that this was a typical or standard letter which the appellant sent to all its clients who used the planning.   The main points made in the tax letter are as follows:

(1)          It was noted that (a) the appellant’s tax advice was based on an opinion from Tax Counsel (b) the client had approached the appellant “for advice on the tax consequences of using financial contracts to produce profits, some of which should be tax free” and (c) the client had informed the appellant that he was in the process of engaging Aston Collie to provide advice in relation to such financial contracts, and that he may shortly be entering into contracts with a suitable counterparty.

(2)          It was noted that the client intended to enter into a Bet and  CSO with Heronden and that the appellant understood that “you are undertaking these contracts entirely of your own volition, and that your employer has not advised, encouraged, or facilitated you in taking out these investments” and that after doing so:

“you may approach a third party to determine whether they would agree to the novation to them of the CSO.  The rationale behind this is to maximise your overall potential profit, as advised by your IFA, because the third party would be agreeing to take on some of the risk.  Whilst it is appreciated that you may wish to seek to novate the CSO to a third-party there is no prior agreement or arrangement with any third party

The outcome of the Transactions is not a certainty.  The financial contracts have several possible outcomes depending on commercial factors, and the decisions you make having undertaken the Transactions yourself will result in the opportunity to make various levels of profit, or indeed losses.”

(3)          It was noted that (a) the individual proposed to take out a Bet and a CSO with the same counterparty, (b) the CSO “will suffer a loss if a defined reference barrier is hit” and “the Bet will make a profit if the same reference barrier is hit”, (c) “prior to the reference barrier being hit, you may seek to novate one of the contracts (the CSO) to a third party for value” and “you might seek to novate one of the contracts (the CSO) to a third party, possibly to your employer, either for value or no value”. 

(4)          It was noted that the individual’s “chances of maximising the overall profit” from the transaction “would be enhanced if the CSO could be novated to a third party…because the third party would be agreeing to take on some of the risk” and:

“Any third party would be likely to require value in return for accepting the novation of the CSO. The valuation of the CSO at novation may be different from, and potentially much less than, the profit that may be derived from the [Bet].

There may be additional benefits to you if your employer would agree to act as the third party accepting the novation of the CSO.  You should bear in mind that any benefits provided to you by your employer are taxable as employment income and therefore although your employer might agree to participate…there may be tax charges on you…”

(5)          It was noted that the events are fully described in a letter of advice that Aston Collie would provide, a draft of which had already been provided to the appellant.

(6)          The tax advice was, in summary, that (a) a capital gains tax charge would be due on the Initial Premium, (b) if the employer decided to accept the novation without payment, it would provide a taxable benefit based on the market value at the point of novation, (c) any other benefit would come from retaining the Bet, and (d) winnings from Bets are tax free. 

(7)          It was noted that it was important to obtain a valuation of the CSO at the novation date and that Heronden would provide this.

(8)          The appellant identified certain risks which included that (a) the “barrier that determines the value of the contracts is not nil” which would mean that the individual would suffer a tax charge on the novation but receive no benefit from the Bet and in fact would suffer an overall loss” and (b) that HMRC would enquire into the arrangements.

43.           I note the following as regards this tax letter:

(1)          The advice given again demonstrates that it was integral to the planning that (a) the relevant individual entered into a “matching” Bet and CSO under which he was expected to “win” under the Bet and “lose” under the CSO and (b) that the CSO would be novated to another party. 

(2)          The appellant described a novation of CSO to another party as having the benefit that it could “maximise your overall potential profit” or “enhance” the “chances of maximising the overall profit” from the transaction.  However, in fact it is plain from the design of the Bet and CSO as matched self-cancelling contracts when entered into as a pair, that the relevant individual simply had no chance of making a profit, and indeed was guaranteed to realise a loss unless he divested himself of one of the contracts.  These statements can be described as “window dressing” designed to create the impression that there was some commerciality to the individual entering into the Bets and CSOs as a pair.

(3)          The appellant (a)  referred to the novation of the CSO to the employer as a possibility only which may have some additional benefits, and (b) repeatedly referred to the individual as entering into the arrangements independently, of his own volition and without reference to or the knowledge of the employer company (a theme which also appeared in the Grant Thornton letter and, as set out below, was a theme of Mr Forsyth’s evidence).  For all the reasons set out in my conclusion in Part C, I view this as the appellant playing-down the intention to novate the CSO to the employer; on the basis of all the evidence, in fact the structure was designed on the basis that CSOs would be novated to the employer (as the Grant Thornton letter plainly indicates) and certainly that was the parties’ intention in entering into the transactions in this case.  The wording used in this letter demonstrates merely that it was thought to be important to the success of the planning from a tax perspective, that the novation of the CSO to the employer was not perceived to be a pre-planned outcome. 

Transactions

44.           Each Individual entered into three iterations of the planning in the tax year 2012/13 which were initiated by the Individual entering into a Bet and a CSO with Heronden in July/August 2012 (the “July transactions”), in October 2012 (the “October transactions”) and in December 2012 (the “December transactions”).  

45.           Whilst these transactions are not the subject of this appeal, the Individuals and the appellant carried out further iterations of the Alchemy scheme, on entering into Bets and CSOs on similar terms to those applicable under the transactions in: September 2013, December 2013, December 2014, April 2015, October 2015, December 2015, March 2016 (two iterations) and December 2016.  Mr Hughes was not involved in the iterations from 2014 onwards, because he was no longer a director of the appellant by that point. 

46.           Under all iterations of the planning, the Individual “won” the maximum winnings under Bet and the appellant “lost” the maximum loss under the CSO (thereby realising the Bet Profits and Heronden Payments respectively) except in December 2013 and October 2015, when the opposite position applied.  At the hearing, Mr Forsyth confirmed that the basic structure was the same under each of the transactions undertaken under the Alchemy structure by the Individuals and by the appellant’s clients with the only differences being that sometimes there were different counterparties, different thresholds and a different Baskets of funds.

47.           At the hearing, Mr Forsyth confirmed that the Individuals also all undertook a similar transaction in 2011 but that the relevant CSOs were novated to an employment benefit trust.  He said that “when we came to trade in 2012, it was not an unfamiliar process but it would have been the first time that we would have…novated to a company of which we were employed”.

BF July transaction

48.           The following transaction entered into by Mr Forsyth, the appellant and Heronden in July 2012 is typical of how all of these arrangements operated:

(1)          On 20 July 2012, Mr Forsyth asked Mr John Swallow of Aston Collie for advice on the process and for a suggestion for a suitable financial counterparty. In the letter he said:

“Specifically, I would like to undertake one financial contract that would stand to win £393,333 if the chosen reference market moved over an approximate 3 months period… as anticipated.  I would also like to hedge that investment with a second contract that stands to lose £393,333 if the same movement in the reference market occurred.

The market movement in question should have an approximate probability of success of around 20% but ideally where historical performance has exceeded this level.  However, in order to make one of the contracts as attractive as possible to a third-party, the bulk of any premium payable should be structured as a final premium.”

(2)          On 21 July Heronden wrote to the Mr Forsyth setting out various terms as regards the proposed July transaction and on 22 July 2012 Mr Swallow of Aston Collie replied to Mr Forsyth’s request for advice.  Further details of these letters are set out below.

(3)          On 26 and 27 July 2012 (but in each case with effect from 26 July 2012), Mr Forsyth entered into a Bet and a CSO with Heronden with the expiry and settlement dates falling on 30 November 2012. The CSO documentation was drawn up in accordance with standard market practices, as established by the International Swaps and Derivatives Association (“ISDA”).

(4)          Both the Bet and CSO related to the value of a Basket consisting of five hedge funds which Heronden had identified as suitable on the basis that they had been historically successful and not volatile. 

(5)          Under the Bet, Mr Forsyth made an “up bet” which he would win only if the value of the Basket had grown by more than 1.4% as at the end of the term: 

(a)          The maximum winnings and maximum loss which he could “win” or “lose” were capped at a specific fixed amount if, at the end of the term, the value of the Basket had increased, as regards “winnings”, by more than 1.6% or, as regards a “loss”, by less than 1.4%.  If the value of the Basket increased by between 1.4% to 1.6%, the amount of “winnings” or “loss” was determined under a formula.  Under these arrangements, the amount that Mr Forsyth could “win” was between £21,224 and £374,400 and the amount that he could “lose” was between £2,321 and £96,502. 

(b)          Mr Forsyth was required to pay a Stake of an amount equal to the maximum loss he stood to “lose”, namely, £96,502. 

(6)          Under the CSO:

(a)          If the value of the Basket had increased in value by more than 1.4% at the end of the term, Mr Forsyth was required to pay Heronden a “Payoff” with the precise amount varying according to the precise growth in value but capped at a maximum loss of £468,000 if the value of the Basket had grown by more than 1.6% as at the end of the term.

(b)          Heronden was required to pay to Mr Forsyth (i) an Initial Premium of £11,286 payable on the date on which the CSO was entered into and (ii) a Final Premium of £74,667 on the expiry date. 

(7)          The Initial Premium due from Heronden under the CSO was set off against the Stake due from Mr Forsyth under the Bet, so that when the contracts were entered into Mr Forsyth paid £85,216 to Herondon. 

(8)          On 30 July 2012, Mr Forsyth sent Heronden an email in which he asked for (a) details of the terms on which Heronden would agree to him novating his obligations and rights under the CSO to a third party, (b) a draft novation agreement, and (c) a valuation of the CSO as at 8 August 2012.  Mr Forsyth said that he asked for the valuation as at 8 August 2012 as he thought that would allow the appellant sufficient time to consider the request for it to accept a novation.  He asked Heronden for the valuation because it was obliged to value assets by reference to the most recent market price and it had itself recently “traded” the CSO.

(9)           On 3 August 2012:

(a)          Heronden sent a letter to Mr Forsyth in which it explained that if the novation went ahead, the appellant would take on the liability to pay the Payoff and that it, rather than Mr Forsyth, would receive the Final Premium.  Heronden stated that the CSO had a value in favour of Heronden so that, if the Individual did not pay the appellant to take on the contract, it may have provided a benefit to the Individual. Heronden noted that the Individual had asked what they would pay for the CSO (the Bid Price) on the novation date and set out the following:

 “For clarity if you transact with us after you have novated the original trade, the effect would be to reinstate your original position.  If we were to transact, we would pay an initial premium to you equal to the Bid Price in order to receive the Payoff Amount and pay the Final Premium.  Currently Heronden would be prepared to pay GBP 74880 in respect of the Payoff Amount less GBP 74667 in respect of the Final Premium due ie a total of GBP 213.  Therefore our Bid Price for the [CSO] is GBP 213.

Our price is based on the “Black-Scholes” option pricing model and uses inputs derived from historical performance data.  The Black-Scholes model is commonly used as a pricing model in the banking markets”

(b)          Mr Forsyth wrote to the board of directors of the appellant (as attached to an email addressed to Ms Baker and Mr Hughes) to request it, as his employer, to consider taking on his rights and obligations under the CSO by way of novation.  He pointed out that (i) under the CSO, the appellant would have a risk of losing up to £393,333 (being the maximum Payoff of £468,000 less the Final Premium of £74,667) but that it could make a significant profit if the funds did not match “the performance hurdle” as set out in the CSO, (ii) he had entered into the Bet with Heronden at the same time as the CSO “to hedge my position, which will profit should the funds achieve the same performance hurdles as the option over the same period as the option”.  He said that the appellant should contact Mr Swallow should it require financial advice in relation to the novation. He enclosed a copy of the valuation obtained from Heronden.

(c)          Acting on behalf of the appellant, Ms Baker wrote to Mr Swallow stating that: “We have recently been approached by Blair Forsyth asking us as their employer to consider agreeing to the novation of an option contract entered into by Blair and Heronden dated 26 July 2012”.  She asked for financial advice on this proposal.

(10)      On 6 August 2012 Mr Swallow of Aston Collie wrote to the managing director of the appellant noting the following:

(a)           He had provided financial advice to Mr Forsyth in respect of various financial transactions he had entered into and his tax advisers, the appellant, have provided tax advice to him in respect of those transactions, including the tax consequences of novating the CSO to the appellant. (As set out above, in fact the appellant did not provide the Individuals formally with tax advice but it provided advice to clients when they used the planning.)

(b)          “Mr Forsyth….has identified a way in which the [appellant] could support financial transactions that he has entered into in order to maximise his ability to profit from those transactions.  Any such support provided will produce tax liabilities for Mr Forsyth.  I refer you to the detailed tax advice given by Mr Forsyth by [the appellant] (which in turn refers to advice given by Tax Counsel) which gives their view of the tax position.”

(c)          He set out the range of outcomes for the appellant were it to take on the CSO and summarised the position as follows:  “The best case for the company is that it will not have to pay the [maximum loss of £468,000] in which case it will receive and keep the Final Premium of £74,667; the worst case is that it will have to pay the [maximum loss of £468,000] in which case it receives £74667 but will have to pay up to £468000”.

(d)          “If this were a stand-alone transaction with no corresponding benefit arising to the employee, it would not be a suitable transaction for the company to enter into.”

(e)           Mr Forsyth would receive a benefit on the novation of the July CSO in that it would relieve him of a potentially onerous contract that was on broadly equal and opposite terms to the Bet.  Mr Swallow set out the figures as set out in the letter to Mr Forsyth.   

(f)            “Heronden stands as counterparty to two separate trades and will require margin payments in order to eliminate its credit risk exposure to these two counterparties”. For that reason the appellant would have to put up a Margin of £393,333 at the point of novation.

(11)      On 6 August 2012, Heronden wrote to the appellant explaining that it had been informed by Aston Collie that the appellant was prepared to consider the novation of the CSO and setting out the conditions on which it was prepared to agree to the novation including the need for the Margin of £393,333.

(12)      On 8 August 2012 the appellant held a board meeting which, according to the record in the minutes, was to consider employee reward arrangements for the year ended 31 December 2012 and, in particular, whether the appellant “should agree to the novation of certain financial contracts from employees”.  The minutes record that:

(a)          There was a general discussion of the likely level of profitability of the appellant based on the management accounts, work in progress and projections for the year.  Consideration was given to the total amount that should be allocated by the appellant for awards to be made to employees in respect of their services for the year ending 31 December 2012.

(b)          The directors were asked to consider the request “from certain employees” for the appellant to agree to the novation of financial contracts “that the employees had entered into of their own volition”, and a copy of each of the financial contracts was provided.  The financial contracts considered were those for the July transactions undertaken by all of the Individuals. 

(c)          The directors considered the advice provided by Aston Collie and noted that (i) the maximum liability to the appellant of accepting the novation of the CSOs each Individual had entered into was approximately £1.2 million and (ii) if the Basket did not perform as the employee hoped, the appellant stood to make a profit of approximately £228,000 from the novated contracts, other things being equal.

(d)         The directors considered valuations of the CSOs provided by Heronden which showed the values of the CSOs at novation would be approximately £563 in aggregate.

(e)          The directors confirmed, subject to the members approving their recommendation, that it was the appellant’s preference to agree to the novation of the contracts, but that this would be taken into account in assessing any other remuneration which might be considered to be paid to the employees for the year ending 31 December 2012.

(f)           The directors confirmed that accepting the novation of the CSOs was within the objects for which the appellant was established. 

(g)          It was resolved that the novation of the CSOs “would promote the success of the [appellant] for the benefit of the members as a whole.  Having particular regard to the likely consequences of the decision in the long term and the interests of the [appellant’s] employees it was resolved to agree to the novations”. 

(h)          It was recorded that the members then approved the novation.   

(13)      The appellant and Heronden signed an ISDA Master Agreement (dated 7 August 2012) to govern all future contracts between the parties. 

(14)      Mr Forsyth, Heronden and the appellant entered into a novation agreement, pursuant to which in effect the appellant assumed Mr Forsyth’s rights and obligations under the CSO and deposited the Margin of £393,333 with Heronden (in a client account held with it).

(15)      The appellant treated the value to Mr Forsyth of the appellant taking on the CSO as a potentially onerous contract as a benefit in kind provided to Mr Forsyth.  The value of that benefit in kind was based on the valuation given by Heronden of the CSO on the date of the novation.  This sum was reported on Mr Forsyth’s P11D for the 2012/13 tax year and Mr Forsyth paid tax on this amount through the self-assessment system.

(16)      On 29 November 2012:

(a)          Mr Forsyth received a confirmation email from Heronden stating that the Bet had been successful.  As a result of the growth of the value of the Basket by more than 1.6% by the end of the term, Mr Forsyth had “won” the maximum winnings of £374,400 under the Bet. Consequently, on 30 November 2012, Heronden paid £470,902 to Mr Forsyth, being the aggregate of the winnings and his Stake.

(b)          Heronden informed the appellant that it had lost under the July CSO.  Heronden, therefore, retained the Margin of £393,333. 

Letters from Heronden and Aston Collie to Mr Forsyth

49.            As noted, on 21 July 2012 Heronden wrote to the Mr Forsyth setting out various terms as regards the proposed BF July 2012 transaction and their comments included the following:

“they [Aston Collie] are recommending that you enter into a pair of transactions, a [Bet] and a [CSO], which they have advised are suitable investments for you.

You will notice that there is a margin requirement of GBP 95502 in respect of this [Bet] but since Heronden will pay you GBP 11286 in respect of the Initial Premium on the [CSO], you will only need to make a payment of GBP 85216; this will be held in Heronden’s Segregated Client account until such time as the trade is agreed….

The cashflows that will arise as a result of you entering into these two transactions simultaneously are as follows:

If Barrier B is breached you will win on the [Bet] and Heronden will pay the winnings from the [Bet] of GBP374400 and will also repay the GBP 96502 held on margin.  The equivalent position with respect to the [CSO] is that Heronden will pay you the Final Premium of GBP 74667 and receive from you the [maximum loss of] GBP 468000.

If Barrier A is not breached, you will lose on the [Bet] and Heronden will keep the GBP 85216 you have paid us, but will then pay you GBP 74667 in respect of the Final Premium due on the [CSO].

For values of the Final Basket Level between Barrier A and Barrier B your potential cash flows are summarised in the table below…..

In summary, either you will be down by £7647, or you will lose up to £10544.

We are obliged to report to you any losses exceeding a predetermined threshold.  Since it is inevitable (by design) that you will suffer a loss on one of the two contracts, these maximum losses will be the predetermined thresholds for the purposes of this reporting requirement.”

50.           The letter from Mr Swallow of Aston Collie dated 22 July 2012 (the “Aston Collie letter”) contained the main points set out below as regards the BF July transaction. 

51.            The scope of the advice given was set out to be as follows:

“You have asked me to advise on a financial contract that would stand to win approximately £393,333 if a chosen reference market moved as anticipated, hedged by a second contract that stood to lose approximately the same amount of the same movement in the reference market occurred.  Only by the use of derivative contracts can these financial objectives be achieved…

I refer you to the detailed tax advice you have been given by [the appellant], which in turn refers to advice given by Tax Counsel, and which gives their view of the tax treatment of your undertaking the proposed investments and subsequently novating one of the investments to a third-party.

You have been advised of the tax treatment of “selling an option” whilst simultaneously “entering into a spread bet” by [the appellant].  Both of these financial instruments are regulated by [the FSA], hence the requirement for an authorised and regulated firm to advise on their suitability…

[Aston Collie was instructed to advise on the CSO and the Bet without receiving details of the Individuals’ financial circumstances]. Aston Collie will therefore limit the scope of its advice solely to the suitability of the [CSO] and [Bet] to achieve the outcome intended by your tax advisers (but not whether the tax outcome will be successful).”

52.           Mr Swallow recommended Heronden as the counterparty but explained that he was a shareholder in that company and would therefore benefit from the transactions with Heronden.  He said it was his intention to charge a fee in respect of his work through Aston Collie and that it followed that a conflict of interest might arise but he intended to be clear and transparent in all his dealings with Mr Forsyth and “will be explicit about the costs and charges that you will incur in your dealings with Aston Collie, and in your dealings with Heronden”.  He noted that the parties had agreed that Aston Collie would charge a fee of £1,120 for “advising on the suitability of these contracts to achieve the outcome intended”.

53.           It was noted that the intention was to treat Mr Forsyth as a retail client “whose overall attitude to risk is that of a speculative investor.  That is to say that you are prepared to take significant risk to your capital in order to achieve very high returns considerably above those that you could normally expect from market-based investments”.  Mr Forsyth said that the fact he was treated as a retail client meant that he benefitted from a high level of consumer protections.

54.            Mr Swallow recommended that Mr Forsyth entered into both the Bet and the CSO with Heronden.  He explained that the Bet and CSO were based on a reference index comprised of five hedge funds “which Heronden has identified as having appropriate characteristics and pricing to facilitate this trade”. 

55.           He explained the following as regards the position if Mr Forsyth retained the CSO and Bet for the full three-month period of the term:

(1)          Under the CSO, (a) Mr Forsyth could expect to receive and keep £85,953 (being the Initial Premium and the Final Premium) if the Basket had not grown in value by more than 1.4% on the expiry date but (b) the risk is that he would have to pay 5% of £468,000 for every 0.01% of growth in the value of the Basket between 1.4% and 1.6% and the full £468,000 if it increased in value by 1.6% or more as at the end of the term.  Taking account of the premium he was to receive under the CSO, Mr Forsyth would have a net loss of £382,047.

(2)          The effect of taking the Bet is “to provide an equal and opposite position” to the CSO:

“in other words apart for a few tweaks…and apart from the detrimental effect of Heronden’s costs, having the two contracts together it is as almost as if you had done nothing….

The terms of the [Bet]…have been constructed by Heronden broadly to oppose the terms of the [CSO].  This means that the worst case liabilities you were exposed to under the terms of the [CSO] are netted off by having (more or less) equal and opposite terms on this [Bet].”

(3)          Under the Bet, Mr Forsyth would “win” if the value of the Basket grew by more than 1.4% on the expiry date and lose if it did not do so.  If the value of the Basket had grown by 1.6% or more by the expiry date he would “win” £374,400.  In that scenario, overall Mr Forsyth would make a loss: 

(a)          He would receive £374,400 of winnings under the Bet (and would get his Stake back) plus premiums of £85,953 under the July CSO, making a total of £460,353.

(b)          He would have to pay Heronden £468,000 under the July CSO.

(c)          He would have a loss of £7,467 (£468,000 less £460,353).  This represented Heronden’s fee for the transactions.

(4)          Under the Bet, if the Basket grew in value by 1.4% or less on the expiry date Mr Forsyth would lose the Bet.  Overall, he would a realise a loss of a slightly larger amount:

(a)          He would lose £96,502 under the Bet.

(b)          He would have no obligation to pay the Payoff under the CSO and would receive total premiums of £85,953.  

(c)          He would have a loss of £10,549 (£96,502 less £85,953).  This represented Heronden’s fee for the transactions of £7,647 in and its liability to betting duty due at 3% of the profit it made on the Bet (£2,895).

(5)          If the value of the Basket on the expiry date was between 1.4% and 1.6%, he would win on the Bet and lose on the CSO with the precise amount depending on the precise increase in value as shown in a table.

(6)          It was noted that the range of possible outcomes under both contracts was that if they were held to term “you will be down by £7,647, or you will lose up to £10,549.  This is because, whether you win or lose, Heronden will receive its costs of £7647 in respect of this transaction”. 

56.           Mr Swallow went on to note however, that if two different parties were on different sides of the trade then “significant benefits and liabilities might arise on one side or the other” and:

“You have been advised therefore that there would be a benefit to you if you were able to “novate” the [CSO] to a third party whilst retaining the [Bet] in your own name…..  [in that case the third party would be at risk to pay the Notional of up to £468000…..

Although it may be your intention to ask a third-party to accept the novation of the [CSO], you run the risk that it may not agree to it [but if it does] you would be left owning a [Bet], with all the rights and obligations that attach to it.  The essence of the tax advice provided to you elsewhere is that if you were to “win” under the [Bet], the payment of winnings would not be taxable”. 

57.           It was also pointed out that the act of novation confers value for which a payment may need to be made and that Heronden would calculate that value and that if “you do not make a payment to the third-party, then it may be the case that the third-party has conferred a benefit on you which may have a tax consequence”. 

58.            The effects of the transaction if the CSO was novated was set out to be as follows:

(1)          If the value of the Basket had grown by 1.6% or more by the end of the term, under the Bet Mr Forsyth would “win” £374,400 and correspondingly, under the CSO, the third party would be liable to pay Heronden the full Payoff of £468,000.  In total:

(a)          Mr Forsyth would receive the Initial Premium under the CSO of £11,286 and the winnings of £374,400 under the July Bet, giving a total of £385,686. 

(b)          Under the CSO, the third party would receive the final premium of £74,667 but would have to pay Heronden £468,000, realising a loss of £393,333. 

(c)          Heronden would receive £7,647 (being the difference between the total amount it paid to Mr Forsyth of £385,686 and the net amount it received from the third party of £393,333).

(2)          If the value of the Basket grew by 1.4% or less by the end of the term:

(a)          Mr Forsyth would lose £96,502 under the Bet. However, he would retain the Initial premium paid under the CSO of £11,286, so that his overall loss would be £85,216, (being £96,502 less £11,286).  

(b)          The third party would receive the Final Premium due under the CSO of £74,667 but would have no obligation to pay any Payoff so that it would be “up” by this amount. 

(c)          Heronden would receive £10,549 (being the difference between the net amounts paid to Mr Forsyth of £85,216 and to the third party of £74,667) as its fee plus betting duty.

(3)          If the value of the Basket was between 1.4% and 1.6 % on the expiry date, Mr Forsyth would “win” on the Bet and the third party would “lose” on the CSO according to a range of possible outcomes as set out in a table.

(4)          Overall, therefore, in the worst case, Mr Forsyth would be down by £85,216 although the third party would gain £74,667 and, in the best case, he would be up by £385,686 and third party would be down by £393,333.  In all cases he would have a capital gains tax liability in respect of the receipt of the Initial Premium and, it was noted if the novation was to his employing company, he may also have an income tax liability in respect of the value of the novated CSO and he was referred to his tax advisers for a detailed explanation of these points.

59.           Mr Swallow said that Heronden would require a margin payment (the Stake) in order to eliminate credit risk exposure to its counterparties.  Mr Forsyth’s margin required under the Bet was £96,502, being the most that he could lose under the Bet.  However, as Heronden would expect to make payments totalling £85,953 in respect of the Premiums due under the CSO, he would only have to pay cash of £10,549 when the contracts were entered into.  However, if the CSO were novated:

“you would also novate the payment of the Final Premium to the third party, meaning you would only receive the Initial Premium of £11,286 against the Margin requirement of £96,502; at that point therefore you would have to post a further £74,667 as margin.

The third-party would have to put up margin of £393,333 at the point of the novation.

In practice, and in order to facilitate the novation if you subsequently get agreement from the third-party, you will be required to make an upfront payment of £85,216 to Heronden.  In the event that…you won on the [Bet], Heronden would pay you your winnings (of up to £374,400) and also pay back the £96,502 which had been held on margin.  In the event that you retained the [CSO], Heronden would not require the additional funds and you would be able to ask for them to be paid back to you before the bet matures.”

60.           Mr Swallow explained how the figures were arrived at as follows:

“The option valuation is lower than historical performance would suggest.  This is because, in general, option pricing takes no account of past performance.

You should understand that the FSA demands that Heronden must always trade “on market” which means that the terms of the [CSO], including its price, will always reflect terms that Heronden expects to be available in the wholesale banking markets, and Heronden’s traders have many years of experience trading derivatives contracts of this kind in the capital markets…..

..when the option is initially traded the market is likely to attach a value to the option of around 20% of the Notional ie the pricing suggests a 20% chance that the Final Basket Level will exceed the Barriers.  If you were to express this in the language of betting…you could say that the odds are 4:1, ie that you put down a stake of £1 in the hope that you might win up to £5 if [the Basket] grows between 1.4% and 1.6% and £5 if [the Basket] grows by 1.6% or more.

[The Basket] is made up of five hedge funds which have particular characteristics of performance, correlation and volatility to suit this trade.  So, although past performance is not guarantee of future performance, had you done this trade in the 3-month periods in the 5 years between November 2006 and November 2011, you would have “won” the maximum 50 time ie 83% of the time, because 50 time out of 60 [the Basket] grew by more than 1.6% over the corresponding 3 month period (and one additional occasion grew by more than 1.4% but less than 1.6%).

What you are contemplating doing, on completely commercial terms, is to place “bets”” where the odds are around 4:1 but where historically you would have won many more times than these odds would suggest.  You are relying on the ability of the managers of the underlying funds within [the Basket] to continue, on average between them, to “beat the market” as they have done successfully on very many occasions.”

61.            He concluded with the following references to the need for tax advice:

“I refer you to the detailed analysis of these points provided to you by your tax advisers, and refer you to them particularly to determine exactly the tax consequences of the novation” of the CSO. 

The actual tax liability that arises on the novation will depend upon the actual value of the [CSO] on the date of novation itself, which obviously cannot be known in advance.  Heronden will provide this value to you and your tax advisers once this figure is known.

….I am not in a position to comment on this tax analysis and you must rely on the advice of your tax advisers.  I have sought in this letter merely to explain the language, the mechanic, the cash flows, the economic risks and benefits of executing these two contracts, and to establish whether these “investment products” will achieve the cash flows that you might anticipate.

As you have seen, it is certainly the case that the contracts proposed meet your stated financial objectives and will (subject to the necessary performance of [the Basket]), provide a mechanism which is likely to deliver the appropriate cash flows, but only if a Novation Agreement is subsequently executed and not otherwise.”

Mr Forsyth’s October and November 2012 Transactions

62.           On 3 October 2012, Mr Forsyth entered into a second Bet and CSO with Heronden.  As summarised below, these transactions were made on substantially the same basis as the BF July transactions and substantially the same subsequent steps were undertaken in respect of them.  The only substantive difference is that, in this case, Heronden had the right, at any time during the life of the Bet and CSO, to “restrike” the value of the Bet and the CSO on a mark-to-market basis.  On that basis Heronden was prepared to accept a lower margin payment from the appellant when the CSO was initially novated to it as set out below. 

63.           In outline, under this transaction:

(1)          On 3 October 2012, Mr Forsyth and Heronden entered into a Bet and CSO using the same Basket and thresholds as were used under the BF July transactions for a term ending on 31 January 2013. 

(a)          The Stake under the Bet was £80,150 and the maximum winnings were set at £310,960. 

(b)          Under the CSO, the Initial Premium was £9,355, the Final Premium was £62,034 and the maximum loss was set at £388,700. 

(2)          On 4 October 2012, Heronden produced a valuation of the CSO as at 10 October 2012 of £158 (which the appellant reported as a taxable benefit).

(3)          On 10 October 2012, the CSO was novated to the appellant on the appellant paying a Margin of £15,705. 

(4)          On 16 November 2012, the appellant paid a further Margin in respect of the CSO of £310,960. 

(5)          On 31 January 2013, Heronden confirmed to Mr Forsyth and the appellant respectively that he had “won” the maximum winnings under the Bet and the appellant had “lost” the maximum loss under the CSO.   In net terms, Mr Forsyth received from Heronden a total of £320,315 (£310,960 plus £9,355) and the appellant paid to Heronden £326,666.  The difference between these two sums represents Heronden’s fee.

64.           In the final iteration in question in December 2012:

(1)          On 21 December 2012, Mr Forsyth and Heronden entered into a Bet and CSO using the same Basket and thresholds as previously for a period ending on 30 April 2013.

(2)          Under the Bet, the Stake was £131,019 and the maximum winnings were  set at £508,320. 

(3)          Under the CSO, the Initial Premium was £15,200 and the Final Premium was £101,400.

(4)          On 27 December 2012, the CSO was novated to the appellant. 

(5)          On 30 April 2013, Mr Forsyth “won” the maximum winnings under the Bet and the appellant “lost” the maximum loss under the CSO with the net result that Mr Forsyth received from Heronden £523,619 (£508,320 plus £15,200) and the appellant paid Heronden £534,000.  The difference between these two sums represents Heronden’s fee.

65.           As regards the transactions undertaken by Ms Baker and Mr Hughes:

(1)          In the July/August 2012 iteration:

(a)           Ms Baker entered into a Bet and a CSO on 24 July 2012 on exactly the same terms and in respect of the same amounts as Mr Forsyth did under the BF July transaction.  Heronden provided a valuation of the CSO on 2 August 2012 and the CSO was novated to the appellant on 8 August 2012.  The outcome of the transactions was exactly the same as under the BF July transaction.

(b)          Mr Hughes entered into a Bet and a CSO on 1 August 2012 on the same terms as Mr Forsyth and Ms Baker did in that iteration except that the maximum loss under the CSO was set at £413,333, which lead to differences in the amounts involved as set out below.  Heronden provided a valuation of the CSO on 7 August 2012 and the CSO was novated to the appellant on 8 August 2012:

(i) The Stake under the Bet was £101,430 and the maximum winnings were set at £393,520.

(ii) Under the CSO, the Initial Premium was £11,777 and the Final Premium was £78,567.

(iii) The valuation of the CSO as at 8 August 2012 was £137.

(v) The net result of Mr Hughes “winning” the maximum winnings under the Bet and the appellant “losing” the maximum loss under the CSO was that Mr Hughes received £405,297 (£393,520 plus £11,777) and the appellant paid £413,333 (with the difference representing Heronden’s fee).

(2)          In the October/November iterations:

(a)          Ms Baker entered into precisely the same transaction on the same dates as Mr Forsyth did in that iteration with the same outcome.

(b)          Mr Hughes entered into a transaction on the same terms and dates as Mr Forsyth and Mr Baker except that again Mr Hughes “traded” £20,000 more than they did which lead to the relevant amounts involved being as follows:

(i) The Stake under the Bet was £85,058 and the maximum winnings were set at £330,000. 

(ii) Under the CSO the Initial Premium was £9,926 and the Final Premium was £65,834.

(iii) Heronden valued the CSO at £166.

(iv) The net result of Mr Hughes “winning” the maximum winnings under the Bet and the appellant “losing” the maximum loss under the CSO was that Mr Hughes received £339,926 (£330,000 plus £9,926) and the appellant received £346,666.

(3)          In the December iteration, Ms Baker and Mr Hughes entered into transactions on precisely the same basis and dates and in respect of the same amounts as Mr Forsyth did with the same outcome.

2013 Transaction

66.           In 2013 Mr Forsyth entered into a Bet and CSO which was unsuccessful:

(1)          On 18 December 2013, the parties entered into a Bet and a CSO relating to a different Basket for a period ending on 30 April 2014. 

(a)          The Stake under the Bet was £124,957 and the maximum winnings were set at £484,800. 

(b)          Under the CSO, the Initial Premium was £14,643, the Final Premium was £96,000 and the maximum loss was set at £606,000. 

(2)          On 20 December 2013 (a) Mr Forsyth wrote to Heronden as regards a novation of the CSO on 420 December 2013, (b) Heronden responded with a valuation of £960 and (c) Mr Forsyth wrote to the appellant as regards the novation of the CSO.

(3)          In accordance with the valuation provided by Heronden the appellant reported the sum of £960 as a taxable benefit. 

(4)          On 29 April 2014, Heronden confirmed to Mr Forsyth and the appellant respectively that he had “lost” the maximum loss under the Bet and that the appellant had “won” the maximum winnings under the CSO.  The result was that Mr Forsyth lost his Stake of £124,957 (less the Initial Premium) and the appellant received £96,000.

Appellant’s accounts

67.           The appellant’s accounts for the year ended 31 December 2012 included the following:

(1)          The accounts showed turnover of £7,018,605 and cost of sales of £7,007,047 resulting in a gross profit of £11,558 and, after deduction of expenses, a profit before tax of £5,371. 

(2)          The notes included a statement that (a) the appellant had agreed to accept the novation of CSOs from the Individuals on 8 August 2012 “as part of the directors’ remuneration for the period” and (b) under the novated CSOs the appellant would benefit from profits under the CSOs of a maximum of £227,901 and would be liable for maximum costs of £1,199,999 which the appellant had in fact suffered. 

(3)          It was also noted that the appellant agreed to accept the novation of CSOs from the Individuals on 10 October 2012 and 27 December 2012 “as part of the director’s remuneration for the period”.  The maximum profits and costs under each set of those novated CSO were stated to be (a) £189,902 and £304,200 and (b) £999,998 and £1,602,000 respectively.  It was stated that “because the board of directors believe that these are onerous contracts, a provision of £2,601,998 has been made in the accounts”. 

(4)          The total directors’ remuneration was, therefore, shown as £3,801,994 (£1,199,999 actually paid in the relevant period under the CSOs entered into under the July transactions plus £2,601,998 provided for in respect of the October and December transactions (which were in fact later paid). 

Mr Forsyth’s additional evidence in his witness statement

68.           In his witness statement Mr Forsyth made the additional main points set out below.  I note that to some extent Mr Forsyth gave modified or different evidence when cross-examined at the hearing as recorded at [70] to [139] and the comments below have to be considered in the light of that further evidence (and I have addressed this further below):

(1)          Mr Forsyth said that he “negotiated” the terms of the Bets with Heronden, “in light of his risk appetite, the length of time that he considered appropriate to allow fund managers to seek to grow the funds, and the probability of the Bet’s success”.  He considered these issues in detail in July 2012 and discussed it with Ms Baker.  He decided that “the optimal transaction” for him was a Bet for three months with a 20% probability of success but where ideally the financial performance of the funds historically exceeded that.  He relied on Heronden to select the relevant funds according to the parameters he specified. 

I do not accept that the levels of the “trades” under the Bets and CSOs were set by reference to the Individuals’ risk appetite.  In fact Mr Forsyth appeared to accept at the hearing that the levels were set by reference to the funds in the appellant which were available to remunerate the Individuals (see his evidence at [92], [104] to [110] and my conclusions at [256] to [263]).

(2)          He gave examples of factors which had varied in Alchemy transactions as follows: (a) the underlying funds in the Basket were different, (b) other  counterparties participated (namely Schroders Private Bank and Capital Financial Markets), (c) the “hurdle rate” varied from 1% to 2.25%, (d) in some cases the funds had to on average collectively deliver three consecutive months of positive returns in order to be successful, (e) in some cases the Bet and CSO were taken out for a shorter two month period, and (f) in some cases only an Initial Premium and in others only a Final Premium was paid (or a larger proportion was paid as one or other other).  Mr Forsyth noted that increasing the amount of the Initial Premium reduced the amount of the Stake the Individual had to put up from his/her own resources but on the other hand the decrease in the Final Premium “would make the CSO less attractive to an entity, such as the Individual’s employer, that may agree to the assignment of the CSO” and he noted that the Initial Premium was subject to capital gains tax. 

(3)          He said that there was a real risk of loss.  In the years 2011, 2012 and 2013, the success rate of Alchemy transactions was 62%, 83% and 25% respectively.

(4)          He relied on Heronden to ensure that the Basket was suitable.  Tim Gledhill explained to him that it was important to ensure that the chosen hedge funds were open to investors (as Heronden was required to trade on market), that they had a historic track record of success and low volatility.   He said that it is “an axiom of all investment advice that past performance is no guarantee of future performance.  However, the aim here was to leverage off the skill and expertise of hedge fund managers, and so identifying managers who had beaten the market previously was a way of meeting that aim”.  He noted that the axiom was proved true in that he and other investors had lost significant sums of money on a number of occasions.  He said that as a general rule the higher the volatility the risker the index. 

(5)          There were benefits in entering into a CSO and a Bet at the same time in that this provided a hedge for Heronden with the result that: 

(a)          This reduced the transaction costs.  In the absence of the CSO, Heronden would have had to obtain a hedge elsewhere and the costs of doing so would have been factored into the pricing.  

(b)          This reduced the amount of the Stake he had to fund compared with the position if he had entered into a Bet only.  Heronden advised that, if no CSO was entered into, Heronden would have required a Stake of around three times the amount of the actual Stake.  Under the BF July transaction he would have had to put up a stake of over £200,000 and he “would not have been in a position to undertake such a transaction, even if I had been willing to risk such an amount”.

Mr Forsyth gave consistent evidence on this at the hearing (see [79] and [80]).  However, I do not accept that this indicates that these contracts were entered into by way of speculative spread betting (see [279]).

(6)          He said that he considered the advice from Aston Collie in the Aston Collie letter and decided that he wanted to enter into the contracts recognising that “there was a risk of my suffering real financial loss…but…there was a real possibility that the [Basket]…would increase sufficiently to result in my winning a significant sum from the Bet.  And in the best case scenario ie if I was able to novate the CSO and the [Basket) moved sufficiently, I stood to achieve a very high return from my investment…”

In my view, for all the reasons set out below, it is apparent that from the outset the Individuals intended to novate the CSOs to the appellant and that the appellant would accept the novation provided the level of its potential liability under the CSO was set at an appropriate level by reference to the amounts it wished to provide as employees’ remuneration.  In referring to the novation of the CSO as “the best case scenario”, in my view, therefore, Mr Forsyth was playing-down the probability that the CSO would be novated to the appellant (as is the case also in many of the documents relating to the scheme).

(7)          He and the other Individuals were not financed by the appellant in entering into these transactions and they acted as individuals in doing so in their “own personal capacity” and “independently” of the appellant. 

(8)          He said that “as a general rule” he and Ms Baker used to discuss their general intentions before concluding the contracts but these discussions were conducted in their own capacity and not as directors of the appellant.  They were aware of the appellant’s overall financial position and were “mindful, on broad terms, of what would be an appropriate level of remuneration for each of us”.  Usually Ms Baker entered into contracts first and then he did on the same or a similar basis.  Any discussions with Mr Hughes were on a much more general basis.  He tended to mirror the transactions which Mr Forsyth and Ms Baker did but invested higher amounts.  Until Mr Hughes approached the appellant to accept a novation of his first Bet and CSO Mr Forsyth was unaware of the amount he had traded.  It was only at the board meetings that the novation requests were considered by the Individuals as directors.

(9)          He noted that Heronden was independent of the appellant and would only have become aware that the appellant was the potential assignee sometime after the Individuals had executed their contracts and subsequently approached the appellant concerning the novation of the CSOs.  He said that the Individual approached the counterparty to see what would be needed for the counterparty to agree to a novation and then asked the appellant regarding the novation.  It was entirely up to the appellant whether it was appropriate to do so on each occasion when it was approached and to determine the price it required from the Individual for doing so. 

(10)      In his view there was no guarantee that the appellant would accept the novation or that Heronden would so do.  In that case the Individual would have four options: The Individual could (a) see if Heronden was willing to move the transaction to a subsequent index if the company was not able to enter into a novation at the relevant time (b) hold both contracts to term (c) ask the counterparty to cancel both contracts in which case typically the cost would be the counterparty’s fees or (d) sell the CSO on the open market although “due to the potential costs associated with this option, one of the above three options would have seemed a better choice”.

(11)      He was keen to novate the CSO as soon as possible as he thought that the longer he held it the more likely that it was that he would have to pay out a substantial amount under it.  He was aware that under the Black-Scholes method the valuation was likely to be higher the longer he held the CSO thereby increasing his tax liability as regards the benefit he received on the valuation.

(12)      He said that there was no agreement that “if I or any of the other directors lost his or her bet that he or she would trade again.  Nor was there any agreement that the [a]ppellant would accept a novation of the CSO if he or she traded again”. 

(13)      He said that when the 2013 Bets were unsuccessful none of the directors wanted to trade again until December 2014 when they were reassured by improved performance of the fund managers although by that time Mr Hughes had left the appellant and he did not trade again.  He noted that the appellant’s business manager Ms Donoghue undertook these transactions in December 2013 and asserted that she used her own resources and acted without any prior agreement with the appellant.  She was not paid any sums to compensate her for her loss on the Bet.

Mr Forsyth expanded on the points made at (8) to (13) in his oral evidence as set out below and I have commented on the matters raised in my conclusions in Part C.

Mr Forsyth’s evidence at the hearing

69.           At the hearing Mr Forsyth gave the following evidence.

Terms and effect of the CSO and Bets

70.           As was consistent with his comments in his witness statement, he said that there were a variety of choices in structuring the transactions:

“You could choose a different length of trade periods…a particular hurdle rate…a split between the Initial and Final Premium.  And weighing those up, we decided that a three-month trade which would last 120 days…; the affordability of the stake that was required and the hurdle rate, the likelihood of success, the historical chance of success, which would be factored into the option price.  So weighing all those up, that was why we were determined to go with that bet structure at that particular time.”

71.           It was put to Mr Forsyth that the position under the Bets and CSOs were “matched” save for an amount equal to Heronden’s fees.  Mr Forsyth accepted this but was somewhat laboured in his responses in getting to that position:

(1)          When initially asked this question Mr Forsyth said that “it’s not true to say that they [the financial outcomes under the CSO and the Bet] were exactly matched” on the basis that:

“There was different losses possible, depending on…the outcome of the transactions and they weren’t always equal.  I mean, the financial advice letter lays out a sort of matrix of what would happen if you novated, what would happen if you didn’t novate, then outlines all the potential financial consequences and…if you look at the ones where you hold and where you retain, they aren’t always equal, depending on what you do.”

(2)          He agreed, however, that in the context of the July transactions and generally as regards those who had entered into the Alchemy structure, the relevant employee always entered into both the Bet and the CSO.  He later agreed that the financial outcomes under the Bet and the CSO are “broadly equal and opposite” although he maintained that they were not “precisely” so. 

(3)          He accepted, using the BF July transaction as an example, that if a person held both the Bet and the CSO to the expiry of the term, whether the person won or lost on the Bet, he would incur a loss.  He said that that “was one of the reasons” that “it’s not quite true to say they precisely match and always inevitably lead to a loss” of the same amount.  He said again that there can be different outcomes depending on what happens and how the funds perform and that to his mind “if something is precisely matched” he would “expect all the losses to be the same all the way through”. 

(4)          He agreed, however, that if both contracts were held to term there would be a guaranteed loss attributable to the fee earned by Heronden.  He said “there’s a spread that Heronden or any market maker operates….so if you were to….buy an equity and then sell it immediately, you would inevitably make a loss because of the spread that they charge”.  He agreed that in that scenario a person would make a loss to this extent whether he won or lost the Bet and that the higher loss simply reflected the fact that Heronden had to pay betting duty on profits. 

(5)          When it was put to him that when he said the positions were not “matched” he was only saying that because of Heronden’s fees and betting duty he said:

“No, that’s not quite right, actually.  I think if you look at the….table [being the table set out in the Aston Collie advice letter], you’ll see that the spread bet payouts are different from the payouts under the premium, so I… wouldn’t regard those two as exactly - I would expect those to be exactly the same - when you talk about matched contracts I would expect the numbers to be exactly the same and they aren’t.” 

(6)          It was put to him that it is the net position that is relevant, after all the fees and premiums. He said:

“Yes, but that's a different question to are the contracts matched……And my view is that the contracts are not matched, as this table makes perfectly clear”.   

72.           I find it baffling that Mr Forsyth insisted at (5) and (6) above that the term “matched” as used by counsel meant something different from “matched” in terms of the overall net result under the Bets and CSOs.  In any event this does not detract from the fact that the overall net position was “matched” in that sense, barring the position as regards Heronden’s fees, and that Mr Forsyth accepted that was the case.  

73.           Mr Forsyth agreed that all three sets of the transactions had “a very similar structure” with differences only in the precise details: “there might be a difference in one of the fund managers or the margins might be different but the basic structure would be the same..”.

Novation of the CSO

74.           Mr Forsyth was asked a number of questions about whether it was intended at the outset that the CSO was to be novated.  Whilst again he was somewhat laboured in getting to this position he confirmed, in effect, as is in any event apparent from the economics of the transaction, that it was not possible for the individual to profit from the transactions unless the novation of the CSO took place.  He agreed that a person who implemented the Alchemy planning could not be described as having a “trading strategy” unless he intended to novate the CSO.

75.           He agreed that the Basket and the thresholds were the same under each of the relevant Bets and CSOs.  It was put to him that, therefore, there is no arbitrage between the two contracts.  He said:

 “the whole transaction is based on an arbitrage between the option price, the market value of the option when it’s novated, and your confidence in - and the managers beating market expectations and delivering market-beating performance.  So arbitrage is central to the… whole transaction”. 

76.           When it was put to him that due to the fact that the contracts only ever produced a loss if held to term there was only a trading strategy if the CSO was novated, he initially did not really answer this but said that he “made a profit on the transaction” and so “the trading strategy was successful”.  However, when these questions were put to him again, he agreed that it was correct that “without divesting yourself of the CSO” there was no trading strategy and that the profit arose because of the novation.  He later agreed that under the BF July transaction he put up a significant Stake in respect of transactions that looked at as a pair and, disregarding novation, could only make a loss as set out above.

77.           In response to what he meant in his witness statement by suggesting that the scheme enabled clients to adopt an “investment strategy” Mr Forsyth said the following as regards when he was first asked to advise on these transactions (expanding on what he said in his witness statement as set out at [28] to [31]):

“what clients presented us with was a similar set of contracts but where the market was pricing, the option price, as opposed to the example of 20% we just looked at, might be 95% or 96% or 99%, so virtually guaranteed.  And when they were novated there was a cancellation feature, so that the novatee could effectively extricate themselves from…the  contract.  Now, because of that cancellation feature, when the novation occurred…the market value would be low because the third party could just cancel…the contracts, and we felt that that omission to act conferred a further value on…the individual in question and therefore we couldn’t advise on those transactions and…what became clear from those discussions was that if you wanted to move away from that and have a purely commercial transaction, you’d need to find transactions which the market said were unlikely to be successful, so gave it a very small chance of success.  An option price of, say, 10 or 20%.  And when we first had…those discussions with the entities mentioned there, they said "Well, you know, if we could beat the market we wouldn’t…be helping you with this transaction.  We can’t do that".  And then it became clear that actually you might be able to piggyback on someone else’s ability to beat the market, hence the - the fund managers.  So when I described an investment strategy here, what I’m talking about is identifying fund managers that can consistently beat market expectations and for you to benefit from their ability to… do that.”

78.           When it was put to him in effect that he could not be saying that in entering into the Bet and the CSO together a person could be viewed as adopting an “investment strategy”, because in doing so the person was inevitably going to make a loss.  He said again that it was a “trading strategy” and suggested it was feasible for the individuals to divest themselves of the Bet in the market (although that is not what happened) and that “if you held them both….to the end [it] would have inevitably resulted in what I would describe as modest loss.”

79.           It was put to him that by getting rid of the CSO a few days after entering into it, the individual merely put himself back in the position he would have been in by taking out the Bet only; he was then exposed to the risk of losing the Bet and having to pay out a significant amount.  He said he disagreed with that and said that there were benefits as he had set out in his witness statement (see [68(5)]:

“the reason for that is it would be much more expensive just to take the Bet out on its own.  The risk would have been much greater, so you would have had to put up a significant more margin.  I had discussions with Heronden, for example, and they said "Well, you could undertake a spread bet on its own but the minimum margin that would be required would be £300,000", for example….So by taking them out both at the same time, you’re providing Heronden with….a hedge, so for them it effectively becomes an opportunity to run a spread.  So they require much less margin in taking out the transactions. So it’s a significant….advantage that you’re going to have to  risk a lot less in…taking them both out together.”

80.           It was put to him that entering into the two transactions meant that the individual had to put up less Stake but the outcome is the same and the ultimate risk is the same.  He was insistent that this was not right:

“No, the ultimate risk is much higher if you just take out the Bet because you’re risking a much greater amount of margin.  So, for example, when we lost our Bets I would have lost a minimum of £300,000 as opposed to £85,000 in this example.  So it’s a way… for the individual to minimise the amount that….he needs to put up….. when I had a discussion with Heronden about taking out the Bet on its own, the sort of margin they were talking about was three/four times the amount of margin that….we’d put up because they would have to go and hedge their position in the market itself and by taking them both out simultaneously you’re providing them with an automatic hedge and, therefore, they don’t require as much margin.  They are just governed by their client money rules…….

….by taking them both out at the same time, it was a way to reduce my risk but still gain exposure to the performance of the funds…. if I had just taken the Bet out and not the option, my losses would have been much - potentially much higher”.

81.           Mr Forsyth was asked a number of questions on the pricing and his expectation of success under the Bet.  Overall he accepted that he divested himself of the CSOs because he expected the Bets to win (or at least to have more than a 50% chance of winning) according to the Basket’s historical performance:

(1)          He agreed that the Basket was picked as the reference index for the transactions on the basis that the market viewed the Bet as having a 20% probability of succeeding but that, on the basis of the Basket’s historical performance, he hoped that those odds would be beaten as Aston Collie had advised.  He noted, however, that:

“of course it’s axiomatic that you don’t… rely on historical performance but…if you believe you’ve identified a basket of fund managers that truly can beat the market and deliver…. alpha, which is returns in excess of the market, then, yes, that’s what you would be achieving by undertaking these transactions.”

(2)          It was put to him that this expectation of beating the market under the Bet was the reason why he kept the Bet and divested himself of the CSO.  He said:

“precisely, but with the very large proviso, as the figures in 2013 were - 75% of the time the Bet failed, the barrier failed to be met but…. There’s a real commercial risk attached to the transaction”. 

(3)          Initially he did not accept that he entered into the transactions on the basis that the chances of the Bet winning were far greater than the chances of him winning under the CSO.  He said “the market pricing suggests only a 20% chance of success.  So we were doing it here with… what the market believed would be a very small chance of success..”  He later agreed, however that:

(a)          “what I hoped for is that the fund managers could beat market expectations” and that he believed that “the market was mispricing…the option price” and that was why he retained the Bet.  

(b)          He believed that it was more likely than not that the Bet would win and the CSO would lose, which is why he kept the Bet and divested himself of the CSO. 

(4)          When it was put to him that the concern was to find a market position and a range of funds that gave the right chance of success under the transactions, namely, that the market saw as having a low chance of success but which he believed had a higher chance of success, he said “yes, absolutely”.  When asked if this meant more than 50% chance of success he said “exactly.  It was piggybacking on the fund managers, yes”.

82.           It was put to him that there was no discussion beyond the risk profile he accepted was of concern and the type of investment or the type of fund that a person would be investing in.  He said that “there were other options” but did not indicate that his concerns related to anything other than identifying the right chance of success:

“a large part of my job was to liaise with Heronden and Aston Collie to discuss the type of investment strategies that were available to discuss the funds and understand their thinking.  So….yes, there were other options out there.  In 2011 we’d opted for this and it had been successful and we repeated it again in 2012, basically because…of the success of  2011 and the actual performance of the trades in 2011 had been relatively successful, you know, 80%…but if it had been unsuccessful, it was for example in 2013, we didn’t…use  that trade again because it had been unsuccessful, we didn't have any faith in it.  So there was lots of discussion around what the possibilities were, what funds could be used.”

83.           It was put to him that within the two thresholds under the Bets and CSOs there is a range of outcomes but looking at the historic position the outcome had only been within that range once out of 60 iterations.  He said: “Yes, and I think in actual fact there was only one or two…iterations where it fell within that range”.  It was put to him that meant that effectively the position was binary in practice.  He said that was not the case “because there’s been examples where it fell between the range but there is a much greater chance that it would not fall within that range but in practice it did fall within that range”. He agreed that it did not fall within that range in any of the iterations in dispute but noted that on:

“iterations on which we advised, it did fall within that range….Once or twice…there was only ever 12 transactions a year, so in that year once or twice out of 12….the way the transactions worked, hedge funds only report on a monthly basis……So there could only be 12 a year, that’s what I mean, yes.  So that’s what those success rates are based on is those baskets, were they successful or not, could only be 12 a year”.

84.            He agreed that, in practice, in the great majority of cases, the result was either that the transaction was below the lower threshold or that it was above the higher threshold.  It was put to him that you cannot have a limit on the downside without accepting the limit on the upside.  He said that:

 “the whole point about the way the transactions are structured is it allows the individual to control, to a fair degree of certainty, the amount of risk that he takes on.  So it’s a very good mechanism to…allow them to do that..”

85.           He agreed that in referring to the amount of risk, one element is the limit on the downside.  It was put to him that the disappointing performance of Heronden in 2013, led the appellant to switch all or most of the transactions to using Capital Financial Markets as the counterparty in 2014.  He said that:

“there was a period where clients effectively had a choice and some traded with Heronden, and some traded with Capital Financial Markets, but…..until Capital had proved their track record, because again history is no good, only a fool believes history is a guide to the future, a lot of clients either basically gave up actually and said I’m not trading again and I think in 2013 that’s probably true of about 10% of the clients, and a number decided to wait and see what happened and then didn’t trade with Heronden again and then traded with Capital Markets and we put ourselves in that basket as well.”

86.           It was put to him that the reason the three Individuals did not trade for a period was because they wanted to wait and see that they had found someone who could deliver the results that they wanted.  He said: “Yes, and then, of course, Garry didn’t trade again because he’d left the company”.

Involvement of the appellant and setting the level of the trades

87.           It was put to him that realistically the only people to whom a person might divest himself of the CSO would be an employer or EBT.  He responded that:

“in practice that’s what happened but realistically….you would obviously go to the person that would agree…..to the novation for no consideration first.  So someone who would have an interest in agreeing the novation.  I think you could have gone into the marketplace and divested it to a third party.  It would have been a more expensive option.  You could in theory find a rich uncle to take it off your hands if he wanted to do you a favour, but in practice, yes, I only would have divested to a trust of which I was a beneficiary and my employer but there would be no reason why one couldn’t do it to someone else.  It made sense to go to those places first, because they hopefully wouldn’t charge me any consideration for doing so.”   

88.           It was put to him that the reason it would have to be a rich uncle or someone else who wanted do a favour is that the CSO was likely to lose.  He said that:

 “the market said not, but - the pricing said not, but there was…obviously a significant risk, a serious risk that it would lose.  That’s what in practice happened…. not of course in 2013, where there was…as it transpired an excellent chance the CSO would have made a profit.”

89.           It was put to him that he was viewing the chance of making a profit under the CSO with hindsight and that the advice received before the novation was that the CSO was likely to lose.  He said:

“We went through the kind of advice earlier, which said that….you’re relying on the ability of the fund managers to continue the past performance…So if they continued - if you believed past performance is a good guide to future performance, then, yes, the CSO would lose….Yes, so the advice to the company is different to the advice I received, I agree with that.”

90.           He was shown the section in the advice letter sent by Aston Collie to the appellant where it was stated that if this were a stand-alone transaction with no corresponding benefit arising to the employee, it would not be a suitable transaction for the company to enter into.  He said that was not the advice he received but the advice given to the appellant and noted that it was addressed to Garry Hughes.  He said:

“we absolutely would have considered that advice..[and]…would not have agreed to the novation without having reviewed the…advice…I’m just pointing out this is not the advice that I received on undertaking the transactions, the advice that the company received following my request”. 

91.           He accepted that in all the Alchemy transactions of which he was aware barring a mistake “to all intents and purposes” novation was only ever to an EBT or an employer. 

92.           Mr Forsyth said that he would have only known of the details of Mr Hughes’ trades when he approached the appellant but he was aware he had an intention to enter into them and that Ms Baker also had such an intention.  He said he had a “greater knowledge” of Ms Baker’s transactions than he did of Mr Hughes.  He confirmed that he was “almost certainly” aware that she entered into trades in exactly the same amount as him prior to her approaching the appellant as regards a novation.

93.           It was put to him that the appellant knew all of this as the Individuals were all directors of it.  He said that was not right and “in entering those transactions we were always acting in our personal capacity in a completely separate way from the  appellant”.  When it was put to him that it is wholly artificial to separate the company’s knowledge from the directors’ knowledge, he said:

“the whole point about establishing a company was it’s a separate legal entity and we were all very aware of our duties qua shareholder, qua director, qua employer and I think given our background and training,….we were very, very assiduous in demarcating those duties as individuals and as directors and employers.  So in my opinion, I do not believe that the company had the knowledge of what I was doing as an individual at that time until I had approached it.”

94.           He did not agree that because the Individuals controlled the appellant they could ensure that the novation took place.  He said that:

“I couldn’t ensure that the novation took place, no….Provided certain hurdles were met, there were certain contingencies that we would have to be seen to be doing, but we couldn’t just, for example, agree to any old novation, but if we felt it was the right thing to do, yes, we could approve the novation, that’s right.”  

95.           It was put to him that it was always intended that the novation would take place because that is how the tax advantage is gained.  He said that it “was always intended, I think, that we would approach our employer or… certainly in these 2012  transactions it was my intention to approach my employer first, yes” but he did not agree that it was the intention of all those undertaking the Alchemy scheme, including the appellant, that this is what would happen:

“I don’t believe that the company had anything to do with me taking out these transactions in the first place….The company agreed to the novation, yes.  When I took out the spread bet and the option, I was absolutely acting in my own personal capacity and completely separate from the company.”

96.           It was put to him that the appellant knew in advance that, if it were asked to novate a CSO or similar financial contract in these circumstances, it would be doing so on the basis that the novation was a benefit to him.  He said that he disagreed in respect of the July 2012 transactions because the appellant had never been approached, so it would be the first time it had been asked to consider the novation. But:

 “once July had happened, maybe because it had gone through the process previously, it may have… had a different view of what it’s gone through but in respect of the July transaction, it didn’t know anything about it……until I had been approached the request.  It couldn’t have known what it was going to do about something it didn’t know about, if that makes sense.”

97.           It was put to him that all the directors knew what was going on and the appellant had taken advice.  He was taken to the Grant Thornton letter.  He said that this was:

“different and separate to the transactions which I undertook.  It was… not in respect of the transactions I undertook, it was in respect of giving advice to clients on what the accounting treatment of the novation might be and - but it was not requested in respect of transactions that its own employees may or may not have undertaken or go on to undertake.”

98.           He said that “the company” referred to in the advice is not the appellant.  Grant Thornton were not “talking about specifically [the appellant]” but “about a generic company” and:

“It is pertinent to any company that undertakes the transactions, yes…Including [the appellant], yes….. but it had nothing to do with whether or not it agreed to the novations that [it] may or may not have been approached to buy it by its employees.  It’s having agreed to that, how would I account for it? So it’s got nothing to do with whether or not the novation should occur… I mean when we transact, we’re talking the first transaction in July, and this was obtained in March.  We weren’t - this had zero impact, I think, on whether or not I was going to undertake the transaction.”

99.            He confirmed that Ms Baker and Mr Hughes novated their CSOs to the appellants on the same day as he did in respect of each iteration of the scheme but said that their employee, Ms Jane Donoghue, carried it out “maybe three times in total, possibly” and she did not novate at the same time, and:

“then Jane traded again way before any of us were even considering another trade.  So I don’t think….I traded again until December 2014, I think, from that failure in 2013.  So Jane would have already traded again and won before I’d even considered…… taking other trades.”

100.       He accepted that she used the scheme not just once but several times between 2012 and 2016 and that she was successful on all these occasions bar one in 2013. 

101.       It was put to him that the level of remuneration the Individuals took from the appellant was small disregarding the winnings from the Bets.  It was put to him that, when he said in his witness statement that the appellant retained flexibility to pay bonuses to employees as and when the company was profitable, this meant that there was flexibility to link the level of the directors’  remuneration to the profitability of the company.  He said:

“I think it means generally that we wanted to reward good performance across all employees and have the flexibility to do that.  So we would pay what I would regard as fairly substantial bonuses, discretionary bonuses to employees for good….performance and for the company performing well.….we didn’t pay bonuses to ourselves but….we discussed in 2014, for example, after the failed trade…. we thought there was a real risk there wouldn’t be any possibility of having a successful trade.  We did discuss the possibility of paying larger salaries or paying a discretionary bonus but we didn’t agree on that at the end.”  

102.        It was put to him that they did not agree on bonuses because they wanted the funds in the appellant to enable it to undertake the novation of other CSOs without risking bankruptcy.  He said:

“No,… that’s not right.  One of the reasons that we didn’t actually undertake the transaction in 2014 is there was a bit of a debate about what we should do with those funds.  So, for example, we wanted to put a larger pool aside to pay for contingencies and, you know, that was the kind of main priority, plus there was also…..the very real risk that the trade simply wouldn’t work, so I don’t think any of us individually were that keen on undertaking another trade under….those circumstances.  And then, of course, what happened was in that 12-month period the ownership and the employees changed.  So Garry left in December 2014.”

103.        It was put to him that in fact in 2012/13 the appellant accepted the novation of CSOs that were likely to lose as part of its remuneration strategy.  He said:

“So the company agreed to the novations to reward its employees and it took that into account, so it wouldn’t be appropriate to pay out a significant bonus to an employee where the company already stood to lose, you know, a potentially large sum. And - yes, and of course what happened was the individuals, they stood - the company might have made a profit, as it did in 2013, but there was certainly no agreement, for example, that the individuals would rerun again or if the company won a significant sum of money under the option, as it did do, that would then be paid out in the bonus, for example, if it failed.”

104.       It was put to him that, in order for that approach to work, the level of potential gains and losses on the Bet and the CSO had to be set at an appropriate level.  He said:

“Yes. It would be inappropriate for the company, as you said earlier, to accept the novation which would bankrupt it and the directors couldn’t - simply - well, I wouldn’t have agreed to a novation that would do that, absolutely”. 

105.       It was put to him that the maximum winnings and loss under the Bet and the CSO had to be set at the right level because the CSO is what the appellant is asked to novate.  He agreed that was correct.  He did not explicitly agree that the maximum loss under the CSO depended on the likely level of available resources within the appellant but said he would put it a different way:

“I would say that one would need to be mindful of what you were asking the company to do.  So, for example, if I turned around and said to the company in July 2012, "I'd like you to agree to the novation where you might stand to lose £2 million", for example, that would be a foolish thing to do because I wouldn’t imagine the circumstances in which it would agree to that.  So when we undertook the transactions, yes, we… would have been mindful about what level we thought the company might be able to….agree to, yes.”

106.       It was put to him that that was why in each case the level is pretty much the same.  He said that he effectively “mirrored what Shelley did certainly for the first two transactions in July and October” and when asked if he copied her he said:

 “No….I would have would have wanted to know personally what it would cost me to undertake the transaction and I wanted to know that she would have thought that the - the company could have afforded to agree to that type of novation….. once I’d confirmed those two facts, then yes, I was happy to mirror what she did.. I wasn’t just blindly saying I will do what you will do.  There was some kind of sense check.”

107.       He agreed that Ms Baker picked the figure and, subject to sense checks, he followed.  He said “there might have been some kind of discussion, a bit of to and fro, but that’s basically what happened”.  He said that Mr Hughes did the same “always traded after us, so as we know he did a little bit more than us in the first two transaction”:

“Garry and Shelley had a greater degree of discussions about what level of transaction than I did.  We…weren’t always in the office together very much.  It was difficult to kind of sit down.  That’s where we kind of said, you know, for the board meeting let's sit down and we’ll all be together and we can discuss this properly and agree what the company agree to.  Now, of course there was an understanding that if we all did something sensible, there would be - it wasn’t an agreement, it was a kind of tacit recognition that if we all did something similar, then we might – the company might be minded to agree to it, but there wasn’t an agreement beforehand as to what we would do and what the company would do…I think it’s true to say that we each individually undertook a transaction of our own volition, and we had outline discussions around what that transaction would look like, yes.” 

108.       However, he did not consider any such “tacit understanding” was made in their capacity as directors:

“I don’t believe we were doing it in our…capacity as directors.  Do I believe that we used our knowledge about what we thought the company might be able to agree to?  Yes, of course we did.  But we weren’t doing it…we certainly wasn’t - I wasn’t doing it qua director, definitely not…I mean, I guess the point is when we were - when we sat down, for example, to consider, you know, what’s the appropriate amount of dividend to pay, we did think hard about that because the fully paid-up share capital of the business was I think £3…and we felt it would have been inappropriate to pay significant dividends to shareholders because they weren't really all that involved in the business…You’ll see in the board minutes, for example, we did actually sit, have a break and then go off and have a discussion qua shareholder and say as shareholders do we agree to this.  Now you may think that is an unimportant distinction, but for us it was important.  We…took our duties very seriously and, of course, we advised clients on the same matters and we advised them to take…care in demarcating their roles.  So, yes, we did.  We stopped - had a board meeting, we would stop and then have a break and then reconvene after that, having voted on it qua shareholder.  That literally did happen”.

109.       He said that Garry probably decided to “do a little bit more” than him and Ms Baker because:

“he thought he probably could…..I suppose it may have been in the back of his mind that you would be raising the argument precisely that three of us are equal shareholders and if we’d all done exactly the same thing, it's clearly in line with shareholdings so maybe that was at the back of his mind, I don't know…”

110.       It was put to him that he was saying that Mr Hughes was wondering what HMRC might be arguing five years later.  He said: “Yes, absolutely, yes, he probably was”.  When it was put to him that it was a variation just intended, perhaps, to defeat a HMRC argument, he said: “No, I am just speculating that was in his mind, that’s    all”.

111.       He confirmed that there was no difference between the baskets or the strikes or the barriers between the transactions undertaken by him and Mr Hughes in 2012.  The 2012 transactions were all the same as the Individuals had done in 2011 except that in 2012 the CSO was novated to the EBT and in 2012 it was novated to the appellant.  He thought that in 2013 the Basket would have changed as well. 

112.        Mr Forsyth was taken to the figures in the accounts as set out above.  It was put to him that the Individuals took all their remuneration from the appellant by means of the novation mechanism.  He said that in this scenario:

 “I didn’t take anything out of the company.  What the company was affording me the opportunity to do was make a profit on the Bet that I took out.  So….the whole genesis of this process was me not taking anything out of the company, the company was affording me the opportunity to profit from a Bet that I’d taken out and of course by doing that it was taking on the risk that it might lose money under the CSO.  And so what these figures here represent are effectively….the potential loss I might have suffered under the CSO which it – which…I novated to it.  So none of that money is money that I’ve taken out of the company.”

113.         It was noted that in the advice letter the company was told that the novation was likely to be onerous and effectively could only be justified because it was a benefit to people that the company was entitled to benefit.  He said:

 “Yes.  …one of the principal reasons would be that it was doing so to benefit its employees, yes. Yes…., the company did ultimately suffer a loss as a result of those novations and then subsequently went on to make a profit as a result of certain novations but, yes.”

114.        He agreed essentially that the result under the Bet and the CSO when novation took place was as set out above and in each iteration where he won that result would be the same except as regards any change in the underlying amounts.  He said that:

“Yes, absolutely, yes, and in the same token when -  you know, when you lost…..the stake as well, so you’d lose the initial premium and the cash that you’d put up….It’s gone……I mean, even those figures are - it is a significant risk - well, for me at the time it was a significant amount of money to…. risk, so I didn’t - one wouldn't have taken it lightly and I didn't take it lightly.”

115.       It was put to him that he wanted to win the Bet.  He agreed because:

“otherwise - I mean, that was the only way you were going to make money -and otherwise you would have lost quite a lot money, you say.  Precisely, yes.  So I definitely wanted the bet to win”. 

116.        It was put to him that on the basis of the advice he received from Aston Collie, he not only wanted it to win but expected it to win.  He said:

“Yes, I hoped that…it would win.  In 2013 did I expect the bet to win?  I don’t know, I’m speculating.  I probably would have had a much greater concern for the bet succeeding in December 2013, but I probably thought it was worth a go…because I had seen this catalogue of failures….All those clients had received the same advice but their bets kept failing and they kept suffering losses, because it’s… a very difficult thing to achieve, that market-beating performance.  So always at the back of my mind there was a little nagging - you know, that something…warning me that past performance cannot…be a guide to future performance.  And every time you flip that coin it's a 50/50 chance.  It doesn’t matter if it's been ten times out of ten come up heads, when you come to flip it again it is 50/50…yes, I always I hoped the bet was going to win.  Did I expect it to win?  Maybe not so much in - certainly not so much in December 2013.”

117.       It was put to him that he would not have undertaken these transactions unless he thought there was more than a 50% chance he would win the Bet. He said: “Yes. I mean, I think that’s probably right, yes..”.  It was put to him that in any event, even if he lost, it was not the end of the world because that just meant that the profit accrued to his company.  He said:

“it was significant - I suffered a significant financial loss and it was not easy to take that loss on the chin.  I mean, I certainly wouldn’t have regarded the company’s money as my money and the company's money is not mine to do with what I want.  So I lost and the company won, yes. I think, you know, Garry had that….issue because he lost in December 2013 and left employment before he even had a chance to benefit from another transaction.  And, of course, he would have lost potentially more than I would have lost because he’d done a slightly higher amount….So, yes, I took - and I was very, very conscious that life gets in the way of things.  So something large could have happened.  So say, for example, in January 2014 we got hit with a - I don’t know, a large, unexpected bill or someone sued us or, I don’t know, something like that, there’s no way we would have undertaken a transaction again under those circumstances.”

118.       It was put to him that barring unforeseen events of the sort he referred to, the failure of the Bet would be mirrored by a success on the CSO that would increase the funds available in the appellant to pay remuneration or dividends, or whatever, to the Individuals.  He said: “Yes, but not in precise proportion to shareholdings but, yes, that’s right, yes”.

119.       It was put to him that such funds could be available to him if he wanted to run the scheme again or that the next time the scheme was run he could increase the amounts involved.  He said that the prospect of losing the Bet:

“was cushioned by the fact that the company benefited from that, but there was absolutely kind of no agreement or any kind of - even an understanding we would automatically just run the transaction again. We would have to consider the circumstances, you know, at the time and determine whether that was appropriate or not”. 

120.       He agreed that Mr Hughes’ loss in 2013 was cushioned by the fact that he was paid a significant sum of money when he left the business in the order of £2 million to acquire his shares.

121.       He was questioned as regards his statement that he was not aware of the amounts that Mr Hughes had staked on the first or second iteration.  He said that he “wouldn’t have been aware of the….amounts until I’d seen the contracts”.  He thought, however, that Mr Hughes would have known “from the documentation…what I had staked because I probably would have approached the company prior to him trading.  I think he traded much later than I did…..He would have known after I….he would have known, yes”.

122.         It was put to him that apart from anything else, however, it is not credible that one of the director shareholders would position himself to take £700,000 out of the company without the knowledge of the other directors.  Mr Forsyth said: “Exactly, and that’s why we had a board meeting to determine whether it was - we were to agree to the novation or not”.

123.          It was noted that under the July transaction he and Ms Baker took out £393,333, being the amount it cost the company when it lost on the CSO and that for Mr Hughes it was exactly £20,000 more, £413,333 (and similarly on the second iteration). It was put to him that this clearly indicated that there was co-ordination. He agreed that there was “co-ordination probably on Garry’s part, yes”.  It was put to him that in December 2012 there was clearly co-ordination because the amounts were exactly the same.  He said:

“Yes, there was a - yes, I agree there was a… better understanding - I think I regard the December transaction as different from the July and October transactions, and I say that because there was a closer - it was the end of the year, so we would have had discussions around, you know, paying employees' bonuses, profitability of the company, things like that. So there would have been….more discussions around the financial state of the company around that time, so it was….a different set of circumstances in which we found ourselves at that period of time than we did earlier in the year….different in that…we would have spent time to sit down and discuss the profitability of the company, discuss paying bonuses to employees, so we would have had a greater understanding of what the profitability of the company was.”

124.         He agreed that he meant that he would have had a greater understanding of the amounts available in respect of the December iterations. He confirmed that the appellant never refused a request for a novation of a CSO. 

125.         It was put to him that the intended tax outcome was ultimately to deliver cash into the hands of the employee.  He said:

“No, the intention was to provide the employee the opportunity to profit from the spread bet.  That was the whole genesis and nexus of these transactions.  We…went through a very difficult time when we thought there just wasn’t a way.  I mean, the way in which that would be achieved would be by having a cancellation feature which the company could just…execute whenever they wanted.  This transaction was a very, very different thing.  So I would say it was precisely the opposite of that intention.”

126.       It was put to him that it is obvious that the intended end result of the transaction was that the employee who had taken out the Bet and the CSO would end up with cash in his or her hands of a significant amount.  He said:

“Oh, I agree.  I would refine that by saying hopefully by way of profiting from the spread bet which he had taken out.  That was the - and the company did that by agreeing to the helping the individual divest itself of… the CSO which was in the employee’s view potentially onerous and - so, yes.  So I agreed with what you said and I would just refine it by adding that…the intended outcome was the individual received cash from the spread bet, yes”

127.       It was put to him that the intended tax consequence so far as the individual was concerned was that there would be a relatively small CGT charge on the Initial Premium received under the CSO and a very small income tax charge on the asserted value of the novation of the CSO.  He said:

“I don’t think that was the intended outcome but that was the outcome that we believed arose as a result of the transactions….spread bets are tax-free, like wins, so we would have thought that winnings received would have been tax-free and we would have thought - we agreed that the novation would benefit only the individual and they were taxed on the market value of the benefit when it was… provided.”

128.        It was put to him that anyone entering into these transactions would, on the basis of his advice, expect that tax outcome.  He agreed that was the case if the person won the Bet but said that the “big proviso, though, was the considerable risk of loss”. It was put to him that he and anyone undertaking the scheme would have believed that they were more likely to win than lose. He said:

“Not in 2013, so, I mean, as I say, many clients just gave up.  I mean, they’d lost - when 75% of these things are failing it does test your faith that this thing is going to be successful and going to work”.

129.         He was asked if he was saying that people would have undertaken this Bet expecting to lose. He said that he thought that:

 “I tend to be an optimistic person but I think some people did undertake the bet with the belief that they were probably going to fail again, after failing, you know, three times and that’s why they gave up.  That’s why a significant number of clients just threw in the towel and said "That's it, I'm not going to trade again"”.

130.        It was put to him that the intention was that the employing company who accepted the novation would be able to claim a deduction, not just for the cost of novation but for the total amount paid if it lost the CSO.  He said that followed on from the Grant Thornton advice.  The appellant claimed that deduction which was why its profits were reduced to almost nil. 

Role of Heronden and Aston Collie

131.       It was put to him that Heronden’s position was that come what may it was guaranteed what was described as its fee.  He said:

“I agree, and that’s…exactly the reason that we could put up such a small amount of margin, because it was structured in that way….we provided them with a hedge, precisely”. 

132.       It was put to him that there was not even a credit risk to Heronden.  He agreed that was the case at the outset and broadly speaking that was correct at any stage in these transactions.  

133.       It was put to him that there was a change as regards the transactions after the July transactions to introduce the restrike mechanism but that just reduced the Margin or Stake requirements.  He said:

“yes, I think that’s right…. Let’s just say, for example, Heronden had its money in Santander, okay, and Santander went bust.  Heronden would still be liable to pay out on its transactions and things like that.  But…. I agree that…the restrike was introduced as a credit mechanism to protect their credit risk, yes”. 

134.       He agreed that the restrike position was to protect Heronden because, following novation, it would only be liable to pay the winner on the Bet once it had received the payment from the company that had lost the CSO.

135.       It was noted that he said that Heronden would only have become aware that the appellant was the potential “novatee” after the individuals had executed the transactions but put to him that Heronden was involved at the design stage of the planning.  He agreed but noted that as regards the transactions in 2011 the CSOs were not novated to the appellant:

“So there was no discussion or conversation with Heronden about, "Oh, I'm going to take this transaction and novate it to this company".  It was more, "I'm taking these transactions out, that I've got - one will be novated", but there was no discussion about their identity or anything like that…. Yes, they weren’t aware of precisely what's going to happen, yes.” 

136.       He agreed that whilst Heronden were not aware of precisely what was going to happen, of course Heronden knew how the transactions were generally going to play out.

137.       It was put to him that when he asked Aston Collie to advise a suitable counterpart, he knew they would put forward Heronden.  He said: 

“Yes.  I mean, the reason that we transacted with Aston Collie providing financial advice is (a) we certainly would have wanted Aston Collie to provide financial advice to the company, and (b) I got extra protections, because I was treated as a retail client, so….it made sense to take financial advice on the transactions.  He thought it would be very unlikely for them to recommend someone else given Mr Swallow was involved in them both and as and the two work closely together, so it would have been…a great surprise if John had recommended a different counterparty, yes”. 

138.        It was put to him that he wanted Aston Collie to confirm the details of the contracts but he knew the shape of it.  He thought that was right and added that he wanted them to liaise with Heronden on his behalf, check the contracts, provide advice to him and then act as a go-between between him and Heronden in order to execute the contracts.

139.         He agreed that Aston Collie essentially gave the same advice each time to all those who undertook the transactions with the only differences being in some of the details and the credit mechanism.

“So the details of the contracts might change, the -- but, yes, that’s right.  I mean, the structure of the advice letter was very similar depending on the iteration…. [although] there might have been other changes that would have affected it.  So basket type, trade length, the split of the initial premium to final premium, things like that, so -- but, yes, taking into account all those changes…the advice would have taken into account the changes and details of contracts.”

Part C - Are the Bet Profits or Heronden Payments “earnings”?

Submissions

140.         It was not in dispute that the correct approach to determining if the transactions involved the payment of employment earnings to the Individuals is to apply a purposive construction of the relevant provisions to the facts viewed realistically as summarised by Lord Nicholls of Birkenhead in Barclays and endorsed most recently  by the Supreme Court in Rangers. 

141.       To re-cap, HMRC submitted that on a purposive approach:

(1)           in the alternative, (a) the Heronden Payments or (b) the Bet Profits are “earnings from an employment” within the employment tax provisions.  On a realistic view, the overall effect of the transactions was to deliver cash to the Individual as a reward for their employment services; and

(2)          the appellant is required to account for income tax in respect of those earnings under the PAYE system and is liable for primary and secondary Class 1 NICs in respect of them. 

142.       As noted Mr Vallat relied on Rangers as authority that (a) the Heronden Payments are not prevented from being “earnings” by the fact that the Individuals were not entitled to receive those payments and (b) there was a “payment” of those earnings for PAYE purposes on the basis that the Individuals agreed to, arranged and/or acquiesced in the various transactions which resulted in the Heronden Payments being made.  As set out in further detail below, in that case the Supreme Court held that the fact that a football club’s employees’ remuneration was routed through a trust arrangement did not prevent it being taxable as their employment earnings in respect of which the club was liable to account for income tax under the PAYE system (and that NICs were due). 

143.       In outline, Mr Vallat considered that the following main conclusions can be drawn from the evidence in support of this analysis:

(1)          The Individuals entered into the CSOs and the Bets as a pair of “matched” transactions with the intention from the outset of novating their rights and obligations under the CSOs to the appellant.  Entering into the contracts otherwise made no sense for the Individuals given that if they held them to term they would inevitably have made a loss. 

(2)          The appellant knew of the Individuals’ intention from the outset and was willing to accept a novation of the CSOs provided the arrangements provided an appropriate level of remuneration for the Individuals.  The Individuals were the controlling minds of the appellant and any knowledge they had should be attributed to it. In any event, by the time of the transactions in 2012 the appellant was well aware of steps involved in the scheme. 

(3)          It is the parties’ expectations that are relevant to the analysis rather than the actual results of the Bets and CSOs: 

(a)          It is apparent, in particular, having regard to the Aston Collie letter, that it was intended that the appellant would “lose” on the CSOs and that the Individuals would “win” on the Bets (albeit that in practice there was a period when this did not occur).

(b)          Each set of Bets and the CSOs could produce outcomes on a spectrum where the growth in value of the basket was between 1.4% and 1.6% but this was a very unlikely outcome.  The likely outcome under both the Bet and the CSO was effectively binary. 

(c)          It is not credible that, as Mr Forsyth said, some taxpayers might have undertaken the Alchemy scheme hoping for a “win” under the Bet but expecting a “loss”.  No one would enter these transactions expecting to “lose” on the Bet and transfer value into the company.  In any event, that is certainly not what these Individuals did. 

(4)          The Individuals set the level of the Bets and CSOs by reference to the amounts which the appellant was able and willing to pay under the CSOs so that it would accept the novation.

(5)          The appellant accepted the novation of the CSOs in the knowledge and expectation that it would lose under them, in particular, given the advice it received from Aston Collie.  It accepted the novation of the potentially onerous CSOs because it wanted to provide remuneration to the Individuals.

(6)          The fact that the potential losses which the appellant was willing to take on in order to remunerate the Individuals were roughly equal as between the three Individuals (except as regards Mr Hughes first two sets of transactions) must have been because the Individuals discussed appropriate amounts.  Following the December iteration, the appellant’s profits for the year ended 31 December 2012 were reduced to almost nil (£5,371 before tax).

(7)          Whilst HMRC accept that the risk of losing the Bets was a real one, it was a contingency that the parties were willing to accept in the interests of the scheme and the risk was limited:

(a)          When the level of that risk became too high to be acceptable, the Individuals waited until they had found a counterparty who could deliver the appropriate results and had a proven track record of doing so.   

(b)          If the outcome was not as planned, the Individual’s personal loss under the Bet was offset in effect by the corresponding gain realised by the appellant under the CSO.  Barring unforeseen circumstances, the appellant’s gains would be available to fund the payment of remuneration or dividends to the Individuals or a further iteration of the scheme. The participants could simply run the scheme again until they got the right result and, following the “failures” in 2013 and 2015, that is what they did. 

(c)          Whilst Mr Hughes did not participate in any iterations after December 2013 because he ceased to be a director in December 2014, his loss on that transaction was cushioned by the sum paid to him by Mr Forsyth and Ms Baker upon leaving the company.

(8)          Heronden and Aston Collie played a standard role in the transaction.  Heronden’s net position was effectively neutral; it bore no risk and simply took a fixed fee for entering into the pair of transactions.  Its role was to enable cash to pass from the appellant to the Individuals.      

144.       Mr Vallat submitted that on the basis of the Scottish Provident Institution v IRC [2004] 1 WLR 3172 (“Scottish Provident”) the risk that the Individual would “lose” the Bet and that the appellant would “win” under the CSO is to be disregarded for the purposes of identifying the overall legal effect of the composite transactions.  In that case, the House of Lords decided that a commercially irrelevant contingency could be disregarded in assessing whether a transaction formed a composite whole under the WT Ramsay Ltd v IRC [1982] AC 300, (1981) 54 TC 101 (“Ramsay”) line of cases.  Whilst the chance of an adverse outcome under the Bet and CSO might have been higher than the chance of the relevant contingency occurring in that case, the position has to be looked at in context.  An important factor is that, as noted above, even if the Individual lost under the Bet, the funds did not go to a third party; they went to the appellant (apart from as regards Heronden’s fee). 

Appellant’s case

145.       The appellant’s stance was that each of the Bet, the CSO and the novation forms a separate transaction with its own distinct legal and commercial consequences, which must be respected on their own terms (by contrast with Griffin v Citibank Investments Ltd [2000] STC 1010).  On that approach:

(1)          Neither the Heronden Payments nor the Bet Profits were “earnings” for income tax and/or NICs purposes on the basis that, under the relevant caselaw, they simply did not derive from the Individuals’ employment (see Hochstrasser v Mayes [1960] 38 TC 673 (“Hochstrasser”), Abbott v Philbin [1961] AC 352, Tyrer v Smart [1979] 1 WLR 113, Shilton v Wilmshurst [1991] STC 88, Wilcock v Eve [1995] STC 18 and Kuehne and Nagel Drinks Logistics Ltd v HMRC [2012] EWCA Civ 34, [2012] STC 840 (“Kuehne”)).

(2)          In fact, as is apparent from these cases, these amounts are from a non-employment source, namely, the contractual rights and obligations under the Bets and the CSOs.  In Abbott v Philbin it was held, in effect, that an employment tax charge arose to an employee on the grant of a share option to him but that there was no further employment tax benefit when the option was exercised; any further benefit arose from his rights under the option and did not relate to his employment.  This establishes, therefore, that once a taxable benefit has been provided by the grant of an asset, further fruits from that asset do not and cannot constitute “earnings”.  The link with employment ends at that point.  

(3)          In this case, as a result of the novation of the CSOs, the Individuals’ rights and obligations under the original CSOs were extinguished and the risks and rewards of the new CSOs were taken on by the appellant, thereby generating  taxable employment benefits for the Individuals.  The effect of the novation, therefore, is that the payments under the Bets and CSOs are not linked with employment. 

146.       Mr Bremner drew very different conclusions from the evidence to the conclusions drawn by HMRC and placed emphasis on different factors.  In outline, he made the following main points:

(1)           The Individuals entered into the transactions on their own account, without the involvement of the appellant, using their own resources and having taken independent advice, on terms that reflected their own requirements and with the benefits Mr Forsyth identified. 

(2)          It is highly relevant to the analysis that the outcome of the Bets and CSOs was entirely uncertain. The risk of a loss under the Bets was very real and significant and in fact materialised in 2013 and 2015. Following the losses in 2013, the Individuals did not enter into any relevant transactions for around a year as they had lost confidence in the fund managers’ ability. There was no certainty that there would be further trading in those circumstances and there were no arrangements for that to occur.  Whether there would be further trades depended entirely on the particular circumstances at the material time (such as confidence in the funds and whether there were sufficient resources to meet Heronden’s liquidity requirements).

(3)          The level of the returns under Bets and CSOs did not reflect any decision by the appellant (or any agreement between the appellant and the Individuals) as to the quantum of remuneration to be awarded.  This is evident from the fact Mr Hughes invested larger amounts than the other Individuals (as reflected his own investment decision). There was no decision by the appellant to pay remuneration to each Individual at a particular level; rather, the appellant would take into account its maximum potential losses and its actual losses under the CSOs when it came subsequently to consider each Individual’s remuneration.  The reward provided by the appellant was removing the risk inherent in the CSOs that a sum of money might have to be paid to the option holder.

(4)          There was no certainty that the CSOs would be novated to the appellant and no arrangement with the appellant that it would.  The Individuals may have hoped and even expected that this would happen but that does not make that outcome preordained. Whether it would take place depended on whether Heronden would agree to it and the decision of the appellant’s directors.  In this context it is notable that (a) each Individual approached the appellant separately as regards the novation, (b) the appellant took advice on the consequences, (c) the appellant’s decision to accept the novation was only made once the directors held a board meeting to consider whether to do so on the basis of the advice, the appellant’s resources (as to whether it could satisfy Heronden’s liquidity requirements) and in light of the directors’ duties and (d) in fact the CSOs could have been novated to a third party.

147.       In support of this view of the facts, Mr Bremner stressed that, in his view, Mr Forsyth was clear that he entered into the transactions in his personal capacity and that he distinguished when he was acting in his personal capacity and when he was acting as a director of the appellant.  He said that the fact that the Individuals may have been mindful of what would be an appropriate level of remuneration simply shows that they were realistic in the requests that they proposed to make of the appellant.  He submitted that the fact that a company may be attributed with the knowledge of its directors does not mean that the company has made a decision or that there is an arrangement in place as to what the company would do.  

148.       Mr Bremner said that, having regard to the facts as set out above, the circumstances of this case are not akin to those in Scottish Provident. The risks of losing on the Bets and winning on the CSOs were not “commercially irrelevant contingencies” artificially inserted into arrangements in order to prevent the various transactions from being treated as a single composite transaction which the parties effectively disregarded.  There is no evidence that the parties proceeded on the basis that these genuine and significant risks could simply be ignored. 

149.       He said that the fact that the Individuals hoped to win the Bets does not alter the fact that the Individuals entered into them and were genuinely exposed to the risks and rewards of doing so.  In his view, it is not permissible, as is the effect of HMRC’s approach, simply to telescope together the Bets and the CSOs and apply the tax code as if the appellant had paid the Individuals directly thereby ignoring the commercial effects of transactions.  Outcomes that the parties want to happen or consider likely to happen are not to be treated as if in fact they have happened for tax purposes.  Those risks were an inherent part of the position taken by the Individuals and the appellant (see McLaughlin v Comrs [2012] SFTD 1003 at [137], Blenheims Estate and Asset Management Ltd v Comrs [2013] UKFTT 290 (TC) at [92] to [97] and Gemsupa Ltd v Comrs [2015] SFTD 447). 

150.       Mr Bremner submitted that HMRC are wrong to suggest that it affects the position that an Individual’s loss under a Bet was mirrored by a corresponding gain for the appellant:  

(1)          To permit such an argument leads to an impermissible piercing of the corporate veil.  Money in a company is not a fund that its employees can draw on freely.  The three directors would have to agree collectively on how to deal with any such funds and there could, of course, be tax consequences for the Individuals if such decisions were taken.

(2)          That argument ignores the fact that there were tax consequences of the appellant winning under the CSOs as it would have to pay corporation tax on the profit.  On that basis this structure would not be fit for purpose as a scheme to extract money tax-efficiently from the appellant. 

(3)          It also ignores the fact that the amounts that were being “traded” may well not match the shareholdings in the company (as was the case in some instances as regards Mr Hughes).

(4)          In any event this does not inform the issue as to whether there is a composite transaction or whether the relevant amounts constitute “earnings”. 

151.       Mr Bremner added that:

(1)          The fact the Individuals risked their own resources in taking out the Bets and the CSOs contrasts with the position in Schofield v HMRC [2010] SFTD 772  (as affirmed in the Court of Appeal [212] STC 2019) where the fact that the individuals did not use their own monies to generate the disputed losses was a significant factor.  Here the individuals had to find a Stake to enter the Bets and were certainly at risk of loss if the Bets did not win. 

(2)          The fact that Heronden is neutral is not relevant.  Heronden is a financial bookmaker who naturally might be expected to take a position that effectively enabled it to earn fees through the spread.

152.       Mr Bremner submitted that the decision in Rangers has no bearing on this case.  In that case, it was clear that the sums paid through the trust arrangements represented remuneration for employment.  The only question for the Supreme Court was whether it affected the tax position that the employees did not have a legal right to receive the monies.  In this appeal, the tribunal is required to assess, in effect at a prior stage, whether the relevant amounts constitute remuneration in the first place.  They do not for all the reasons already given. 

153.         Mr Bremner said that the test is not simply whether money ends up in the Individuals’ hands at the expense of the appellant in some sort of economic sense.  For the reasons set out above, there is no evidence to support the view that the Heronden Payments are derived from the Individuals’ employment.  He described it as unintelligible in that context, to characterise the Individuals as agreeing or acquiescing in the appellant making a payment of remuneration.  It is irrelevant that, in deciding whether to take on the CSOs, the appellant took account of what remuneration might be appropriate for the Individuals in the future.

154.        Mr Bremner added that HMRC’s stance on the Rangers argument, in picking out one element of the transaction (the Heronden payments) is inconsistent with their stance that, on a realistic view of the facts, the overall effect of the transactions must be considered.  In any event, looking at the broader context, the effect was simply that the Individuals were put in the position where they were exposed to the risk only of the Bet.  No view of the facts supports the proposition that the overall effect of the transaction was a payment by the appellant, as employer, of the Individuals’ remuneration to Heronden.  This an unprincipled and misconceived attempt to stretch the decision of the Supreme Court in Rangers far beyond its proper scope. 

Further submisisons

155.       Mr Vallat submitted that the facts and circumstances here, as outlined above, are materially different from those in Abbott v Philbin.  Looking at the broader context, as is required under a purposive approach, it was always intended and expected that the employee should reap as a reward for his services not only the benefit of the novation of the CSO but also the Heronden Payments and/or the Bet Profits.  Unlike in Abbott v Philbin, the reward for the employee’s services was not confined to the immediate event (in this case, the novation of the CSO) but also encompassed the outcome of the CSOs/the Bets.  That decision certainly does not support a general proposition that if an option or other asset is granted to an employee the tribunal or court cannot look at the wider circumstances to determine, on a proper realistic view of the facts, what constitutes the employee’s remuneration. 

156.       As regards Rangers, Mr Vallatt said that he was not suggesting that Rangers itself answers the question of whether the relevant payments were remuneration.  Rangers provides the answer that, if the payments were remuneration, it does not affect the position that they were paid to Heronden rather than the Individuals.  Clearly, matters were set up so that the appellant made the Heronden Payments pursuant to the CSOs, but the tribunal must look at the broader context as to the reasons why the CSOs were entered into.  On the evidence the appellant was willing to pay remuneration of the relevant amount to the Individuals and the Individuals and the appellant together put in place the Alchemy transactions in a clear attempt to change the tax consequences of a simple payment to the employee.  In particular, that the payments are remuneration is very clear from the appellant’s accounts and the accounting advice.  The fact that it was in substance remuneration explains why it is shown as such in the accounts. 

157.       Mr Bremner said that HMRC mischaracterise the effect of the decision in Abbott v Philbin.  The whole point of the grant of the option in that case was to enable the employee to profit from any increase in value of the company.  The tribunal must look at the nature of the benefit provided and cannot ignore the fact that here the benefit lies in the individual being released from the rights and obligations under the CSO and left free to speculate in relation to the Bet.  It is simply a mischaracterisation of the facts to label the outcome of the Bet as the “reward for services”.   

Discussion and decision

Caselaw – meaning of earnings from employment

158.       To re-cap, it was central to the appellant’s case that, on the basis of the caselaw set out below, there was no sufficient link between the Bet Profits and the Heronden Payments and the Individuals’ employment for them to constitute employment earnings.

159.       The starting point is the often-quoted decision of the House of Lords in Hochstrasser v Mayes [1960] AC 376.  In that case ICI operated a scheme under which it made a tax-free loan to an employee to enable him to purchase a house.  Under the terms of the scheme when the employee was later transferred to another place of work and sold his house at a loss, ICI made good the loss.  The majority of the House of Lords held that the payment by ICI in making good the loss was not taxable as an emolument “from” the employee’s employment as a “perquisite” or “profit” (under the legislation then in place) on the basis that it was not a reward for past services.

160.       Viscount Simmonds said, at page 390 and 391, that the test of taxability is “whether from the standpoint of the person who receives it the profit accrues to him by virtue of his office” and acknowledged that fine distinctions may arise in cases where the question is whether a “payment is made to an employee as a reward for his services or…is made out of affection or pity”.  In this case he thought there was little doubt on which side of the line this case fell; the payment was not a reward for services. 

161.       Lord Radcliffe agreed with Viscount Simmonds but added some comments of his own. He also acknowledged, at page 391, in effect that it is not easy to draw the line but noted that (a) the test to be applied “is the same for all, namely that the payment “must arise “from” the office or employment” and (b) in the past several explanations have been offered in the courts as to the significance of the word “from”, such as that the payment must have been made to the employee “as such” or “in his capacity of employee” or “by way of remuneration for his services”, and that this is what is meant by payment to him “as such” but:

“these are all glosses, and they are all of value as illustrating the idea which is expressed by the words of the statute.  But it is perhaps worth observing that they do not displace those words.  For my part, I think that their meaning is adequately conveyed by saying that, while it is not sufficient to render a payment assessable that an employee would not have received it unless he had been an employee, it is assessable if it has been paid to him in return for acting as or being an employee.” (emphasis added)

162.        Lord Radcliffe decided, at 392, that in that case this test was not met because the payment was “in substance a free benefit conceded to his employee”.  In his view, the employer wanted “to ease the mind and mitigate the possible distress of an employee” as regards his housing situation  such that it “was paid to him in respect of his personal situation as a house-owner who had taken advantage of the housing scheme and had obtained a claim to indemnity accordingly”. 

163.       Lord Denning said, at [397], that it was essential to focus on the words of the statute.  The question was simply whether the relevant amount was a profit from the employee’s employment.  He thought that was not the case for the simple reason “that it was not a remuneration or reward or return for his services in any sense of the word”.

164.        The appellant relied particularly on the later decision of the House of Lords in Abbott v Philbin in relation to an employee who was granted an option to subscribe for shares in an employer company, for which he paid £20.  HMRC argued that he should be subject to income tax when the option was exercised (on the substantial difference between the market price of the shares at that time and the amount he paid for the shares (plus a proportionate part of the price of the option)) on the basis that there was no taxable “perquisite” when the option was granted. 

165.       Viscount Simmonds said, at pages 365 to 366, that the option was a valuable right that could be turned to pecuniary account.  He noted that the words “perquisites or profits whatsoever are as wide and general as they well could be”.  He thought there was no relevant limitation of their meaning except in the words of Lord Watson in Tennant v Smith, [1892] AC 150, at page 159, that they:

“denote something acquired which the acquirer becomes possessed of and can dispose of to his advantage - in other words, money - or that which can be turned to pecuniary account.”

166.       He continued that it could not then be said that an option to take up shares at a certain price is not a valuable, or at least a potentially valuable, right:

“Its genesis is in the desire of the company to give a benefit to its employees, and at the same time, no doubt, to enhance their interest in its prosperity.  It is something which the employee thinks it worth his while to pay for: not a large sum, truly, but £20 deserves a second thought. And it is something which can assuredly be turned to pecuniary account…”

167.       At page 366, he rejected a number of arguments that the option could not be turned to pecuniary account; in his view, the point was that the option holder had “a right which is of its nature valuable and can be turned to pecuniary account”. He noted that HMRC’s argument “appeared to demand for its success that the individual did not acquire a perquisite at the date of the grant”.  He said that, on the basis that there could not be one perquisite at the date of the grant and a second perquisite when the shares were taken up, HMRC’s case “fails at the initial step”.  However, he thought that there were “other grave difficulties in the way of its success” in that the increase in value of the shares in future years could not, in his view, be said in any event to relate to a reward for employment services:

“The taxable perquisite must be something arising “therefrom”, i.e., from the office, in the year of assessment.  I do not want to embark on the notoriously difficult problem as to the year to which, for the purpose of tax, a payment should be ascribed if it is not expressly ascribed to any particular year.  But I do not find it easy to say that the increased difference between the option price and the market price in 1956 or, it might be, in 1964, in any sense arises from the office.  It will be due to numerous factors which have no relation to the office of the employee, or to his employment in it.  The contrast is plain between the realised value, as it has been called, of the option when the shares are taken up (though the realisation falls short of money in hand) and the value of the option when it is granted.  For the latter is nothing else than the reward for services rendered or, it may be, an incentive to future services.  Unlike the realised value it owes nothing to the adventitious prosperity of the company in later years. On this ground also I should reject the claim of the Crown.”

168.       The other members of the majority, Lord Reid and Lord Radcliffe, agreed with Viscount Simmonds that the option was a taxable “perquisite” in the year of assessment in which it was granted.  Lord Radcliffe agreed that in any event there could be no taxable event when the option was exercised and Lord Reid thought it was highly doubtful that there was but did not want to express a conclusive view on the point. 

169.       Lord Reid noted, at page 372, that a person exercising an employment is taxable on, amongst other items, “perquisites” therefrom for the year of assessment.  He said that it may be difficult to relate a “perquisite” strictly to a particular year but said that if the option is itself the “perquisite” it would generally be sufficiently related to the year in which it is given properly to be such for that year but if there is no perquisite until the option is exercised and shares issued possibly many years later:

“- in what sense would the shares be a perquisite for the year when they were issued? There would be no relation whatever between the service during that year and the giving of the option many years earlier, or the exercise of the option during the later year. I do not wish to express any concluded opinion on this point, but it does seem to lend support to the conclusion which I have reached on other grounds.”

170.       Lord Radcliffe expressed his more definitive view as follows, at page 379:

“The advantage which arose by the exercise of the option, say £166, was not a perquisite or profit from the office during the year of assessment: it was an advantage which accrued to the appellant as the holder of a legal right which he had obtained in an earlier year, and which he exercised as option holder against the company. The quantum of the benefit, which is the alleged taxable receipt, is not in such circumstances the profit of the service: it is the profit of his exploitation of a valuable right.  Of course, in this case the year of acquiring the option was only the year immediately preceding the year in which, pro tanto, it was exercised.  But supposing that he holds the option for, say, nine years before exercise? The current market value of the company’s shares may have changed out of all recognition in that time, through retention of profits, expansion of business, changes in the nature of the business, even changes in the market conditions or the current rate of interest or yield. I think that it would be quite wrong to tax whatever advantages the option holder may obtain through the judicious exercise of his option rights in this way as if they were profits or perquisites from his office arising in the year when he calls the shares…”

171.       In Tyrer v Smart the House of Lords held that an employee received a taxable employment “perquisite or profit” when he was given the right to subscribe for shares in the parent company of his employer when the parent decided to “go public” at what was expected to be (and in fact turned out to be) a preferential rate.  At page 114, Lord Diplock described it as well established that the relevant test is “whether the benefit represents a reward or return for the employee’s services, whether past, current or future, or whether it was bestowed upon him for some other reason”.  He said that the “borderline may be a fine one” as is illustrated by Hochstrasser and Laidler v Perry [1966] AC 16.  However, he considered that the employer’s purpose may be a relevant factor:

“Where the benefit is granted by and at the expense of the employer or its parent company, as distinct from benefits derived from third parties, such as a huntsman’s field money or a taxi driver’s tips, the purpose of the employer in granting the benefit to the employee is an important factor in determining whether it is properly to be regarded as a reward or return for the employee’s services.  The employer’s motives in conferring the benefit may be mixed and the determination of what constitutes his dominant purpose is a question of fact for the Commissioners to determine.”

172.       Lord Diplock said, at page 116, that the crucial finding of fact was that the company’s purpose in making the shares available to employees was “to encourage established employees of the company and of companies within the group to become shareholders in the parent company” with its aim being “to achieve a better relationship with the employees so that they would become and continue to be loyal employees, having an understanding of and a sense of involvement in the affairs and fortunes of [the group]”. He noted that the Commissioners held that this was an advantage afforded to the taxpayer “in return for acting as or being an employee”, within the meaning of that expression as used by Lord Radcliffe in Hochstrasser v Mayes.

173.       Lord Diplock rejected the view (as Brightman J had held) that there was no taxable benefit as a result of the grant of the right to subscribe for shares to the employee due to the uncertainty when the employee applied for his shares as to whether the price was in fact a preferential one or not.  He concluded that that did not affect the Commissioners’ finding as to the purpose of the offer of shares as set out above which he considered was a clear finding that:

“the offer was made as a reward for past (since he had to have served five years to qualify for the offer) and more particularly for future services and accordingly was made to him in return for acting as or being an employee.”

174.       In Shilton v Wilmhurst the House of Lords held that an amount of £75,000 paid to Mr Shilton by the football club he was then employed with (Nottingham Forest) as an inducement to take a contract with a new club (Southampton) was an emolument of employment with the new club.  In giving the judgement, with which the other Lords agreed, that it was an emolument Lord Templeman in effect endorsed the approach taken by Lord Radcliffe in the earlier Hochstrasser case. 

175.       He said, at page 91d to e, that because the relevant provision embraces all “emoluments from employment” it must therefore comprehend an emolument provided by a third party.  He said that the term applies:

“first to an emolument which is paid as a reward for past services and as an inducement to continue to perform services and, second, to an emolument which is paid as an inducement to enter into a contract of employment and to perform services in the future. The result is that an emolument “from employment” means an emolument “from being or becoming an employee.” (emphasis added)

176.       He considered that the authorities are consistent with this analysis and are concerned to distinguish between an emolument which is derived “from being or becoming an employee”, and “an emolument which is attributable to something else …..for example, to a desire on the part of the provider of the emolument to relieve distress or to provide assistance to a home buyer”.  He said that if “an emolument is not paid as a reward for past services or as an inducement to enter into employment and provide future services but is paid for some other reason, then the emolument is not received “from the employment”” although he appreciated applying this distinction is “frequently difficult and gives rise to fine distinctions” (at page 92e to f).

177.       He noted, at page 92, that the authorities have been concerned with those cases in which it is not clear whether an emolument has been paid to an employee for acting or agreeing to act as an employee or has been paid for some other reason.  Having referred to the passages in Hochstrasser v Mayes set out above, he noted that Lord Radcliffe was dealing with an emolument paid to an existing employee but “applying his words to an emolument paid to a prospective employee “it is assessable if it has been paid to him in return for his agreement to act as or become an employee” (at page 92f).  In the present case Nottingham Forest paid £75,000 as an emolument in return for Mr Shilton agreeing to act as or become an employee of Southampton and for no other reason and he accepted it in return for agreeing to do so (just as he accepted £80,000 from Southampton for the same reason).   It did not matter to him whether these sums were paid by the clubs or by some other third party (at page 92j).

178.       He rejected, at page 94, the view expressed in the lower courts that an “emolument from employment” only applies to an emolument provided by a person who has an interest in the performance by the employee of the contract of service with the employer as that raises difficulties in “defining the “interest” which makes the employee liable to pay tax on the emoluments.  He preferred “the simpler view that an emolument arises from employment if it is provided as a reward or inducement for the employee to remain or become an employee and not for something else” (at page 94h).

179.       In Wilcock v Eve Carnwarth J held that no employment tax was due on a payment to a taxpayer by the group within which he was formerly employed which was made to compensate him for his loss of share option rights when he ceased to be employed (on a management buyout of the employer company).

180.       At page 27d to f, Carnwath J said that the question was:

“whether…. the loss of rights under the share option scheme is “intimately connected with the employment” in the same way as the union rights in Hamblett v Godfrey; or alternatively whether it is to be treated as something distinct, such as the loss incurred on the sale of a house in Hochstrasser v Mayes”. 

181.       He had earlier set out that in Hamblett v Godfrey [1987] STC 60 a payment made by the Crown to an employee for the loss of trade union rights was held to be taxable as employment income, in his view, on the basis that it was “for the loss of rights which were in effect part of the employment” (page 25f to j).   He noted that in that case Knox J drew a distinction with the Hochstrasser case on the basis that the House of Lords there found “a separate source for the payment in question, namely the housing agreement” which dealt with the taxpayer’s position as householder” whereas in the Hamblett case “there is no such independent source other than the Crown’s desire to recognise the loss of rights intimately linked with employment.” 

182.       He continued, at 27e, that he would have found this a more borderline case but for the decision in Abbott v Philbin and he cited and referred to the passages from that case set out above.  He said, at 28d, that (a) that case demonstrates that, apart from specific statutory provisions, the value realised by the exercise of the taxpayer’s option right, had it been exercised at the relevant time, would not have been a taxable emolument and (b) that the Commissioners found that the payment was made in recognition of the loss of that benefit and, therefore, “for tax purposes it should have the same character as that for which it was being given”.

183.       He also considered whether a different statutory provision applied which was based on different wording “by reason of” employment.  Overall, he concluded that there was little difference between the two formulations.  He noted, at page 30b, the formulation of the “therefrom test” set out by Neill LJ in Hamblett v Godfrey [1987] STC 60 at 71 that: “The question is, was the payment of an emolument from the employment? In other words, was the employment the source of the emolument?”   He said at 30c that the difference between such formulations and the expression “by reason” of is hard to detect.  He held that that provision also applied.  

184.       More recently what is required for there to be an emolument “from” employment has been considered by the Court of Appeal in Kuenhe.  The court upheld the tribunal’s decision that (a) sums paid to individuals on their transfer to a new employer company were taxable as earnings from an employment on the basis that they were paid and received as an incentive to work willingly and without industrial action for the new employer and (b) the fact that they were also paid and received as compensation for the loss of the pension scheme previously available to the individuals did not affect the conclusion that they were paid for the services the employees rendered and as  a reward or inducement for future willing service.

185.       At [32] Mummery LJ said the following as regards the word “from”, at [33]:

“All I need say at this point is that the use of "from" in the idea expressed in the statutory expression "earnings from an employment" and "earnings derived from an employment" in a fiscal context indicates, as matter of plain English usage, that there must, in actual fact, be a relevant connection or a link between the payments to the employees and their employment.” (emphasis added)

186.       At [50] Patten LJ noted that what constitutes an emolument or other benefit from an employment has been the subject of judicial analysis for almost 100 years and the court’s task it to apply the statutory test to the fact and not to apply some other test based on a gloss (referring to Hochstrasser).  However, he thought some gloss is inevitable because:

“it is accepted that it is not enough merely to show that the payment was received as an employee and would not have been received if the individual had not been an employee.  Something more must be established.  This has been expressed in terms of the difference between causa sine qua non and causa causans but it does, on any view, require a sufficient causal link to be established between the payment and the employment.” (emphasis added)

187.       He said, at [51], that he thought that the  ways in which that necessary link has been described and analysed in the earlier cases has to be respected even though the ultimate question is whether the “from” question can be answered in the affirmative. He noted that Neill LJ in Hamblett v Godfrey describes those explanations as valuable and authoritative (see page 726 G and H).  He thought that the cases show:

“that the question of taxability involves one being able to characterise the payment as one "from employment" if it derives "from being or becoming an employee" and is not attributable to something else such as a mark of esteem or a desire to relieve distress. I take this formulation from Lord Templeman in Shilton v Wilmshurst….”

188.       He said, at [52,] that it must follow from this that, in order to satisfy the test, one must be able to say that the payment is from employment rather than from a non-employment source.  He said that this has certainly been the approach of the courts in most of the decided cases referring to the comments of Viscount Simmonds in Hochstrasser, Lord Wilberforce in Brumby v Milner [1976] STC 534 at 536 (where he said this is “not an easy question to answer”), Lord Diplock in Tyrer v Smart at 36: (as regards the “determination of what constitutes his dominant purpose”;) and Carnwath J in Wilcock v Eve at page 25 (where he said that where there is more than one operative cause “there is an element of value judgment in deciding on which side of the statutory line the payment falls”).

189.       At [53] he said that this process of evaluation requires the judge to make findings of fact “based on the evidence as to the reasons and background to the payment and then to apply a judgment as to whether the payment was from the employment rather than from something else..” He considered, at [56], that the tribunal judge was “obviously” right to say that “a payment can be from employment even though there are other reasons for it”.  He noted that in all the cases he had referred to there are competing causes.  In each case the payments were in part motivated by feeling of generosity towards the recipient. However:

“Employment does not have to be the sole cause but it does have to be sufficiently substantial as to characterise the payment as one from employment.” 

190.       He added at [59]:

 “If the employment is a substantial and equal cause of the payment, it becomes open to the judge to say that the statutory test is satisfied.  The payment is then from the employment even if it also substantially attributable to a non-employment cause.”

Caselaw - purposive approach to construction

191.       As noted, the parties were agreed that the tribunal must adopt a purposive approach to construction of the meaning of the term “earnings from employment” for the purposes of assessing whether the relevant amounts are subject to income tax.  In Barclays Lord Nicholls set out a detailed examination of the authorities on applying a purposive approach, from the seminal decision in Ramsay onwards.  In more recent decisions in the Supreme Court, such as that in Rangers, Lord Nicholls’ judgement in this case is relied on as the definitive word on this topic.  I have not set out the facts which are far removed from those in this appeal.

192.       At [28] Lord Nicholls noted that as Lord Steyn explained in Inland Revenue Commissioners v McGuckian [1997] 1 WLR 991, 999 the modern approach to statutory construction is:

“to have regard to the purpose of a particular provision and interpret its language, so far as possible, in a way which best gives effect to that purpose”. 

193.       He noted that until the Ramsay case, however, revenue statutes were “remarkably resistant to the new non-formalist methods of interpretation”. The “particular vice” of formalism in this area was “the insistence of the courts on treating every transaction which had an individual legal identity ….as having its own separate tax consequences, whatever might be the terms of the statute”.  He said that as Lord Steyn said, it was:

“those two features - literal interpretation of tax statutes and the formalistic insistence on examining steps in a composite scheme separately - [which] allowed tax avoidance schemes to flourish.”

194.       At [29], he said that the Ramsay case “liberated the construction of revenue statutes from being both literal and blinkered”.  He cited two passgaes from Ramsay:

(1)          First at page 323:

“What are ‘clear words’ is to be ascertained upon normal principles: these do not confine the courts to literal interpretation. There may, indeed should, be considered the context and scheme of the relevant Act as a whole, and its purpose may, indeed should, be regarded.” 

(2)          Secondly at pages 323-324:

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

195.         He concluded, at [30], that the application of these two principles:

“led to the conclusion, as a matter of construction, that the statutory provision with which the court was concerned, namely that imposing capital gains tax on chargeable gains less allowable losses was referring to gains and losses having a commercial reality (“The capital gains tax was created to operate in the real world, not that of make belief’) and that therefore (p 326):

“To say that a loss (or gain) which appears to arise at one stage in an indivisible process, and which is intended to be and is cancelled out by a later stage, so that at the end of  what was bought as, and planned as, a single continuous operation, there is not such a loss (or gain) as the legislation is dealing with, is in my opinion well and indeed essentially within the judicial function.”

196.       Lord Nicholls commented, at [32], that the essence of the new approach was:

“to give the statutory provision a purposive construction in order to determine the nature of the transaction to which it was intended to apply and then to decide whether the actual transaction (which might involve considering the overall effect of a number of elements intended to operate together) answered to the statutory description… however one approaches the matter, the question is always whether the relevant provision of statute, upon its true construction, applies to the facts as found. As Lord Nicholls of Birkenhead said in MacNiven v Westmoreland Investments Ltd [2003] 1 AC 311, 320, para 8:

"The paramount question always is one of interpretation of the particular statutory provision and its application to the facts of the case."” (emphasis added)

197.       He continued that Ramsay did not introduce “a new doctrine operating within the special field of revenue statutes but on the contrary:

“as Lord Steyn observed in McGuckian [1997] 1 WLR 991, 999 it rescued  law from being "some island of literal interpretation" and brought it within generally applicable principles.

198.       He noted, at [34], that unfortunately, “the novelty for tax lawyers of this exposure to ordinary principles of statutory construction” meant there was a tendency to regard Ramsay as establishing a new jurisprudence governed by special rules of its own. He thought that was encouraged by two features characteristic of tax law, although by no means exclusively so:

“The first is that tax is generally imposed by reference to economic activities or transactions which exist, as Lord Wilberforce said, "in the real world". The second is that a good deal of intellectual effort is devoted to structuring transactions in a form which will have the same or nearly the same economic effect as a taxable transaction but which it is hoped will fall outside the terms of the taxing statute. It is characteristic of these composite transactions that they will include elements which have been inserted without any business or commercial purpose but are intended to have the effect of removing the transaction from the scope of the charge.”

199.       He said that there have been a number of cases, such as Inland Revenue v Burmah Oil Co Ltd 1982 SC (HL) 114, Furniss v Dawson [1984] AC 474 and Carreras Group Ltd v Stamp Commissioner [2004] STC 1377 in which it has been decided that “elements which have been inserted into a transaction without any business or commercial purpose did not, as the case might be, prevent the composite transaction from falling within a charge to tax or bring it within an exemption from tax”.  Thus “in each case the court looked at the overall effect of the composite transactions” and:

“On the true construction of the relevant provisions of the statute, the elements inserted into the transactions without any commercial purpose were treated as having no significance.”

200.       At [35] he said that cases such as these gave rise to a view that, in the application of “any taxing statute, transactions or elements of transactions which had no commercial purpose were to be disregarded”.  However, he thought that was “going too far”.  He said:

“It elides the two steps which are necessary in the application of any statutory provision: first, to decide, on a purposive construction, exactly what transaction will answer to the statutory description and secondly, to decide whether the transaction in question does so. As Ribeiro PJ said in Collector of Stamp Revenue v Arrowtown Assets Ltd [2003] HKCFA 46, para 35:

"The driving principle in the Ramsay line of cases continues to involve a general rule of statutory construction and an unblinkered approach to the analysis of the facts. The ultimate question is whether the relevant statutory provisions, construed purposively, were intended to apply to the transaction, viewed realistically."”

201.       He said, at [37], that the need to avoid sweeping generalisations about disregarding transactions undertaken for the purpose of tax avoidance was shown by MacNiven v Westmoreland Investments Ltd [2003] 1 AC 311 and, at [38] that that case shows:

“the need to focus carefully upon the particular statutory provision and to identify its requirements before one can decide whether circular payments or elements inserted for the purpose of tax avoidance should be disregarded or treated as irrelevant for the purposes of the statute.” 

202.       In the same passage he noted that in MacNiven Lord Hoffman drew a distinction between cases where was “a statute laid down requirements by reference to some commercial concept such as gain or loss”, where “it would usually follow that elements inserted into a composite transaction without any commercial purpose could be disregarded” and those made “purely by reference to its legal nature” (in MacNiven, the discharge of a debt) in which case “an act having that legal effect would suffice, whatever its commercial purpose may have been”.  He thought that this is not “an unreasonable generalisation but:

“we do not think that it was intended to provide a substitute for a close analysis of what the statute means.  It certainly does not justify the assumption that an answer can be obtained by classifying all concepts a priori as either "commercial" or "legal". That would be the very negation of purposive construction: see Ribeiro PJ in Arrowtown at paras 37 and 39 and the perceptive judgment of the special commissioners (Theodore Wallace and Julian Ghosh) in Campbell v Inland Revenue Commissioners [2004] STC (SCD) 396.”

203.       He said, at [39], that the present case, like MacNiven, illustrates the need for a close analysis of what, on a purposive construction, the statute actually requires and then proceeded to apply the approach he had set out to the facts of that case.

204.       In Rangers, the taxpayer company (RFC) was a member of group of companies which set up a trust arrangement for the remuneration of employees.  When it wished to benefit an employee, it made a payment to a trust, asked the trustee to resettle the sum on to a sub-trust and requested that the sub-trust income and capital should be applied in accordance with the employee’s wishes. The trustee had a discretion whether to comply with those requests, but, in practice, the trustee without exception created the requested sub-trust.  The employee was appointed as protector of the sub-trust with the power to change its beneficiaries.

205.       HMRC assessed RFC to tax on the basis that under the PAYE system it should have accounted for income tax and NICs on amounts paid into the main trust on the basis they comprised payments of emoluments/earnings from an employment.  The Supreme Court unanimously decided in favour of HMRC.  Lord Hodge gave the judgment with which the other Lords agreed.

206.       Lord Hodge started with general comments on the correct approach to take to the construction of the relevant provisions.  He noted, at [10], that the legislative code for the taxation of income has developed over time to reflect changing governmental policies in relation to taxation, to remove loopholes in the tax regime and to respond to the behaviour of taxpayers.  He considered that as a result, “the legislative code is not a seamless garment but is in certain respects a patchwork of provisions”.   He said, at [11], that the courts at the highest level “have repeatedly warned of the need to focus on the words of the statute and not on judicial glosses, which may clarify or illustrate in a particular case but do not replace the statutory words”. He referred amongst other cases to Hochstrasser and Lord Radcliffe’s comments as set out above.

207.       He continued, at [12], that “another, more recent, judicial development in the interpretation of taxing statutes is the definitive move from a generally literalist interpretation to a more purposive approach”. He said that this can be traced to the speech which Lord Nicholls of Birkenhead in Barclays, in which he explained the true principle established in Ramsay and the cases which followed it:

“As he explained (para 28), the modern approach to statutory construction is to have regard to the purpose of a particular provision and interpret its language, so far as possible, in a way which best gives effect to that purpose. In the past, the courts had interpreted taxing statutes in a literalist and formalistic way when applying the legislation to a composite scheme by treating every transaction which had an individual legal identity as having its own tax consequences.  Lord Nicholls described this approach as “blinkered” (para 29). Instead, he removed the interpretation of taxing statutes from its literalist enclave and incorporated it into the modern approach to statutory interpretation which the court otherwise adopts.”  [He cited [32] as set out above]

208.       He continued to explain, at [13], that Lord Nicholls (at [34]) recognised two features which were characteristic of tax law.

“First, tax is generally imposed by reference to economic activities or transactions which exist, as Lord Wilberforce said (in W T Ramsay, 326) “in the real world”. In the Court of Appeal in Barclays Mercantile [2003] STC 66, para 66, Carnwath LJ made the same point: taxing statutes generally “draw their life-blood from real world transactions with real world economic effects”. Secondly, the prodigious intellectual effort in support of tax avoidance results in transactions being structured “in a form which will have the same or nearly the same economic effect as a taxable transaction but which it is hoped will fall outside the terms of the taxing statute”. He continued:

“It is characteristic of these composite transactions that they will include elements which have been inserted without any business or commercial purpose but are intended to have the effect of removing the transaction from the scope of the charge.”

The correct response of the courts was not to disregard elements of transactions which had no commercial value. That, he said, was going too far. Instead the court had, first, to decide, on a purposive construction, exactly what transaction would answer to the statutory description and secondly, to decide whether the transaction in question did so (para 36).”

209.       At [14] he said that Lord Reed in UBS AG v Revenue and Customs Comrs [2016] 1 WLR 1005 (“UBS”), at [62], has helpfully summarised the significance of the new approach, which Ramsay, as explained in Barclays, has brought about, in these terms:

“First, it extended to tax cases the purposive approach to statutory construction which was orthodox in other areas of the law. Secondly, and equally significantly, it established that the analysis of the facts depended on that purposive construction of the statute.”

210.       He summarised the position, at [15], noting that three aspects of statutory interpretation are important in determining the appeal.

“First, the tax code is not a seamless garment. As a result provisions imposing specific tax charges do not necessarily militate against the existence of a more general charge to tax which may have priority over and supersede or qualify the specific charge…….Secondly, it is necessary to pay close attention to the statutory wording and not be distracted by judicial glosses which have enabled the courts properly to apply the statutory words in other factual contexts.  Thirdly, the courts must now adopt a purposive approach to the interpretation of the taxing provisions and identify and analyse the relevant facts accordingly. 

211.       He concluded, at [16], that accordingly the proper approach was, first, to interpret the relevant statutory provisions purposively and, secondly, to analyse the facts in the light of those statutory provisions so construed.

212.       He said, at [36], that the central issue was “whether it is necessary that the employee himself or herself should receive, or at least be entitled to receive, the remuneration for his or her work in order for that reward to amount to taxable emoluments”.  In his view, at [37] and [38], a careful and detailed examination of the provisions of the primary legislation revealed no such requirement.  Moreover, at [39] and [40], he saw “nothing in the wider purpose of the legislation” which excluded from the charge or the PAYE regime, remuneration which the employee is entitled to have paid to a third party and thought that the relevant subordinate legislation points in the same direction .

213.       He concluded, at [41], that as a general rule, therefore, the charge to tax on employment income:

“extends to money that the employee is entitled to have paid as his or her remuneration whether it is paid to the employee or a third party. The legislation does not require that the employee receive the money; a third party, including a trustee, may receive it.”

214.       He said that whilst there are certain exceptions from this rule there is no exception as regards ss 62(2)(a) or (c).  

215.       He said, at [50], that the advice of the Privy Council in Hadlee v Comr of Inland Revenue [1993] AC 524 is in point. That case concerned legislation in New Zealand which provided that income tax was payable by every person on income derived by him during the year for which tax was payable.  A partner in an accountancy firm assigned a proportion of his share in the partnership to a trust under which the primary beneficiaries were his wife and child.  The New Zealand courts rejected his argument that he was not liable to income tax on that proportion of his annual partnership income. The Privy Council upheld their decision, holding that income tax was a tax on income which was the product of the taxpayer’s personal exertion and that the taxpayer could not escape liability to pay that tax by assigning a part of his share in the partnership.  Lord Hodge noted that:

“While the relevant provision of the New Zealand statute was worded differently from the United Kingdom legislation, the latter, by its emphasis on emoluments arising from a taxpayer’s employment, adopts a similar concept of the tax charge. It supports the view which I have reached that a charge to income tax on employment income can arise when an arrangement gives a third party part or all of the employee’s remuneration.”

216.       He continued, at [51], that it was also necessary to decide whether under the PAYE provisions there has been a “payment” of emoluments/earnings from which deductions were required.  In that context he considered that misplaced reliance, of the type he had warned against, had been placed on judicial glosses in earlier cases on the meaning of the term “payment” in this context.

217.       In his view, at [52], this stemmed from the decision in Garforth v Newsmith Stainless Ltd [1979] 1 WLR 409.  In that case, a taxpayer company voted to award bonuses to its two directors and controlling shareholders and credited the sums to accounts with the company from which the directors were free to draw.   The directors did not draw on those sums.  HMRC assessed the company to tax, arguing that the company should have deducted tax under the PAYE system on the full sums credited to those accounts. 

218.       Lord Hodge noted, at [52], that Walton J said that the word “payment” had no one settled meaning but took its colour from its context.  He held that there was no need for the directors to withdraw the money from their loan accounts for there to have been payment by the company, stating “when money is placed unreservedly at the disposal of directors by a company, that is equivalent to payment”.  Different considerations would have arisen if a further decision by the board of directors or by the shareholders in general meeting was required before the money could have been withdrawn. Lord Hodge said that the interpretation or gloss which Walton J placed on “payment” (as “money placed unreservedly at the disposal …”) “was a practical and sensible one in the context of the circumstances which he was addressing, which later became the subject of statutory provision..”.  However, at [54], there was a limit to this in that:  

“the gloss is no basis for establishing a general rule or “principle” that a payment is made for the purposes of PAYE only if the money is paid to or at least placed unreservedly at the disposal of the employee.  Yet it has been so used.”

219.       He gave some examples of the misuse of this “gloss” in the decision by the Inner House in Aberdeen Asset Management plc v Revenue and Customs Comrs 2014 SC 271 and of the Special Commissioners in Sempra Metals Ltd v Revenue and Customs Comrs [2008] STC (SCD) 1062 (as set out at [54] to [57]) which he considered were wrongly decided applying this judicial gloss to the meaning of the term “payment”. 

220.       He concluded at [58] and [59] that:

“In summary, (i) income tax on emoluments or earnings is due on money paid as a reward or remuneration for the exertions of the employee; (ii) focusing on the statutory wording, [none of the relevant provisions]…. (except section 62(2)(b)), provide that the employee himself or herself must receive the remuneration; (iii) in this context the references to making a relevant payment “to an employee” or “other payee” in the PAYE Regulations fall to be construed as payment either to the employee or to the person to whom the payment is made with the agreement or acquiescence of the employee or as arranged by the employee, for example by assignation or assignment; (iv) the specific statutory rule governing gratuities, profits and incidental benefits in section 62(2)(b) of ITEPA applies only to such benefits; (v) the cases, to which I have referred above, other than Hadlee, do not address the question of the taxability of remuneration paid to a third party; (vi) Hadlee supports the view which I have reached; and (vii) the special commissioners in Sempra Metals (and in Dextra) were presented with arguments that misapplied the gloss in Garforth and erred in adopting the gloss as a principle so as to exclude the payment of emoluments to a third party.

Parliament in enacting legislation for the taxation of emoluments or earnings from employment has sought to tax remuneration paid in money or money’s worth. No persuasive rationale has been advanced for excluding from the scope of this tax charge remuneration in the form of money which the employee agrees should be paid to a third party, or where he arranges or acquiesces in a transaction to that effect…..”

221.       Applying the legislation to the facts Lord Hodge held, at [64], that the relevant provisions for the taxation of emoluments/earnings were and are “drafted in deliberately wide terms to bring within the tax charge money paid as a reward for an employee’s work”.  The scheme was designed to give each footballer access without delay to the money paid into the trust, if he so wished, and to provide that the money, if then extant, would ultimately pass to the member or members of his family whom he nominated. He concluded therefore that “having regard to the purpose of the relevant provisions….the sums paid to the trustee of the main trust for a footballer constituted the footballer’s emoluments or earnings”.

222.       At [65], he said that the fact that there was a chance that the trust company as trustee of the main trust might not agree to set up a sub-trust and that as trustee of a sub-trust it might not give a loan of the funds of the sub-trust to the footballer, did not alter the nature of the payments to the main trust.  He based that conclusion on the approach taken in the Scottish Provident case: 

“In applying a purposive interpretation of a taxing provision in the context of a tax avoidance scheme it is legitimate to look to the composite effect of the scheme as it was intended to operate. In Inland Revenue Comrs v Scottish Provident Institution [2004] 1 WLR 3172 Lord Nicholls stated (para 23):

“The composite effect of such a scheme should be considered as it was intended to operate and without regard to the possibility that, contrary to the intention and expectations of the parties, it might not work as planned.”

The footballers, when accepting the offer of higher net remuneration through the trust scheme which the side letters envisaged, were prepared to take the risk that the scheme might not operate as planned. The fact that the risk existed does not alter the nature of the payment to the trustee of the Principal Trust.”

223.       Accordingly, at [67], payment to the trust should have been subject to deduction of income tax under the PAYE Regulations.   I note that also in an employment tax context Lord Reed expressed a similar view, that a composite scheme should be considered as it was intended to operate, in UBS (see [68] to [78]).

224.       The Scottish Provident case, on which these views were based, concerned a scheme designed to take advantage of a change in the law governing the taxation of gains and losses made by mutual life offices on the grant or disposal of options to buy or sell gilts. Under the scheme:

(1)          The life office, SPI, granted Citibank the option to buy a quantity of gilts from it at a “strike price” of 70, well below their anticipated market value at the time the option was exercised, in return for a premium. Under the law then in force, the premium was exempt from tax.

(2)          After the law had changed, Citibank exercised the option, requiring SPI to sell the gilts to it at a loss. Under the law then in force, the loss was allowable for tax purposes. In order to ensure that no real loss could be suffered by either party, the scheme also provided for Citibank to grant an option to SPI, entitling it to buy a matching quantity of gilts from the bank at a strike price of 90, calculated so that the overall movements of money between the parties were equivalent.

(3)          It was anticipated that both options would be exercised, but there was a possibility that they might not be. In the event, both options were exercised, and neither gilts nor money changed hands.

225.       Lord Nicholls set out, at [18], that whether SPI was entitled to treat the loss suffered on the exercise of the option granted to the bank as an income loss essentially depended on whether the option gave the bank an “entitlement” to gilts.  At [19] he  noted that if attention was confined to that option, it “certainly gave [the bank] an entitlement, by exercise of the option, to the delivery of gilts” but “if the option formed part of a larger scheme by which [the bank’s] right to the gilts was bound to be cancelled by SPI’s right to the same gilts, then it could be said that in a practical sense [the bank] had no entitlement to gilts”. He then referred to the purposive approach set out in caselaw:

“Since the decision of this House in WT Ramsay Ltd. v. Inland Revenue Commissioners [1982] A C 300 it has been accepted that the language of a taxing statute will often have to be given a wide practical meaning of this sort which allows (and indeed requires) the Court to have regard to the whole of a series of transactions which were intended to have a commercial unity. Indeed, it is conceded by SPI that the Court is not confined to looking at the Citibank option in isolation.  If the scheme amounted in practice to a single transaction, the Court should look at the scheme as a whole. Mr. Aaronson Q.C., who appeared for SPI, accepted before the Special Commissioners that if there was “no genuine commercial possibility” of the two options not being exercised together, then the scheme must fail.”

226.       Lord Nicolls continued, at [20] and [21], that:

(1)          The taxpayer’s counsel submitted that “even if the parties intended that both options should be exercised together…the Court could treat them as a single transaction only if there was “no practical likelihood” that this would not happen”. 

(2)          SPI had the benefit of the findings of fact by the Special Commissioners who adopted (at para 24) the analogy of horserace betting as follows:

“If the chance of the price movement occurring was similar to an outsider winning a horse race we consider that this, while it is small, is not so small that there is no reasonable or practical likelihood of its occurring; outsiders do sometimes win horse races.”

(3)          The test of “no practical likelihood” derived from the speech of Lord Oliver of Aylmerton in Craven v White [1989] A C 398, at p 514.  In that case, however, “important parts of what was claimed by the Revenue to be a single composite scheme did not exist at the relevant date” (see Lord Oliver (at p 498)). 

227.       Lord Nicholls continued at [22] to note that in Craven v White “thus there was an uncertainty about whether the alleged composite transaction would proceed to completion which arose, not from the terms of the alleged composite transaction itself, but from the fact that, at the relevant date, no composite transaction had yet been put together” whereas in the present case:

“…the uncertainty arises from the fact that the parties have carefully chosen to fix the strike price for the [option granted to SPI] at a level which gives rise to an outside chance that the option will not be exercised.  There was no commercial reason for choosing a strike price of 90.  From the point of view of the money passing (or rather, not passing), the scheme could just as well have fixed it at 80 and achieved the same tax saving by reducing the Citibank strike price to 60.  It would all have come out in the wash.  Thus the contingency upon which SPI rely for saying that there was no composite transaction was a part of that composite transaction; chosen not for any commercial reason but solely to enable SPI to claim that there was no composite transaction.  It is true that it created a real commercial risk, but the odds were favourable enough to make it a risk which the parties were willing to accept in the interests of the scheme.”

228.       At [23] Lord Nicholls held that:

“We think that it would destroy the value of the Ramsay principle of construing provisions such as [the relevant provisions in the Finance Act 1994] as referring to the effect of composite transactions if their composite effect had to be disregarded simply because the parties had deliberately included a commercially irrelevant contingency, creating an acceptable risk that the scheme might not work as planned. We would be back in the world of artificial tax schemes, now equipped with anti-Ramsay devices. The composite effect of such a scheme should be considered as it was intended to operate and without regard to the possibility that, contrary to the intention and expectations of the parties, it might not work as planned.”

229.       At [24] he concluded that it follows that the Special Commissioners erred in law in finding that there was a realistic possibility of the options not being exercised simultaneously meant, without more, that the scheme could not be regarded as a single composite transaction.  He said: “We think that it was and that, so viewed, it created no entitlement to gilts and that there was therefore no qualifying contract”.

Conclusion

Summary

230.       To recap, the question is whether (a) the Heronden Payments or Bet Profits are subject to income tax and NICs as “earnings from” the Individuals’ “employment” with the appellant and, (b) if so, whether there was “payment” of such earnings within the meaning of the PAYE rules so that the appellant was required to account for tax in respect of them. 

231.       As set out in Barclays and endorsed most recently in Rangers, it is long established, following the seminal decision in Ramsay, that in determining this question the tribunal must (a) give the employment tax provisions a purposive construction to determine the nature of the transaction to which they are intended to apply and (b) decide whether the actual transaction (which might involve considering the overall effect of a number of elements intended to operate together) answers to the statutory description.

232.        In carrying out that exercise, the correct response is not simply in all cases to disregard elements of transactions which have no commercial value. The tribunal must decide, on a purposive construction, exactly what transaction answers to the statutory description and whether the transaction in question did so.  As it was put succinctly in Arrowtown (as cited in Barclays), the principle established in the cases involves “a general rule of statutory construction and an unblinkered approach to the analysis of the facts”.  In other words, the “ultimate question is whether the relevant statutory provisions, construed purposively, were intended to apply to the transaction, viewed realistically”.

233.       As Lord Hodge recognised in Rangers, citing the decision in Scottish Provident, in applying this purposive approach to the interpretation of taxing provision in the context of a tax avoidance scheme it is legitimate to look to the composite effect of the scheme as it was intended to operate without regard to the possibility that, contrary to the intention and expectations of the parties, it might not work as planned.  In Scottish Provident Lord Nicholls said that it would destroy the value of the Ramsay principle if “the composite effect of transactions had to be disregarded simply because the parties had deliberately included a commercially irrelevant contingency, creating an acceptable risk that the scheme might not work as planned”.  He held that whilst it was true that the relevant contingency in that case created a real commercial risk, “the odds were favourable enough to make it a risk which the parties were willing to accept in the interests of the scheme”.

234.       As Lord Hodge said in Rangers, the purpose of the employment tax provisions is to impose income tax on money or money’s worth paid as a reward or remuneration for the exertions of the employee.  The provisions are “drafted in deliberately wide terms to bring within the tax charge money paid as a reward for an employee’s work”; there is no requirement that the employee must receive remuneration to which he is entitled for it to be taxable in his hands.  Lord Hodge held that the references to making a “payment” of employment earnings to an employee in the PAYE rules may encompass payment either to the employee or to the person to whom the payment is made with the agreement or acquiescence of or, as arranged by, the employee.

235.       Nothing in Rangers detracts from the fact that, as set out most recently by the Court of Appeal in Kuenhe, for the employment tax provisions to apply “there must, in actual fact, be a relevant connection or a link” between the relevant payments and the employee’s employment or “a sufficient causal link” between them; the payments must be from an employment source.   As it was put in the earlier cases, to be taxable as employment income money or money’s worth must be paid, “in return for acting as or being an employee”, or as “a reward or return for the employee’s services, whether past, current or future”, or “for being or becoming an employee” and not for some other reason (as set out in Hochstrasser, Tyrer v Smart and Shilton v Wilmhurst respectively). 

236.       With these principles in mind and, on the basis of the factual findings drawn from the evidence set out below, my view is that the Heronden Payments made by the appellant to Heronden, which in effect enabled it to pay the Bet Profits to the Individuals:

(1)           are “earnings from” the Individuals’ “employment” with the appellant paid as a reward for their services as employees/directors employed in its tax consultancy business;

(2)          to which the PAYE rules apply on the basis that they were paid to Heronden under arrangements put in place by the Individuals and with their acquiescence or agreement (on the basis that they would in effect pass to the Individuals as the corresponding Bet Profits). 

237.        In summary, in my view, it is appropriate to apply the widely drawn term “earnings from an employment” by reference to the overall effect of the arrangements under consideration in this appeal on the basis that the elements involved were plainly intended to operate together as a composite whole with commercial unity. Taking an unblinkered realistic view according to their overall effects, the transactions were put in place as nothing more than devices intended to extract from the appellant sums, which were intended to be a reward for the Individual’s employment services for it, without attracting any substantial employment tax charges: 

(1)          Viewed individually the Bets and CSOs had the commercial characteristics of a spread bet/option contract in that their outcome depended on the movement in a specified market index, namely, the growth in value of the Basket over a short period.  However, viewing the overall effect of their carefully constructed payment profiles under the steps, which from the outset the parties intended to undertake in relation to them, it is plain that neither the Individuals nor the appellant undertook them in order to speculate on movements in the market. 

(2)          In fact, the transactions and their commercial effects, as set up and implemented to operate as composite whole, were designed solely with the intent that:

(a)          if the selected Basket grew in value as the parties expected in reliance on its historic performance, the contracts would operate to extract cash, specifically set at the level in which the appellant wished to remunerate the Individuals, from the appellant to Heronden (as the Heronden Payments) and thereby into the hands of the Individuals (as the Bet Profits), on the face of it, largely in the form of tax-free “winnings” from a spread bet, or

(b)          if the Basket did not perform to deliver that result, the contracts would operate to transfer the funds “lost” by the Individual (in placing his Stake under the Bet (less the Initial Premium)) to the appellant as corresponding “winnings” under the CSO.  In that case, the funds spent in seeking to extract the employment earnings from the appellant in a tax free form, became available, within the employer, a vehicle owned and operated by the three Individuals, to be returned to them in one form or another (barring any wholly unexpected events).

238.       In all the circumstances, the fact that the contingent nature of the outcomes under the Bets and CSOs was at the heart of the scheme, so that there was a real risk that the planning would not work as intended, does not evidence that the Heronden Payments (or Bet Profits) were payments generated under speculative betting transactions rather than employment earnings:

(1)          The risk profile under the contracts, in the form of the chosen Basket, was not chosen by reference to risk appetite or any other commercial imperative which might normally drive the terms of spread betting transactions.  Rather it was carefully selected with a view to:

(a)          striking a balance between ensuring that the receipt of the Heronden Payments and matching Bet Profits was subject to (i) a sufficiently material degree of uncertainty as the parties considered necessary to reduce the prospect that the transactions would be viewed as a composite whole designed to deliver employment earnings but (ii) at the same time,  a sufficient degree of certainty to maximise the prospect that the planning would in fact succeed to deliver those earnings (see, for example, Mr Forsyth’s comments at [77]); and

(b)          ensuring that the value attributable to the novated CSOs at the date of novation was minimal thereby, in the parties’ view, triggering a correspondingly low taxable employment related benefit. 

(2)          There may well have been a greater likelihood of an unsuccessful outcome under the Bets and the CSOs than that the “commercially irrelevant” contingency would be triggered under the option arrangements in Scottish Provident.  However, it would be out of accord with the general principles set out above to regard that case as setting down a “one size fits all” test that only a contingency of that type may legitimately be disregarded in deciding on whether a particular statutory provision applies on a realistic view of the composite effect of a series of transactions intended to take place as a commercial unity. 

(3)          On the facts of this case, the contingent nature of the outcome under the contracts has to be viewed in the light of the fact that the scheme was designed so that, if the risk of an adverse outcome materialised, the funds “lost” by the Individuals were simply channelled into the appellant, a vehicle which together they controlled.  Viewed in that context, whilst the Bets and CSOs were constructed to create a real commercial risk that the planning would not work, the odds of success were favourable enough and the odds of failure were sufficiently tempered by the design of the scheme, to make it a risk which the parties were willing to accept in the interests of the scheme.  The risk of an adverse outcome, mitigated as it was by the channelling of the “lost” funds into the appellant, was simply the price which the parties were prepared to pay to maximise their chances of realising employment earnings in what was intended to be a virtually tax free way. 

(4)          In other words, taking a realistic unblinkered view of the facts, the fact that a contingent outcome carrying a level of risk was specifically built into a scheme with a view to achieving a particular tax outcome and that, if the risk of an adverse outcome materialised, it was substantially mitigated by design, does not evidence that these transactions are to be viewed for tax purposes as generating “winnings” or “losses” from speculative spread betting transaction rather than employment earnings.  For the same reasons it does not evidence that the sums were not “payments” for PAYE purposes.

239.       Overall, therefore, I have concluded that, on a purposive approach, the provisions are broad enough to capture the sums emanating from this structure in the form of the Heronden Payments as “earnings from an employment” given the appellant’s plain purpose in entering into the transactions (acting through and as facilitated and arranged by the Individuals) was to provide them as a reward or return for the Individuals’ services.  The fact that the sums were paid via Heronden in the form of “losses” (thereby funding a matching profit for the Individuals) under contracts designed as spread bets/option contracts with contingent outcomes does not detract from their character as employment earnings given that the only purpose of that contractual construct and each element involved in the pre-determined set of transactions was simply to avoid income tax and NICs charges.  I have addressed the appellant’s argument based on Abbott v Philbin below.

Conclusions on the facts

240.       In undertaking the transactions, the Individuals and the appellant implemented for their own benefit the planning or, as Mr Forsyth would put it, the “investment” or “trading” strategy, known as Alchemy.  In their capacity as directors and employees of the appellant, the Individuals were at least in part responsible for devising the planning, they advised many clients on it (in the standardised terms set out in the tax letter) and, prior to the transactions, they took valuation, tax and, in March 2012, accounting advice on it.  The Individuals had previously entered into similar arrangements in 2011 (although that involved a novation of the relevant CSOs to an EBT) and did so subsequently in 2013 to 2016.  Mr Forsyth accepted that by the time of the July transactions the Individuals were familiar with the process for implementing the Alchemy scheme.

241.       The accounting advice given to the appellant by Grant Thornton was based on the fact that the Alchemy structure was intended to provide employers with a strategy for remunerating their employees (see [40] and [41]).  It is reasonable to suppose from the scope of that letter that the appellant informed Grant Thornton that this was the case.  Accordingly, Grant Thornton advised that the employer should account for payments it made under the scheme as employees’ remuneration.  The appellant did so in respect of the Heronden Payments it made under the transactions and claimed a deduction for those amounts as employees’ earnings in computing its profits for corporation tax purposes.

242.       The scheme was operated essentially on the same basis in each of the iterations undertaken by the Individuals and, as Mr Forsyth confirmed, as regards the appellant’s clients also, using what can be described as a standardised process and standardised documentation:

(1)          Each of the transactions was made on precisely the same basis except that Mr Hughes “traded” £20,000 more than the other Individuals in the July and October 2012 iterations.  The Heronden Payments made under the transactions together reduced the appellant’s accounting profits for the year ended 31 December 2012 to a few thousand pounds only.

(2)          The parties followed the same procedure and used the same documentation to implement the transactions (changed as regards the details of each “trade” only).  For example, the parties used the same form of letter or document as regards (i) the letter the Individuals sent to Aston Collie to initiate the transaction, (ii) the Aston Collie letter (iii) the letter Aston Collie sent to the appellant in response to its request for advice on the financial effects of the novation of the CSOs, (iv) the documents implementing the Bet, the CSO and the novation, and (v) the board minutes of the appellant’s board of directors approving the novation of the CSOs.

(3)          In each iteration of the transactions, the Individuals each novated their CSOs to the appellant on the same day.  Mr Forsyth confirmed that in all cases where the Alchemy structure was used that he was aware of, barring some error, the relevant CSO was novated to the employer.

243.       In my view it is reasonable to conclude from the evidence summarised above that (i) the Alchemy structure was intended to enable employers to remunerate employees for their employment services on a virtually tax-free basis and (ii) by the time the July transactions were undertaken, the Individuals and the appellant were fully aware of how the scheme was intended to operate. That this was specifically the intention and purpose of the Individuals and the appellant in implementing the transactions is evidenced by the design and financial effects of the scheme and the factors referred to below.

Initiation of the transactions and Aston Collie advice

244.       The Individuals initiated each transaction formally by writing to Aston Collie asking for advice on implementing a transaction in the specified parameters and for a recommendation of a financial counterparty (the “initiation letter”).  However, as in relation to much of the documentation recording the steps involved in the transactions, it appears that the initiation letters were formed and framed in this way to create or enhance the impression that the Individuals were entering into the relevant contracts as a form of commercial “investment” relating to spread betting and that in doing so were acting independently of the appellant. For all the reasons set out, I do not consider that is reflected in reality.

245.       Given that Heronden was closely involved in the design of the Alchemy scheme and, that it and Aston Collie were connected through Mr Swallow, it would have been highly surprising if Heronden was not recommended by Aston Collie as the counterparty (as Mr Forsyth accepted (see [137]).  Mr Forsyth said that there was no discussion with Heronden about the identity of the party to which the CSOs would be novated although he accepted that they were aware how things were likely to play out (see [135] and [136]).  However, in light of their involvement in the development of the scheme and, given that they and Aston Collie played precisely the same role repeatedly in transactions where in all cases (barring accident) an onerous CSO was novated to the employer, I consider it improbable that they were not fully aware that the intention was to extract employment earnings from the appellant in a largely tax free form.  

246.       In any event, from the content of their speedy responses to the initial initiation letter, there can be no doubt that both Heronden and Aston Collie fully understood that their role was to source an appropriate Basket and to structure the precise payments under “financial contracts” with a view to maximising the chances of those contracts delivering, through the medium of Heronden as counterparty, a specified amount of cash into the hands of the Individuals from another party. 

247.       In each initiation letter, the Individuals essentially set out the key parameters and essential elements of the relevant transaction (see [48]).  Mr Forsyth accepted that when he initiated the transactions he was already aware of the shape of them and was looking to Aston Collie to provide the details and to liaise with Heronden who he relied on to source a Basket with an appropriate risk profile (see [138]).  In summary, in the initiation letters:

(1)          The Individuals specified that they wanted to enter into a Bet and a CSO in effect as a “matched” pair of contracts and set the parameters of the overall maximum and minimum net amounts they stood to “win” or “lose” in specifying the maximum loss and maximum winnings (as precisely matching sums). 

(2)          The Individuals stated that they wanted the reference index applicable to both contracts to be chosen on the basis that they had a relatively low probability of success under the contracts (of around 20%) according to how the market judges such matters but a much higher chance of success according to the index’ historical performance.

(3)          They made it plain that they were considering divesting themselves of one of the contracts as they said that they wanted one of the contracts to be as attractive as possible to a third-party so that the bulk of any premium payable should be structured as a final premium.

248.       Aston Collie responded in each case with a standard form letter in identical terms except that the details/figures of each “trade” changed as necessary.  I note the following from the Aston Collie letter sent in relation to the BF July transaction (see [50] to [61]):

(1)          The comments on the scope of the firm’s role indicate that Mr Swallow considered that Aston Collie and Heronden were being asked to structure “financial derivative” contracts with a view to achieving a particular cash flow and tax result on the basis that the CSO would be novated to a third party.  Mr Swallow (a) referred at the start of the letter to the tax advice on the tax treatment of the proposed “investments” and subsequent novation of one of them, (b) he later noted that Aston Collie would limit the scope of its advice “solely to the suitability” of the CSO and Bet “to achieve the outcome intended by your tax advisers (but not whether the tax outcome will be successful)”, and (c) he concluded from the figures he set out that the contracts would meet Mr Forsyth’s “stated financial objectives” and would (subject to the necessary performance of the Basket), “provide a mechanism which is likely to deliver the appropriate cash flows, but only if a Novation Agreement is subsequently executed and not otherwise”.

(2)          From Aston Collie’s explanation of the possible outcomes under the contracts and the illustrative figures, it is plain that, in structuring the transaction within the parameters Mr Forsyth had specified in the initiation letter, the financial outcomes under the proposed Bet and CSO were “matched” so that:

(a)          If Mr Forsyth were to retain both the Bet and the CSO to term, whether he “lost” or “won” under the Bet or the CSO he would necessarily suffer a loss to the tune of Heronden’s fee.  As Aston Collie put it, in that case it would be as if nothing had happened apart from the fact that Heronden’s fees were due.  Aston Collie noted that, therefore, Mr Forsyth had been advised that there was a benefit to him in novating the CSO to a third party. I would put it rather that there was in fact a detriment to him unless that were the case.

(b)          If the CSO was novated to another party, leaving aside Heronden’s fee, whatever the outcome, the payments under both contracts were designed so that the other party would be required to pay to Heronden a net amount exactly equal to the net amount Heronden would have to pay to Mr Forsyth.  Moreover, the payments were structured so that:

(i) the maximum amount Mr Forsyth could “lose” under the contracts was confined to a much smaller amount than the Bet Profit he could “win”.  As regards the BF July transaction, for example, he stood to lose £85,216 (the Stake less the Initial Premium) and stood to gain £385,686 (the maximum winnings under the Bet plus the Initial Premium); and

(ii) correspondingly the maximum gain the other party could realise  under the CSO was confined to a much smaller amount than its potential maximum loss.  As regards the BF July transactions, the other party could gain a maximum of £74,667 (in the form of the Final Premium) but stood to lose a maximum of £393,333 (the maximum loss under the CSO less the Final Premium).

(3)          Whilst there were a range of possible outcomes under the Bet and the CSO, it is clear from discussion in the letter on the chances of success under the contracts and the focus of the illustrative figures that the Basket was chosen with a view to maximising the likelihood that the “investor” would “win” the maximum winnings under the Bet and “lose” the maximum loss under the CSO.  As Aston Collie put it, Mr Forsyth was relying on “beating the market” by reference to the Basket’s historical performance which, as Mr Forsyth had requested, was chosen specifically according to the much higher prospects of success on that measure than according to how such matters are typically judged in the market.  The CSO was specifically designed, therefore, as a contract under which the “investor” was expected to lose the maximum loss according to the statistical measure the parties relied on, namely, historic performance.

(4)          It is evident from the explanation of the outcomes under the contracts that Heronden’s role was confined merely to facilitating the transaction and acting as a conduit through which the funds would flow from the party which took on the CSO to the Individual (or vice versa if the planning was unsuccessful) from the facts that:

(a)          It was to receive a fixed fee (plus where it “won” under the Bet an amount to cover it for the betting duty cost) whatever the outcome of the CSO and Bet and whether or not the CSO was novated to another party.

(b)          It took no credit risk in that it was envisaged in the letter that (i) on the novation of the CSO, Heronden would require a Margin from the relevant party equal to the Heronden Payment which it would be required to pay to the Individual if the planning was successful (plus its fee) and (ii) it would require a Stake from the Individual under the Bet equal to the maximum amount Heronden would be required to pay to the other party if the planning was unsuccessful and the Individual “lost” the maximum loss under the Bet (plus its fee). I note that in later iterations of the transactions, the re-strike mechanism was introduced but the overall economic effect was the same under that revised structure as under the July transactions as Mr Forsyth accepted (see [134]).

249.       It is plain from the very design of the contracts, therefore, that, viewed as a pair, these were not spread betting and related option transactions in any normal commercial sense.  As might be expected under a spread bet the outcome of the Bets and the CSOs depended on an external factor which the parties could not control, namely, the growth in value of the Basket.  However, the Individuals can hardly be viewed as entering into either the Bet or the CSO as transactions in their own right, with a view to speculating on the outcome of the chosen reference index, given that they entered into them as a pair and, if they were to hold them both to term, whatever the outcome, they had no chance of realising a profit under the “matched” position but were certain to make a loss. 

250.       There can be no doubt that when they entered into the July transactions the Individuals were fully aware that this was the effect of entering into the contracts as a pair given their involvement in devising and advising on the structure, that they had previously implemented the structure and that they themselves in effect set the nature and level of the “matched” “losses” and “winnings” in the initiation letters.  In any event, Aston Collie made it entirely clear that there was no point in them entering into the Bets and the CSOs from a financial perspective and, indeed, there was a detriment to them (albeit a modest one) unless they were able to novate one of the contracts to another party. 

251.       It is also readily apparent that the Individuals would want to divest themselves of the CSOs rather than the Bets, given that, as already set out, it is plain from the Aston Collie letter that the CSOs were designed as onerous contracts under which the Individuals stood to “lose” much more than they could gain and that they expected they would “lose” the maximum losses under the CSOs (and realise the corresponding maximum winnings under the Bets).  I also note Mr Swallow’s comment that the desired cashflows would be achieved only if the CSOs were novated, that the Individuals specified in the initiation letter that the CSOs should be attractive to a third party and the comments in the standard form letter which Aston Collie in each case sent to the appellant describing the onerous effect of the CSOs.

252.       At the hearing Mr Forsyth accepted ultimately that (a) an individual could not be said to have a “trading strategy” in relation to the Alchemy transactions apart from through divesting himself of either the Bet or the CSO and (b) he entered into the relevant Alchemy transactions on the basis that it was more likely than not that he would “win” under the Bet and that was why he wanted to divest himself of the CSO.  He hoped that the fund managers could “beat market expectations” and he believed that “the market was mispricing…the option price” and that was why he retained the Bet (see [74] to [81] and [115] to [117]).  (I do not accept that the Individuals had a “trading strategy” as that term may be understood in any commercial sense for the reasons set out below.)  Mr Forsyth remarked in his witness statement and on a number of occasions at the hearing (see, for example, [81] and [116]) that it is self-evident or “axiomatic” that an investor should not rely on the historical performance of an index such as the Basket as a guide to its future performance.  However, axiomatic or not, in the light of his other evidence and the design of the structure as explained in the Aston Collie letter, it is plain that the Individuals and the appellant were expecting that they would realise the desired cashflows under the Bets and CSOs (in the form of the Heronden Payments and the Bet Profits) relying on the Basket’s track record.

253.       At the hearing Mr Forsyth to some extent continued to refer to Alchemy as devised as an “investment” strategy as he had described it in his witness statement (see [77]).  However, his comments, in referring again to the “ability to beat the market” and the levels of the chance of success, suggest nothing more than that he considered that Alchemy had to deliver transactions carrying the right chance of success in terms of (a) a relatively low chance of the realisation of the Bet Profits and Heronden Payments according to market perception (but not too low) and a high chance of that result according to historical performance (but not so high as to be virtually guaranteed).  Viewed in the context of the design and operation of the scheme and his other comments (see, in particular, [30]), it is plain that that feature was viewed as integral (a) in order to enhance the prospect of the scheme not being viewed as a composite scheme to deliver employment earnings (dependent as it is on a real contingency), but (b) to provide a high chance that the scheme would in fact succeed, and (c) to ensure that a low value was ascribed to the CSO at the novation date (thereby minimising the taxable benefit it was accepted then arose).  

254.       Overall, I conclude from the above evidence, as viewed in the light of the background to the implementation of these transactions, that the Individuals arranged for the Bets and CSOs to be set up and entered into them, as part of a broader plan, with the intention and expectation from the outset of being able to divest themselves of the relevant CSOs for no or only minimal cost.  In all the circumstances, it is not credible that they would put in place and take on such a pair of contracts unless that were the case or that they would want to divest themselves of the Bets rather than the CSOs.   As set out in detail below, I also consider that in fact there was never any real possibility that the CSOs would be novated to anyone other than the appellant.

Role of the appellant

255.       The Aston Collie letter, like the tax letter and other documents, was drafted on the basis that it was possible that the CSO could be novated to someone other than the appellant or relevant employer.   For example, it was noted in that letter that there was no certainty that the Individual would be able to novate the CSO and that, if the appellant took on the CSO, an employment related benefit may arise as regards which Heronden would provide a valuation. 

256.       However, in my view, for all the reasons set out below, there was no realistic possibility that the relevant CSOs would be novated to anyone other than the appellant (and, for the reasons set out above, I regard it as improbable that Aston Collie and Heronden were not aware that was the case (see [245])).  In fact, it is apparent that from the outset when setting up and entering into each transaction:

(1)          The Individuals intended to novate each of the relevant CSOs specifically to the appellant, as their employer, with a view to it remunerating them for their employment services through the Heronden Payments they expected to become due under the CSOs (which in effect funded the Bet Profits). 

(2)          As the Individuals were the sole directors and shareholders of the appellant and, in effect, its controlling minds, the appellant was fully aware of that intention and expectation. 

257.       As regards the parties’ intention that the CSOs would be novated to the appellant, I note the following:

(1)          For the reasons set out below, I have concluded that the level of the “trades” under the CSOs and the Bets (in the form of the maximum loss under the CSOs) was set by reference to the funds available in the appellant and the amounts in which it wished to remunerate the Individuals for their employment services (see [258] to [263]).  That accords with the terms of the Grant Thornton advice and the fact that the appellant reflected the Heronden Payments as directors’ remuneration in its accounts. 

(2)          In any event, it is very difficult to see who would be willing to take on such contracts on being required to provide a Margin equal to the Heronden Payment, other than a party who wished to confer a benefit on the Individuals, such as their employer.  For the reasons set out above, it is plain from the Aston Collie letters that each CSO was designed as an onerous contract under which the “investor” was expected to lose the maximum loss (and stood to gain only a much lower sum).  The onerous nature of the CSO was highlighted by Aston Collie also in the letters sent to the appellant when that firm was asked to advise the appellant on the financial consequences of taking on the novated CSOs.  Having set out the range of outcomes for the appellant, Mr Swallow concluded that “if this were a stand-alone transaction with no corresponding benefit arising to the employee, it would not be a suitable transaction for the company to enter into” (see [48(10)]). 

(3)          Mr Forsyth suggested in his witness statement that a novation to the appellant was not inevitable and that another party might have taken on the CSO (see [68(10)]).  However, under cross-examination he accepted that in practice it was not contemplated that the CSOs would be novated to anyone other than his employer or a related entity such as an employee benefit trust.  He initially said that it was always intended that the Individuals would approach their employer as regards the novation of the CSOs but only in the first place.  He then accepted, however, in effect that a sale in the market was unfeasible given the likely cost.  The best he could suggest was that a rich uncle might have helped him out (see [87]).

(4)          Moreover, any contention that it was seriously contemplated that the CSOs would be novated to a party other than the appellant lacks credibility given that:

(a)          As Mr Forsyth confirmed, in all the Alchemy transactions of which he was aware, barring accident, the CSO was in fact novated to the appellant or relevant employer company (see [91]). 

(b)          As part of the overall standard process used, the parties followed precisely the same procedure to give effect to the novation using the same standard documents (for example, as regards the request for the novation made by the Individual, the valuation from Heronden and the letter of advice from Aston Collie to the appellant), that process took a very short time and novation of all of the CSOs took place on the same day under each iteration of the transactions.

258.       In my view, in order to give effect to their intention that the CSOs would be novated to the appellant, the Individuals co-ordinated between themselves in initiating and implementing the transactions, in particular, in setting the levels of the maximum losses under the CSOs as specified in the initiation letters which then set the level of the Heronden Payments it was expected that the appellant would have to pay (and the corresponding Bet Profits).  They set these amounts at an appropriate level which they knew that the appellant could and would be willing to pay to them individually and collectively as remuneration for their services as directors/employees so that (acting through them as its directors and shareholders) it would agree to take on the novated CSOs.   

259.       That the Individuals must have co-ordinated on the level of the “trades” is apparent simply from the facts that (a) under each transaction the Individuals “traded” at precisely the same level of specified maximum losses except as regards Mr Hughes’ first two transactions when he “traded” precisely £20,000 more than the other two Individuals and (b) the Heronden Payments which the appellant made under the relevant transactions in aggregate equate to virtually all of the available accounting profits in the appellant in the relevant year (see [67]).  The only reason Mr Forsyth put forward for Mr Hughes “trading” in larger sums in July and October 2012 was that it may have been in the back of Mr Hughes’ mind that HMRC could later raise a point on the Individuals receiving exactly equal amounts in seeking to attack the scheme.  

260.       It would be a remarkable coincidence if, without any collaboration and co-ordination, the Individuals happened to pick the same sums (barring the exact £20,000 differences) as the maximum losses under the relevant CSOs which together amounted to nearly the whole of the appellant’s available accounting profits in the year ended 31 December 2012.  It can also hardly be coincidental that the Individuals entered into each iteration of the transactions at more or less the same time or shortly after each other using precisely the same process, that in each iteration of the transactions the novation of all of the relevant CSOs took place on the same day and that the terms of the transactions were exactly the same (with the exception only of the £20,000 difference).  

261.       Whilst Mr Forsyth did not accept the above proposition in terms he did accept, in effect, that there was at least an understanding or “tacit recognition” between the Individuals as regards the shape of the transactions and the level of “trades” as summarised below (and see, in particular, his evidence at [103] to [110], [123] and [124]):

(1)          Mr Forsyth initially said only that he was “aware” that Mr Hughes and Mr Baker intended to undertake the transactions but he did not know the details of Mr Hughes’ trades until he approached the appellant regarding the novation.  He later said he had a “greater knowledge” of Ms Baker’s transactions than of Mr Hughes’s transaction and that he was “almost certainly aware” that she entered into “trades” in exactly the same amount as him prior to her approaching the appellant as regards a novation. 

(2)          In his later comments he accepted that he essentially followed what Ms Baker did in terms of setting the level of the maximum loss under the CSO.  He said that he “mirrored” what Ms Baker did in this respect certainly for the first two transactions in July and October 2012.  He did not just blindly follow and there was “some kind of a sense check” in that he wanted to know personally what it would cost him to undertake the transaction and that she thought that the appellant could afford to agree to that type of novation but otherwise he “was happy to mirror what she did”.  He seemed to accept that Mr Hughes mirrored what he and Ms Baker did (although he traded a bit more in the first two transactions); Mr Hughes entered into the transactions after they did and knew the amounts which they had traded.  

(3)          He said that Mr Hughes and Ms Baker had a greater degree of discussions about setting the level of the transactions than he did.  He noted that the Individuals were not always in the office together and suggested that was why they sat down to discuss it properly at the board meeting.  He accepted, however, the following:

(a)           In entering into the transactions “one would need to be mindful of” what the company would be asked to do and the level of “trades” it would agree to.

(b)          “Of course, there was an understanding that if we all did something sensible….it wasn’t an agreement, it was a kind of tacit recognition that if we all did something similar, then…the company might be minded to agree to it”.  

(c)          Whilst he asserted that there was no prior agreement as such and the Individuals “each individually undertook a transaction of our own volition”, they “had outline discussions around what that transaction would look like, yes”. 

(d)         Whilst he did not consider any such “tacit recognition” or outline discussions took place in the Individuals’ capacity as directors, “of course” the Individuals used their “knowledge about what we thought the company might be able to agree to”.  He then went on to stress that they considered matters separately as directors and shareholders of the appellant.   

(e)          There was “co-ordination” as regards the picking of the amounts, at least as regards Mr Hughes.  

(f)           In December 2012 there was a “better understanding” of the appellant’s profitability at that point as that was the time when the directors discussed paying employees’ bonuses and profitability and that meant that he had a greater understanding of the amounts available in respect of the December transactions.

262.       Whilst Mr Forsyth only accepted that there were “outline discussions” as regards the transactions, I would not think it surprising if the Individuals did not have lengthy detailed discussions between themselves.  By the time they entered into the first set of July transactions the Individuals were all entirely familiar with the planning and the standardised process and documentation for implementing it.  In view of that, I cannot see that there would be a need for discussion beyond the Individuals co-ordinating on the timing of and appropriate levels of the transactions (as they plainly did).

263.        In any event, viewing Mr Forsyth’s evidence in the context of all the factors highlighted above, I consider it plain that there was, at the very least, an understanding between the Individuals that (a) they would each put in place Bets and CSOs with maximum losses under the CSOs (which drove the nature and level of the other payments under the contracts) set at the level at which the appellant wanted to remunerate them for their employment services, (b) on the basis that, barring any wholly unforeseen circumstances, they would cooperate, as the appellant’s shareholders and directors, in taking steps to ensure that the remuneration was delivered to each of them by the appellant taking on the novated CSOs (albeit that ultimately success depended on the growth in value of the Basket).  Acting collaboratively, they each set their transactions up so that there was no reason for the appellant to refuse to accept the novated CSOs;  they ensured that they were made on terms which it was known at the outset would be acceptable to the appellant on the basis that it wished to remunerate the Individuals in the specified amount.  It is evident from these conclusions that I do not accept Mr Forsyth’s repeated assertions that in setting up the transactions the Individuals were acting entirely independently from the appellant. However, I have considered Mr Forsyth’s evidence in that respect and the appellant’s related submissions at [266] to [276].  As part of that discussion I have considered the process whereby the appellant approved the novated CSOs at [273] to [276].

264.       In my view, it is plain that the appellant accepted the novation of the CSOs in the expectation that it would lose the maximum losses under the CSOs, in particular, given that its directors/shareholders arranged for the CSOs to be set up on the basis that was the likely outcome (according to historical performance of the Basket) as reflected in the comments on the onerous nature of the contract made by Aston Collie (see [48(10)]).  There is no plausible explanation for the appellant’s willingness to enter into such onerous contracts with that expectation other than that it wanted to provide the Individuals with employment earnings to that extent.  Moreover, it is readily apparent from the economics of the transactions, as set out in the Aston Collie letters of advice, that the consequence of the appellant “losing” the maximum loss under the relevant CSO, as expected, would be to deliver, via Heronden, cash in that amount (less the Final Premium) to the Individuals in the form of the matching Bet Profit (as pre-funded by the Margin).  In any event, that the scheme was intended to provide a mechanism for employers to remunerate their employees is readily apparent from the Grant Thornton letter, the accounting treatment adopted and the fact that the level of the maximum losses was set by reference to the appellant’s available resources.  

265.       I do not accept Mr Forsyth’s view that the content of the Grant Thornton letter does not indicate that the parties intended from the outset that the CSOs would be novated to the appellant, on the basis that the advice was provided to the appellant before the transactions were undertaken in a generic form as to how an employer should account for payments made under a CSO if it decided to accept a novation of a CSO (see [95] to [98]).  It is plain from the content of the letter that Grant Thornton must have been told by the appellant/the Individuals acting on its behalf that the commercial aim behind the Alchemy structure was to remunerate an employer’s employees.  That stated aim surely applies to the appellant and the Individuals, as much as to any of the appellant’s clients, should the appellant and the Individuals decide to implement the planning for themselves (as indeed they did). 

266.       Mr Forsyth maintained (see [103], [112], [125] and [126]) and as was the focus of much of the appellant’s case, that the Heronden Payments and Bet Profits were somehow disassociated from the Individuals’ employment.  He said that the remuneration/employment benefit provided by the appellant on taking on the novated CSOs was confined to a small benefit as valued by Heronden.  In his view, in taking on the CSOs the appellant simply provided him with the opportunity to profit from his contractual rights under the Bets on the basis that he had entered into those contracts and the CSOs initially entirely independently of the appellant.  He asserted that the appellant did not intend to make the Heronden Payments as employment earnings but simply took into account that it would not be appropriate to award the Individuals a large bonus in light of its potential liability to make those payments under the novated CSOs.  That view was also reflected in the board minutes of the board meeting at which the novation of the CSOs was approved (see [48(12)]).  

267.       To some extent this involves a legal argument that the link with the Individuals’ employment was necessarily broken when the CSOs were novated to the appellant with the result that the subsequent payments under the CSOs and Bets flowed from the parties’ contractual rights and obligations under them.  I have addressed that below.  However, to the extent this involves the assertion that, as a factual matter, the appellant’s purpose in entering into the novated CSOs was not to provide the Heronden Payments as employment earnings, it is not accepted.  Any such assertion is wholly unrealistic in light of the above findings, in particular, that (a) the level of the “trades” under the CSOs were set by reference to the available funds in the appellant and the amounts in which it wished to remunerate the Individuals, (b) the lack of any plausible explanation as to why the appellant would take on such onerous contracts otherwise than with a view to remunerating the Individuals, (c) the fact that for the Individuals to profit from the transactions necessarily required that the appellant suffered a matching net loss to fund that profit, and (d) the clear evidence that the scheme was intended to be a remuneration strategy in the Grant Thornton letter and from the accounting treatment adopted.

Mr Forsyth’s evidence on “independence”

268.       In my view,  there is no viable foundation for Mr Forsyth’s repeated contentions throughout his evidence that (a) in initiating the transactions the Individuals were acting of their own volition or entirely on their own account or independently, (b) any outline discussions the Individuals had on the levels of the “trades” and the shape of the transactions were not made in their capacity as directors of the appellant, and (c) the appellant had no knowledge of each transaction until it was formally approached by the relevant Individual (see [103] for example).  As noted, similar themes run through the tax letter provided by the appellant to its clients and some of the other documents.

269.       The appellant is, of course, as a corporate entity, a separate legal person distinct from the Individuals but it can only operate through them, as its directors, who operate its business for it.  As HMRC submitted, the Individuals were, as its directors, in effect the controlling minds of the appellant.  With that in mind and, in all the circumstances of this case, I consider that Mr Forsyth’s view of matters is not reflected in reality.

270.       As noted above, by the time of the transactions the Individuals plainly fully understood the Alchemy planning (and indeed it was at least in part devised by them) including the intended role of the employer company.  That knowledge came to them whilst acting as directors/employees of the appellant in carrying out its tax consultancy business.   I cannot see any argument that all of that knowledge is not to be attributed to the appellant as an entity.  The appellant was aware certainly in general terms, therefore, of the planning and an employer’s role in it.

271.       The appellant’s stance rests on the view that, in undertaking the planning on their own behalf and in initiating each relevant Bet and CSO, the Individuals were acting wholly independently from the appellant.  Mr Forsyth asserted that, on that basis, at least as regards the July transactions, the appellant is not to be attributed with any knowledge of them until the relevant Individual formally contacted its board of directors as regards a novation of the relevant CSO, as recorded in the standard documents evidencing the steps taken, and the board then met to consider it.  He did appear to accept that, given its involvement in the July transactions, the appellant would have had some inkling of what was planned as regards the later transactions although he asserted that there was no prior agreement with the appellant that the novation of the CSOs would take place (see [96]).

272.       However, it is difficult to see that the Individuals were not acting, at least in part, in their capacity as directors on behalf of the appellant from the very outset in taking steps to put in place a scheme that, on the findings set out above, was, by its very design, plainly intended to operate as a mechanism for extracting cash from the appellant into the hands of the Individuals as a reward for their employment services without attracting any significant tax charges.  In my view in putting in place the “matching” Bets and CSOs, as the initial required steps, with a view to novating the CSOs to the appellant, the Individuals can only be taken to have acted both (a) on their own personal behalf, as the directors/employees who hoped to receive employment earnings under the overall operation of the scheme, and (b) on behalf of the appellant in their capacity as its directors, as the employer who intended to provide those earnings.  

273.       Even if the Individuals could be said to be acting on their own account and independently of the appellant in setting up and entering into the Bets and CSOs initially (which I do not accept), it does not follow that the appellant was not fully aware of its intended role in the transactions until the holding of the board meeting at which it approved the novation of the relevant CSOs.  I cannot see that the Individuals’ intention to call on the appellant to accept a novation of the CSOs can be viewed as confined somehow to being known to them only in their own personal capacity until the point at which it was necessary for them formally to ask the appellant to accept the novation. 

274.       Of course, persons can and do act in different capacities as regards a company and may legitimately take decisions on its behalf, “wearing different hats” as directors and shareholders.  However, I cannot see that the status of a company as an entity with its own legal personality justifies the claim that directors of a company who implement planning which, as they know from the outset, involves them taking steps (i) not only in their own capacity, but (ii) also as directors of the company, become aware of the plan in their capacity as directors only once they formally approach the company in their other individual capacity.  It is unrealistic and artificial to suggest that, in such circumstances, such a person acquires the knowledge of the overall plan initially solely in his individual, non-director’s capacity and that he can somehow put that knowledge to one side or in a separate compartment until he chooses to reveal it to himself in his capacity as a director. 

275.       I accept that it does not necessarily follow from the fact that the appellant knew from the outset that the plan was for the CSOs to be novated to it that the appellant agreed to the novation of the CSOs in advance of its directors formally considering whether it should do so at the relevant board meeting.  However, as set out above, it seems to me that, barring a wholly unexpected event, there was no real likelihood that the appellant would refuse to take on the CSOs given, in particular, that (a) the Individuals set the levels of the maximum loss under them (which in turn set the level of the Heronden Payments) by reference to the amounts which they knew that the appellant would want to and would be able to pay them as earnings and (b) on Mr Forsyth’s own evidence, there was an understanding between the Individuals that the appellant would be minded to accept the novation of the CSOs assuming they all did something sensible. 

276.       In that context, Mr Forsyth’s assertion that the directors could not just agree to “any old novation” and each one had to be considered on its merits by the board at the relevant meeting as to whether it was “the right thing to do” (see [94]) simply has no teeth.  In effect, the CSOs were set up and arranged by the Individuals to ensure that they, acting as directors and shareholders of the appellant, had no reason to refuse to take them on.  The CSOs and related Bets were carefully crafted with a view to maximising the chances that the contracts would operate to extract cash from the appellant in amounts which were set by the Individuals by reference to the resources which the appellant had available and wanted to remunerate them with. 

277.       When the board met to consider the novation of the CSOs, given nothing unexpected had occurred, there was simply nothing of substance for the board of the appellant to consider.  The level of the “trades” (and consequent Margin payment) was already set by reference to what the appellant wanted to achieve.  The advice received from Aston Collie can hardly have been a surprise given that the appellant was well aware of how the scheme was intended to operate.  In those circumstances, the approval of the novation of the CSOs was merely a formality as the standard wording of the minutes also suggests. 

278.       This conclusion is not affected by:

(1)           The fact that in the paper trail the parties created it is recorded that each Individual approached the board separately as regards the novation of their CSOs, that one of the other directors responded and asked Aston Collie for advice on the financial consequences of the novation and that the board considered the request on the basis that the Individuals had entered into the contracts of their “own volition” and approved the novation having considered that advice.  In my view, this demonstrates only that the parties considered it important to the success of the scheme to create a documentary record as part of the standardised process adopted which was intended to show that each step in the structure was implemented and considered “independently” by the relevant party.  Given the findings I have made, the presentation of the scheme in this way is merely window dressing which is not reflected in reality. 

(2)          Mr Bremner’s submission that whether the novation took place depended on whether Heronden agreed to it.  It is plain that Heronden played a standard role in facilitating these transactions.  Its only requirement as regards the novation was that it received its Margin requirements from the appellant.  For all the reasons set out, the Margin was plainly going to be forthcoming (barring wholly unforeseen events) because from the outset the level of the Heronden Payments (and thereby the Margin) was set by reference to the appellant’s available funds and the amounts in which it wished to remunerate the Individuals. 

Other points on commerciality

279.       Mr Forsyth made a number of other assertions which he presented as supporting the view that the Individuals’ purpose in entering into these transactions was to speculate under spread bets, albeit that this “trading strategy” was dependent on the ability to divest themselves of the CSOs.  In his witness statement, for example, he said that the level of the Bets was set by reference to the Individuals’ “appetite for risk” and their own requirements.  As set out, at the hearing he appeared to accept (and I have found on all the evidence) that the level of the “trades”, which in effect determined the Heronden Payments under the CSOs (and the corresponding Bet Profits), was set by reference to the remuneration which the appellant wished to pay to the Individuals.  In this context, however, I also note the following:

(1)          I accept that the Individuals were concerned with the probability of the Bets’ success and, in particular, in identifying a Basket as the reference index which gave the Bets a low probability of success according to the market but where the historical performance of the Basket exceeded that. 

(2)          However, it is apparent that this was required to ensure the success of the scheme in delivering remuneration with the intent that the usual tax charges would not apply (see [238], [252] and [253]). In addition to the comments already noted in these conclusions (see [250] and [251]), Mr Forsyth said that the whole transaction “is based on an arbitrage” between “the market value of the CSO” when it’s novated, and “your confidence in…the managers beating market expectation” so “arbitrage is central to the… whole transaction” (see [75]). 

(3)          Mr Forsyth did not expressly accept that there was no discussion in deciding on the “trade” beyond this risk profile and the type of investment or the type of fund that a person would be investing in.  However, whilst he said that “there were other options” and his job was to discuss “investment strategies” with Heronden and Aston Collie he did not give any examples which indicate anything other than that his concern related solely to identifying what reference index could be used in terms of presenting the opportunity for “successful” trades, in the sense of delivering the right result in the form of the Heronden Payments and Bet Profits (see [77] and [82]). 

280.       Mr Forsyth was insistent that there was a benefit to the Individuals in entering into each CSO and each Bet together (see [68(5)], [79] and [80]).  He asserted that, as Heronden took no risk at all under the “matched” position under the relevant CSO and Bet, the Individual had less transaction costs and was required to put up less as a Stake than if he had taken out a Bet only at the required level.  I accept that Heronden may well have required a higher Stake if the Individual had entered into a Bet only.  However:

(1)          If an individual wanted simply to speculate on a financial market, he could take out a spread bet set at whatever level of “trade” he could afford to undertake (according to the Stake required) under which he would have had at least some prospect of making a profit. 

(2)          It remains the case that the ability to enter into “matched” and, thereby inherently loss making contracts, of this type cannot be said to provide any benefit to the Individual unless as part of a broader plan, namely, that the Individuals expected to divest themselves of the CSOs for no or minimal cost to them.  However, it is inherent in the “matched” design of the contracts, whereby the CSO was constructed as an onerous contract, that there was no realistic possibility that the Individuals could “trade” by selling the CSOs in the market to a wholly independent party with little cost.  For the reasons already given, the appellant was the only real candidate who would be willing to take on these onerous contracts on that basis and, indeed, I have concluded that the CSOs were specifically designed with a view to the appellant doing so.  In that context, the only “benefit” the Individuals obtained from entering into the pair of contracts was that the “matched” design specifically facilitated the appellant, as their employer, funding their receipt of Bet Profits through the corresponding Heronden Payments (albeit that this outcome was dependent on the Basket growing in value as expected according to its historical performance).

281.       Finally:

(1)          Mr Forsyth also identified factors which varied in later transactions undertaken by the Individuals or by the appellant’s clients again it seems with a view to demonstrating the commerciality of the arrangements as a spread betting “trading” strategy.  He said, for example, that a different Basket was used, that the counterparty was different, that different thresholds were used and that there was a different split between the Initial Premium and the Final Premium (see [68(2)]). There was no suggestion, however, that any of these differences affected the overall effect of or, outcome under, transactions undertaken under the Alchemy scheme. 

(2)          I do not accept Mr Bremner’s submission that the fact that the Individuals and the appellant received letters of advice from Aston Collie regarding the transactions evidences their commerciality as spread betting and related option contracts.  The advice was provided in a standard form as part of the standard process repeatedly followed in all Alchemy transactions. The Individuals and the appellant must have been fully aware of its content in substantive terms before its receipt given their role in the design of the Alchemy planning.  In my view, therefore, the provision of the advice, as with much of the documentary record, was included in the structure and its content framed as “window-dressing”.  It was included with a view to creating the impression that the appellant and the Individuals were all acting independently of each other and not in collaboration with a view to arranging for employment earnings to be extracted from the appellant.  For all the reasons already set out, that is not reflected in reality. 

Funding and contingency of the scheme’s success

282.     As noted, the appellant also drew support for its position on the basis that the Individuals provided their own funds as the Stakes for the Bets and that the outcome of the Bets and CSOs was dependent on a real contingency beyond the control of the parties, namely, the collective growth in value of a number of hedge funds.  Whilst it was the hope and, I would say the expectation, that the Basket would grow in value sufficiently to trigger the maximum loss under the CSO, (and thereby the Heronden Payments and corresponding Bet Profits) that was not guaranteed.  In fact, in some periods, the opposite result arose in respect both of transactions entered into by the Individuals and the appellants and by other users of the Alchemy scheme. 

283.     It is clear from Mr Forsyth’s evidence on the development of the Alchemy structure that it was considered to be essential to its success that it was specifically designed to be dependent on a factor beyond the parties’ control (see [28] to [35] , in particular, [30]).  Whilst he appeared to suggest that it was key to the success of the planning as he described it in his witness statement as an “investment” strategy or at the hearing, as a “trading” strategy, for all the reasons already given, it is plain it was considered essential to its success in providing the Individuals with employment earnings without attracting the usual tax charges. 

284.     Mr Forsyth stressed that he put up his own money as the Stake and said that the risk of loss was very real to him (see, for example [114]).  He went so far as to suggest that individuals entered into the transactions in the expectation that they would “lose” under the Bets, in particular, where they entered into transactions following earlier cases where the individuals had in fact lost under the Bet (see [128] and [129]). As noted, he referred on several occasions to the “axiom” that historical performance of an index is no reliable guide to its future performance.  However, it is plain that, notwithstanding this “axiom”, the Individuals and the appellant undertook the scheme in the expectation that the Basket’s historical performance would be repeated in relation to the transactions.  Moreover, it is simply not credible that individuals undertook this carefully constructed scheme without at least an expectation that it was more likely than not that it would work out as planned. 

285.     I note that whilst there was a real risk of loss under the structure, it was specifically designed to mitigate the adverse effects should such a loss arise.  In effect a loss made by the Individual under the Bet (in the maximum amount of the Stake less the Initial Premium) was simply recycled to a vehicle controlled by the Individuals in that it was wholly off-set by a corresponding gain for the appellant (except to the extent of Heronden’s fees).  Mr Forsyth agreed that the fact the appellant made a gain in those circumstances provided a “cushion” and that the resulting funds within the appellant were potentially available to him (see [119] and [120]).  He said, however, that he did not regard the appellant’s monies as his own and that there was no prior agreement or understanding that, following the structure failing, it would be re-rerun or the funds made available “automatically”; it would depend on circumstances at the time.  He noted that, for example, following the scheme’s failure the appellant may have had some large unexpected bill or be sued for a large amount.  He said that one reason why it was not run in 2014 was because the Individuals “wanted to put a larger pool aside to pay for contingencies” and there was the “very real risk that the trade simply wouldn’t work” (see [116] to [120]).

286.      I note, however, that the Alchemy structure was in fact re-rerun on a number of occasions following the failure of two of the later iterations undertaken by the Individuals and the appellant (in December 2013 and October 2015).  Whilst Mr Hughes did not participate in any iterations after December 2013 because he ceased to be a director in December 2014, his loss on the December transaction was cushioned by the sum paid to him by Mr Forsyth and Ms Baker when he left the company (see [120]).  Following the failure of the relevant transactions, the Individuals may well have been concerned that they/Heronden had not identified funds with a sufficient chance of “beating the market” and, hence, there was a relatively prolonged period in 2014 when they did not undertake Alchemy transactions.  However, in all the circumstances, I cannot see that their concern was sparked by anything other than a desire to source the right sort of index reference to use in the transactions to maximise the chances of the planning when re-used delivering the wanted level of employees earnings. 

287.      I accept, of course, that money in a company is not a fund that its employees/directors can draw on freely.  In this case the three directors would have to agree collectively on how to deal with any funds representing a profit made by the appellant under a CSO.  I accept that there may well have been no prior formal agreement that the scheme would be re-run or agreement that monies would be paid out “automatically” as whether this could be done would depend on the circumstances at the time. 

288.     Overall, however, it is reasonable to suppose that the Individuals intended and expected that they would co-operate and collaborate to ensure that the appellant returned funds they “lost” under the Alchemy structure to themselves in the appropriate amount (whether under the Alchemy structure or in some other way) in view of the fact that they were the only three shareholders/directors of the appellant, who together controlled it, and that they all undertook the same planning on the same occasions and were in the same boat should it not succeed.  Moreover, that is in fact what happened in practice given that the scheme was re-run, following its failure, on many occasions and given the substantial amount paid to Mr Hughes when he left the business.   

289.     On that basis, the fact that the Individuals may have used their own funds to finance the net amounts required to participate under the Bets (the Stake less the Initial Premium) is not a material factor which detracts from any of my conclusions.  In effect, barring a wholly unexpected event, the Individuals could reasonably expect that any funds they “lost” when the scheme was unsuccessful would be returned to them in one way or another.  In any event, I note that no evidence was provided of the precise source of the funds which the Individuals used to pay these net amounts.  Plainly it could well be the case that the funds received as Bet Profits from the July transactions could have been used to fund the net amounts required for the October and December transactions (and funds from the earlier 2011 transaction could have been used to fund the net amounts required for the July transactions).  

290.       Overall, as Mr Forsyth’s own comments indicate, structuring the contracts to include a real commercial risk of loss was intended to give the hoped-for payments under the CSOs and Bets the character for tax purposes of financial payments made in settlement of commercial financial transactions.  However, in my view this in-built risk, on which the success of the planning hinged, does nothing more than cast a thinly veiled disguise over the true purpose and tax effect of the transactions.  For all the reasons set out in my conclusions above (see [238]), in all the circumstances the fact that the contracts which delivered the Heronden Payments and corresponding Bet Profits were specifically constructed to be dependent on a contingent outcome, does not detract from their character as employment earnings.

291.     Finally, I note that the appellant argued that the structure would not be fit for purpose as a scheme to extract money tax efficiently from the appellant because the appellant would have been subject to corporation tax on its winnings. I do not know if the appellant did in fact pay such tax but in any event I would view any such tax costs simply as a further price the parties were willing to pay with a view to achieving the objective of extracting employment earnings from the appellant without the usual income tax and NICs charges.

Appellant’s argument on Abbott v Philbin

292.     It is evident from my conclusions set out above that, applying the employment tax provisions on a purposive construction to the facts viewed realistically, I consider that (a) the Heronden Payments made by the appellant to Heronden under the relevant CSOs (thereby funding its payment of Bet Profits to the Individuals) are correctly to be characterised for tax purposes as earnings “from” an employment source and (b) neither those sums nor the Bet Profits are from a non-employment source, namely, as “losses” or “winnings” from undertaking financial spread bet contracts. 

293.     In my view, there is nothing in the cases Mr Bremner referred to at some length, on what is required for earnings to be “from” an employment, which detracts from the conclusion that the required factual link between the Heronden Payments and the employment is present in this case.  I do not accept Mr Bremner’s argument, made, in particular, by reference to Abbott v Philbin (as also applied in Wilcock v Eve) in effect that any employment tax charges arising in respect of the transactions are necessarily confined to the very small employment related benefits which the appellant accepts it provided to the Individuals in taking on the novated CSOs. 

294.     Mr Bremner argued that, as a result of the novation, any employment related tax charge was triggered and crystallised in full in the form of an employment related benefit as valued by Heronden.  He said that this had the effect of breaking any link or connection between the subsequent payments under the Bets and CSOs and the Individuals’ employment with the appellant; those payments necessarily flowed from the parties’ rights and obligations under the contracts, as a spread bet and related financial call option and not “from” their employment.  In that context, he emphasised that the legal effect of each novation of a CSO was that the original contract between the Individual and Heronden was extinguished and a new one was put in place between the appellant and Heronden on the same terms as the original one.

295.     I cannot see any parallels between these arrangements and the award of the share options to the individual in Abbott v Philbin.  To recap, in that case it was not in doubt that the share option was awarded as a reward for the employee’s services.  The question was whether the taxable benefit or “perquisite” arose only when the option was granted and/or when it was exercised.  The House of Lords found that (a) the grant of the option was a taxable “perquisite” when it was granted (as a valuable asset that could be turned to pecuniary account) and that (b) in any event, the exercise of the option could not give rise to a taxable “perquisite”.  That was on the basis that a profit generated by the increase in the price of the shares after the grant of the option does not relate to the employee’s services but rather:

(1)          as Viscount Simmonds said, would be due to numerous unrelated factors or the “the adventitious prosperity of the company in later years” or,

(2)          as Lord Radcliffe put it, represents “the profit of his exploitation of a valuable right” in circumstances where, given the option may be exercised many years after its grant, the market value of the company’s shares may have changed out of all recognition due to a variety of commercial factors.  He thought it would be “quite wrong to tax whatever advantages the option holder may obtain through the judicious exercise of his option rights” as if they were profits or perquisites from his office arising in the year when he exercised the option.   

296.       I do not take this decision to have set down a definitive rule that, in any case where the grant or award of an asset generates a taxable benefit or earnings, any future benefits/payments arising from that asset are necessarily not taxable as employment income or further employment-related benefits.  As HMRC submitted, this decision does not detract from the need for the tribunal to analyse whether, on a purposive approach to the construction of the legislation taking a realistic view of the facts, the relevant payments constitute “earnings from an employment”. 

297.       On that approach, this is not a situation like that in Abbott v Philbin where a clear demarcation can be drawn between (a) the realisation of an employment related benefit on the immediate award of an asset, and (b) the subsequent uncertain increases in value flowing from the ownership of that asset, which may be realised many years later and, when eventually realised, may be held to have no connection with the provision of the employee’s services.  In this case, by contrast, it is plain that, when the CSOs were novated to the appellant, the appellant expected to pay the Heronden Payments, within a relatively short period of around three months, as specifically set from the outset at the level in which the appellant wished to remunerate the Individuals.  The appellant accepted the novation of the onerous CSOs in order to provide those pre-determined cash sums through the medium of Heronden (albeit their payment was subject to the outcome of a contingency).  

298.        It requires an unrealistic and formalistic approach to the facts, which is wholly out of kilter with the modern purposive approach, in effect to draw an artificial line, as at the date when the CSOs were novated to the appellant, as the appellant seeks to do, on the basis that (a) any immediate benefit in the appellant taking on the onerous CSO for no charge is employment related but (b) the subsequent payments made under the CSOs and Bets necessarily derive their nature solely from the design of the arrangements as financial “spread bet” contracts.  On the required purposive approach and for all the reasons already set out, that is simply plainly not the case. 

299.       I cannot see that it makes any difference to this conclusion that, as a legal matter, the mechanism used in effect to transfer the Individual’s rights and obligations under the CSOs to the appellant, involves the termination of the original contract and the creation of a new one under a novation.  The fact is that it was always envisaged that the appellant would take on the Individual’s rights and obligations under the CSOs.  It seems to me that the precise legal mechanism by which that was achieved is immaterial to the analysis. 

Part D - Decision on employment benefit issue

300.     I have considered HMRC’s alternative arguments in case I am wrong in my conclusions set out above.  As noted, it was common ground that, in taking on the CSOs, the appellant provided an employment related benefit to the Individuals.   The appellant’s position is that the benefit is of a very small value under the Black-Scholes method of valuation.  The parties did not want the tribunal to consider that method of valuation at this stage. 

Submissions

301.       HMRC submitted that, in any event, it is wrong to assess the value of the benefit by reference to the market value of the CSOs as at the date of the novation to the appellant.  In their view, on a purposive approach and a realistic view of the facts, the Heronden Payments constitute the “cost” of the benefit (s 203) as defined as “the expense incurred in or in connection with the provision of the benefit” (under s 204).  They argued that, essentially, for the same reasons as set out in relation to the Ramsay issue, on the evidence, the appellant accepted taking on the potential expense of the Heronden Payments in order to reward the Individuals for their services (as reflected in the appellant’s accounts).

302.       As noted, HMRC said that if the tribunal accepts this position, HMRC will collect the resulting income tax from the Individuals.  The issue is relevant to this appeal because on HMRC’s analysis, the benefit in the form of the Heronden Payments would be treated as “general earnings” in respect of which the appellant, as employer, is liable for class 1A NICs (under s 10 SSCBA).

303.       Mr Bremner raised similar arguments to those raised in relation to the Ramsay issue:

(1)          On the basis of Abbot v Philbin (and Wilcock v Eve) the benefit provided to the Individuals on the novation of the CSOs had to be valued as at that date.  Moreover, once the novation had taken place, thereby transferring all the risks and rewards of the CSOs to the appellant, the link with the Individuals’ employment ended.  Consequently, any later payments made by the appellant on “losing” under the CSOs cannot constitute payments for the benefit of the Individuals.  On that basis, the appellant did not agree to be liable for future losses accruing to the Individuals (by contrast with, for example, Hartland v Diggines [1926] AC 289).  Moreover, as a matter of statutory construction the payments did not arise “in connection with” the employment.  The payments made under the new novated CSOs simply arose from those new contracts and not “in connection with” the novation itself. 

(2)          That the “cost” of an employment-related benefit must be determined on the date on which the benefit is provided, is consistent with the structure of the benefits code; the code treats the cash equivalent of benefits as earnings for the tax year in which the benefit is provided (under s 203(1)).  HMRC’s view leads to the difficulty that if, for example, the CSOs were novated to the appellant on 30 March 2012 and the appellant “lost” on the CSOs on 30 June 2012, the value of the benefit provided in the 2011/12 tax year could only be established by referring to events occurring in the subsequent tax year.  Similar difficulties were highlighted in Abbott v Philbin.

(3)          The benefit of the novation was valued using an industry-standard valuation method (indeed, one relied upon by HMRC in their manuals).  As part of the application of the Black-Scholes method, the risk of losing on the relevant CSO was already factored into the valuation of the novation.  HMRC’s approach, therefore, leads to double counting.

304.       HMRC responded that:

(1)          The authorities relied on by the appellant relate to the interpretation of the provisions taxing “perquisites”.  Here, it is common ground that an employment-related benefit has been provided; the question is what the “cost” of that benefit is.  The “cost” is not restricted to expenses incurred before or at the same time as the benefit is provided.  Future and contingent expenses must also be taken into account; otherwise it would be a straightforward matter for an employer to minimise the “cost” of a benefit.

(2)          The complexity that, in some cases, the “cost” of the benefit could only be quantified in the tax year after the novation took place arises from the artificial nature of the scheme implemented by the appellant.  As a practical matter, the “cost”, when quantified, can be brought into account by an amendment to the Individual’s tax return for the year in which the novation occurred or, where necessary, by an assessment by HMRC.

(3)          Alternatively, on the facts, the benefit was “provided”, for the purposes of s 203, only when the “cost” to the appellant was quantified.  The benefit was in effect enjoyed by the Individuals only when the appellant’s liability under the relevant CSO crystallised.  It was only when the Bet and CSO simultaneously came to an end that the Heronden Payments arose thereby enabling the Individuals to receive the Bet winnings.  Mr Vallat referred to the decision in Templeton v Jacobs 68 TC 735 in support of this.

305.        Mr Bremner responded that the timing difficulty is a consequence of HMRC’s analysis; it is a clear sign that that analysis is wrong.  Moreover, in his view, Templeton v Jacobs does not support HMRC’s argument.  In that case, the issue was when an employment related benefit was provided as a result of a company paying for the taxpayer’s loft to be converted into an office.  The High Court rejected the taxpayer’s argument that the benefit was not taxable on the basis that it was provided when the company contracted and paid for the work in the tax year before that in which he was engaged as the company’s employee and the loft was completed.  It was held that the benefit was provided in the later tax year.  The benefit was provided only when the office was available to be enjoyed by the taxpayer.  Mr Bremner said that, in this case, there is no doubt as to when the benefit was provided, namely, when the novation took place.  He concluded that this really is a valuation issue.  In his view, the right analysis is that there is a single benefit in kind provided at the date of novation and the cost of that benefit is given by the valuation of the option.

Decision

306.       In short, my view is, that in these circumstances, establishing the “cost” in terms of the “the expense incurred in or in connection” with the provision of any employment related benefit by the appellant on accepting the novated CSOs essentially comes down to a valuation issue.

307.       The relevant provisions do not specifically address how the cash equivalent of an employment related benefit (in the sense of the cost/expense incurred in or in connection with the provision of the benefit) is to be quantified when, as here, a person takes on a contract, for no payment, under which its liability to make and right to receive future payments is wholly dependent on a contingency beyond the contracting parties’ control.  However:

(1)        The overall effect of s 203 and s 204 is to impose a tax charge in respect of the cash equivalent of the employment related benefit in the tax year in which the benefit is provided.  That indicates that the amount of the cost/expense is to be established as at the date within the relevant tax year when the benefit is provided. 

(2)        It did not appear to be disputed that it is the appellant’s acceptance of each novated CSO which constitutes an employment related benefit.  On that basis I cannot see any scope for the view that the benefit is “provided” at any time other than when each novation took place. It was the act of relieving the Individual from the relevant onerous CSO that the benefit was enjoyed by the Individual; the Individual was then free to reap any future rewards from the Bet unencumbered by the “matched” CSO including the contingent liability to pay the Heronden Payment. 

(3)        On the assumption that the appellant provided the relevant benefit when the novation took place, in the circumstances, I can see no basis for establishing the expense incurred by the appellant in or in connection with relieving the relevant Individual of the CSO other than by valuing the bundle of contingent contractual liabilities and rights taken on as at that date.  The fact that the appellant took on the contingent liability to make the Heronden Payment is very much part of the benefit it provided but the “expense” incurred in doing so as at the time of the novation necessarily has to be valued due to its contingent nature as part of the valuation of the overall bundle of contingent rights and liabilities.

(4)        The benefits code does not appear to provide a basis for re-calculating the resulting tax charge by reference to the actual liabilities or receipts as and when any payments are made or received under the contract. 

308.       On that basis, therefore, the tribunal would require evidence on the correct valuation approach to be adopted to determine the amount of any employment related benefit provided by the appellant on the novation of the CSOs to it.  Given the parties did not present any such evidence on the basis that the question of the validity of the Black-Scholes valuation method used was to be “parked”, this issue will have to be considered further, if necessary, at a later date.

Part E - Decision on the application of Part 7A

Law

309.         Finally, I have considered whether the requirements for Part 7A to apply to the transactions are met. 

310.       Part 7A applies if the following criteria set out in s 554A are met:  

“(1) Chapter 2 applies if -

(a) a person (“A”) is an employee, or a former or prospective employee, of another person (“B”),

(b) there is an arrangement (“the relevant arrangement”) to which A is a party or which otherwise (wholly or partly) covers or relates to A,

(c) it is reasonable to suppose that, in essence -

(i) the relevant arrangement, or

(ii) the relevant arrangement so far as it covers or relates to A,

is (wholly or partly) a means of providing, or is otherwise concerned (wholly or partly) with the provision of, rewards or recognition or loans in connection with A’s employment, or former or prospective employment, with B,

(d) a relevant step is taken by a relevant third person, and

(e) it is reasonable to suppose that, in essence—

(i) the relevant step is taken (wholly or partly) in pursuance of the relevant arrangement, or

(ii) there is some other connection (direct or indirect) between the relevant step and the relevant arrangement .

(2) In this Part “relevant step” means a step within section 554B, 554C or 554D.

(3) Subsection (1) is subject to subsection (4) and sections 554E to 554Y…

…..

(5) In subsection (1)(b) and (c)(ii) references to A include references to any person linked with A.

(6) For the purposes of subsection (1)(c) it does not matter if the relevant arrangement does not include details of the steps which will or may be taken in connection with providing, in essence, rewards or recognition or loans as mentioned (for example, details of any sums of money or assets which will or may be involved or details of how or when or by whom or in whose favour any step will or may be taken).

(7)  In subsection (1)(d) “relevant third person” means -

(a) A acting as a trustee,

(b) B acting as a trustee, or

(c) any person other than A and B…..

……..

(11) For the purposes of subsection (1)(e)—

(a) the relevant step is connected with the relevant arrangement if (for example) the relevant step is taken (wholly or partly) in pursuance of an arrangement at one end of a series of arrangements with the relevant arrangement being at the other end, and

(b) it does not matter if the person taking the relevant step is unaware of the relevant arrangement.

(12) For the purposes of subsection (1)(c) and (e) in particular, all relevant circumstances are to be taken into account in order to get to the essence of the matter.

311.       For the purposes of s 554A, an “arrangement” includes any “agreement, scheme, settlement, transaction, trust or understanding (whether or not it is legally enforceable)”.

312.       HMRC argued that the for the purposes of the above provisions there was a relevant step under s 554C(1)(a) which provides as follows:

( 1)  A person (“P”) takes a step within this section if P—

(a) pays a sum of money to a relevant person….

(2)  In subsection (1) “relevant person”—

(a) means A or a person chosen by A or within a class of person chosen by A, and

(b) includes, if P is taking a step on A’s behalf or otherwise at A’s direction or request, any other person.

(3) In subsection (2) references to A include references to any person linked with A…..”

313.       Where these provisions apply:

(1)          The value of the relevant step identified counts as employment income of A in respect of A’s employment with B (where the step is taken during A’s employment) for the tax year in which the relevant step is taken (s 554Z2(1)).   In this case, therefore, if the provisions apply, the value of the relevant steps would count as the Individuals’ employment income in the tax year in which the relevant step was taken. 

(2)          When the value of a relevant step counts as employment income under Part 7A, and that relevant step is the payment of a sum of money, an employer is treated as making a payment of PAYE income (s 687A).  It did not appear to be disputed, therefore, that if Part 7A applies the appellant would be liable to account for income tax and Class 1 NICs under the PAYE system  (under ss 687A and 710 and regulation 22B of the Social Security (Contributions) Regulations 2001/1004 (the “Social Security Regulations”) 

314.        Mr Vallat noted, as also did not appear to be disputed, that if Part 7A applies, there would be additional income tax and NICs charges as a result of the application of s 222.  This applies where (a) an employer has made a notional payment which includes a payment of earnings under Part 7A (under s 710), (b) the employer is required, by virtue of s 710(4), to account to HMRC for an amount of income tax in respect of that notional payment; and (c) the employee does not, before the end of the period of 90 days from the date on which the employer is treated as making the notional payment, make good the due amount of income tax to the employer.  The due amount of income tax is treated as earnings from the employment for the tax year in which the notional payment is treated as having been made.

315.       To the extent that an amount counts as employment income under s 554Z2, that would produce a “due amount” of income tax within s 222 which the Individuals have not made good.  The additional income tax arising would be due from the Individuals.  However, any such amounts are also treated as earnings for NICs purposes so that the appellant would be liable to class 1 NICs in respect of the relevant amount (under regulation 22(4) of the Social Security Regulations).

Submissions

316.         Mr Vallat argued that:

(1)          On the basis that factual conclusions drawn from the evidence are as set out in HMRC’s submissions in relation to the Ramsay issue, all the transactions, which form the Alchemy structure together constitute an “arrangement” within the meaning of s 554A.  He noted that, even on the appellant’s own case, the Alchemy transactions are concerned with remuneration albeit that the appellant only accepts that there was a limited employment related benefit arising on the novation of the CSOs. 

(2)          Heronden, as a “relevant third party” took relevant steps in paying (a) the Initial Premium and the maximum winnings to the Individuals under the Bets, (b) the Stakes to the Individuals when returned on the Individuals “winning” under the Bets, and (c) the Final Premium to the appellant (as a linked party) (within the meaning s 554(C)).  (I note that it was not disputed that the appellant is a linked party.) The relevant steps were taken in pursuance of the relevant arrangement (within the meaning of s 554A(1)(e)).  Each of these steps was taken by Heronden in pursuance of the relevant arrangement; they were all part of the pre-ordained series of transactions undertaken in order to implement the Alchemy scheme.  Mr Vallatt noted that this requirement is satisfied even if Heronden was not aware of the relevant arrangement (due to s 554A(11)) (although he thought the evidence confirmed that Heronden was aware of the plan). 

317.  Mr Bremner submitted that Part 7A does not apply on the basis that the threshold requirements in s 554A are not met.  He made similar points to those raised in relation to the other issues in this appeal:

(1)           There was only one discrete transaction, the novation of the CSOs, which involved the provision of an employment-related reward; only that transaction could be a “relevant arrangement”.  However, the payments made by Herondon under the Bets and CSOs were not made under that arrangement but pursuant to the terms of the separate contractual agreements between Heronden and the Individuals and between Heronden and the appellant.  Transactions which were not made to provide, and were not concerned with, the provision of rewards in relation to the Individuals’ employment with the appellant cannot form part of a “relevant arrangement”.

(2)          In any event, even viewing the “relevant arrangement” in a broader sense the requirement of sub-s 554A(1)(c) is not satisfied on the basis of the evidence set out above, in particular, that: (a) each Individual approached Heronden to enter into the relevant Bet and the CSO, having taken advice from Aston Collie;  (b) the terms of the Bets and the CSOs were determined by each Individual, in accordance with his/her individual requirements, including appetite for risk, and not by any arrangement or plan on the part of the appellant to provide employment-related rewards to the Individuals.  

(3)          Even if Heronden could be regarded as taking a relevant step in making payments to the Individuals and the appellant, it is not reasonable to suppose that (in essence) such steps were made in pursuance of the relevant arrangement or that there some other connection between the relevant step and the relevant arrangement (under s 554A(1)(e)).  Heronden is an unconnected third party which contracted with the parties on arm’s length terms.  The payments it made were not connected with arrangement to provide rewards to employees; they were made under the contractual terms of the Bets and CSOs, which Heronden entered into initially with the Individuals acting on their own account and, having agreed to the novation of the CSOs, separately with the appellant.

(4)          The step by step approach HMRC wish to take is inconsistent with their Ramsay approach as set out above. 

(5)          It cannot be correct that the Stake should be brought into tax.  That is a sign that HMRC’s analysis has gone badly wrong. 

318.          Mr Vallat responded that the appellant’s position that, for the purposes of s 554, the “relevant arrangement” is restricted to the novation of the CSOs is an absurdly narrow interpretation.  However, in his view, the analysis remains the same even if the relevant arrangement is regarded as being confined in that way.  Nonetheless the relevant payments identified at [316(2)] were made in pursuance of or, in connection with, that arrangement because that arrangement was set up in advance to allow for those payments to be made for the Individuals’ benefit. 

319.       As regards the Stake, he said that there is simply nothing in the provisions to limit their application to cases where the payment is intended to confer a reward or an economic gain.  In certain cases, where a relevant step is made using already taxed funds, there is an exclusion to prevent the recycling giving rise to a second charge (under s 554Z5).  But here the Individuals have made a virtue of the fact that they used their own resources to fund the Stake, which they assert was not taxed employment income, so that this exclusion does not apply. 

Conclusion on application of Part 7A

320.       In short, I have concluded that, assuming that my conclusion on the Ramsay issue is not correct, in any event, the provisions of Part 7A apply for the reasons put forward by HMRC.  It is readily apparent from my detailed conclusions on the facts that the conditions of s 554A and 554C are satisfied.   On my findings of fact, it is plain that:

(1)          There was a relevant arrangement to which the Individuals were a party, in the form of all the steps involved in the transactions, which it is reasonable to suppose was, in essence, (wholly or partly) a means of providing, or was otherwise concerned (wholly or partly) with the provision of, rewards in connection with the Individuals’ employment with the appellant. 

(2)          A relevant step was taken by a relevant third person, namely, Heronden, in the form of making the relevant payments, which, it is reasonable to suppose, in essence, was taken (wholly or partly) in pursuance of the relevant arrangement.  For the reasons set out above, I consider that it is implausible that Heronden was not aware that the intention was that, under the transactions, the appellant would provide sums as a reward for the Individuals’ employment services.  In any event, as HMRC noted, even if Heronden was not aware of the relevant arrangement, that would not itself prevent this requirement being satisfied (due to s 554A(11).

321.       However, I would question whether the funds paid by Heronden to the Individuals in respect of the Stakes they placed under the Bets should be regarded as falling within s 554C(1)(a). 

(1)          The Individuals were required to pay the Stakes to Heronden when they entered into the Bets as representing the maximum loss they stood to “lose” under the Bets.  When they in fact “won” the maximum winnings under the Bets, Heronden returned the funds representing the Stakes to them because they were not, in those circumstances, required to forfeit those amounts.  The Individuals did not, therefore, in an economic sense gain anything from the return of those funds. 

(2)          In my view, it is unlikely that Parliament intended to capture within these provisions, as a payment of a sum of money, a movement of monies from one person to another where in substance there is no economic gain for the recipient.  The whole tenor of these provisions, broadly drawn as they are, is to capture sums which are in reality paid by way of economic gain as disguised employment earnings.

Part F Decision on application to amend HMRC’s statement of case

Background

322.       On 16 May 2018, HMRC applied to the tribunal to amend its statement of case to include the Rangers argument.  The appellant objected at length in submissions sent to the tribunal on 22 May 2018.  The tribunal decided that this issue would have to be dealt with at the start of the hearing. 

323.       By way of background:

(1)          The appellant submitted its appeal to the tribunal on 18 November 2016. HMRC produced a statement of case on 27 June 2017 (having initially submitted what appeared to be a draft on 26 May 2017) and, on 17 October 2017, served their witness evidence from Mr Barrett.

(2)          HMRC pointed out that on 22 January 2018, they wrote to the appellant stating they intended to issue “follower notices” consequent on the Supreme Court’s decision in Rangers under the provisions set out in Part 4 of the Finance Act 2014. 

(3)          On 25 January 2018, the appellant wrote to HMRC noting amongst other things:  “We note that no reference is made to [the Rangers argument] either in HMRC’s Statement of Case or in your statement….There is no doubt that the Tribunal is the appropriate forum for the matter to be determined”.

(4)          On 16 March 2018, HMRC sent the appellant determinations and decisions in relation to further iterations of the Alchemy structure in respect of the tax years 2013/14, 2014/15 and 2015/16.   

(5)          On 5 April 2018, HMRC issued a “follower notice” to the appellant relating to the 2012/13 tax year.

(6)          On 12 April 2018, the appellant submitted a notice of appeal to HMRC in respect of the decisions and determinations issued in respect of the later tax years.

(7)          On 1 May 2018, HMRC wrote to the appellant stating that they wished to rely on the Rangers argument.  

(8)          On 14 May 2018, the appellant responded to HMRC’s letter stating that HMRC would have to make a formal application to the tribunal to amend its statement of case and that the appellant would resist it. 

324.       It was common ground that, to the extent that the tribunal has to consider whether the application should be allowed under the overriding objective governing the tribunal, of dealing with matters fairly and justly, the tribunal should follow the approach set out by Carr J in Quah Su-Ling v Goldman Sachs International [2015] EWHC 759 at [38]: 

“(a) whether to allow an amendment is a matter for the discretion of the court. In exercising that discretion, the overriding objective is of the greatest importance. Applications always involve the court striking a balance between injustice to the applicant if the amendment is refused, and injustice to the opposing party and other litigants in general, if the amendment is permitted;

(b) where a very late application to amend is made the correct approach is not that the amendments ought, in general, to be allowed so that the real dispute between the parties can be adjudicated upon.  Rather, a heavy burden lies on a party seeking a very late amendment to show the strength of the new case and why justice to him, his opponent and other court users requires him to be able to pursue it. The risk to a trial date may mean that the lateness of the application to amend will of itself cause the balance to be loaded heavily against the grant of permission;

(c) a very late amendment is one made when the trial date has been fixed and where permitting the amendments would cause the trial date to be lost. Parties and the court have a legitimate expectation that trial fixtures will be kept;

(d) lateness is not an absolute, but a relative concept. It depends on a review of the nature of the proposed amendment, the quality of the explanation for its timing, and a fair appreciation of the consequences in terms of work wasted and consequential work to be done;

(e) gone are the days when it was sufficient for the amending party to argue that no prejudice had been suffered, save as to costs. In the modern era it is more readily recognised that the payment of costs may not be adequate compensation;

(f) it is incumbent on a party seeking the indulgence of the court to be allowed to raise a late claim to provide a good explanation for the delay;

(g) a much stricter view is taken nowadays of non-compliance with the Civil Procedure Rules and directions of the Court. The achievement of justice means something different now. Parties can no longer expect indulgence if they fail to comply with their procedural obligations because those obligations not only serve the purpose of ensuring that they conduct the litigation proportionately in order to ensure their own costs are kept within proportionate bounds but also the wider public interest of ensuring that other litigants can obtain justice efficiently and proportionately, and that the courts enable them to do so.”

325.       HMRC said that the tribunal plainly has jurisdiction to hear the Rangers argument and it is in the interests of justice and fairness for it to be heard.   The appellant disputed both of these points.

Jurisdiction

Submissions

326.       It is common ground that determinations are treated as assessments which the tribunal can, on an appeal made to it, reduce the amount of or increase under the provisions of s 50 of the Taxes Management Act 1970 (“TMA”).  The dispute was whether the Rangers argument falls within the scope of the conclusions made on issue of the determinations and thereby within the scope of the appeal proceedings and the tribunal’s powers under s 50 TMA.   HMRC referred to the leading authority on this topic, namely, Tower MCashback LLP 1 and Another v Revenue and Customs Commissioners [2011] 2 AC 457, [2011] STC 1143 as the decision in that case was interpreted by the Court of Appeal in Fidex Ltd v HMRC [2016] EWCA Civ 385, [2016] STC 1920.  Whilst those cases are concerned with the scope of closure notices rather than specifically determinations, it was not disputed that the same considerations apply.

327.       In Fidex, Kitchen LJ set out, at [45], that the principles to be applied as regards the scope of a closure notice are those set out by Henderson J in Tower MCashback as approved by and elaborated upon by the Supreme Court which, so far as material to that appeal, he thought could be summarised in four propositions:

“i) The scope and subject matter of an appeal are defined by the conclusions stated in the closure notice and by the amendments required to give effect to those conclusions.

ii) What matters are the conclusions set out in the closure notice, not the process of reasoning by which HMRC reached those conclusions.

iii) The closure notice must be read in context in order properly to understand its meaning.

iv) Subject always to the requirements of fairness and proper case management, HMRC can advance new arguments before the FTT to support the conclusions set out in the closure notice.”

328.       Mr Vallat said that it is entirely in accordance with these principles for HMRC to be permitted to raise the Rangers argument.  HMRC are seeking to include a legal argument which further supports HMRC’s existing position, namely, that on a purposive construction of the employment tax provisions each iteration of the transactions constituted a composite scheme to deliver taxable employment earnings into the hands of the Individuals.  This simply involves the application of the principle in Rangers to the particular facts of this case which, in any event, the tribunal will have to consider in detail.  The current appeal necessarily involves the question whether the payments made by the appellant to Heronden under the CSOs are tax-free.  Broadly, the appellant’s stance rests on the assertion that the sequence of steps results only in minimal tax charges and necessarily that the Heronden Payments (as well as the Bet Profits) are not liable to tax as employment earnings for any reason.

329.        Mr Vallat said that the broad scope of the determinations is clear from the covering letter accompanying them, which described them as being “in respect of the spread betting/option arrangements under the [Alchemy] scheme”.   The letter stated: 

“…We have taken this step now to protect tax and national insurance contributions in relation to payments for the services of an individual or number of individuals.

We anticipate that the company will have some knowledge of the transactions to which the liability relates.  Whatever the level of knowledge of the officers of the company has been before now, you need to know that HMRC believes that the payments are within the Income Tax (Earnings and Pensions) Act 2003 and that PAYE and NICs should have been deducted.  Without prejudice to any other arguments HMRC may wish to advance, we think the payments are either employment income on first principles or are brought within the charge to employment income by Part 7A of the Act referred to above.” (emphasis added)

330.       Mr Bremner argued that the tribunal does not have jurisdiction because HMRC have simply not issued any determination seeking to tax the Heronden Payments; they have issued them on the basis of seeking to tax the winnings under the Bet as is clear from the figures in the determinations.  The tribunal does not, therefore, have jurisdiction to increase an assessment/determination by reference to the Rangers argument; there is no assessment/determination before the tribunal that it can act in relation to.  In his view, HMRC were attempting to introduce an additional charge to tax by the back door which is wrong in principle.  Mr Bremner did not, however, make any points by reference to the caselaw set out above other than that he said that those cases did not address this issue.

331.       The appellant also said that for the tribunal to consider this argument raises the difficulty that, if the argument succeeds, the December iterations of the transactions would be taxable in the tax year 2013/14 but the tribunal only has before it an appeal in respect of the tax year 2012/13.  In response, Mr Vallat said that HMRC’s statement of case has always stated that the December transactions could result in tax charges arising in the 2012/13 period. Moreover, given the determinations were not restricted to specific iterations of the scheme, the appellant has always been on notice that the December transactions form part of the present appeal.  The appellant has not previously tried to argue that this is not before the tribunal.

Decision on jurisdiction

332.       I cannot see why the appellant appeared to think the caselaw set out above is not relevant to this issue.  In my view, on the basis of that case law, it is clear that the scope of the relevant determinations is not limited to the taxation of the “winnings” under the Bet only as is clear from the context provided by the covering letter (albeit that HMRC may have computed the outstanding tax by reference to those “winnings”).  The Rangers argument is simply another legal argument which, subject to case management considerations, HMRC may raise in support of their determinations.  It is not the case that HMRC would have to have issued determinations specifically referencing the Rangers argument in order for them to be permitted to raise it.  I can see no basis for the appellant’s stance on this.

Case management considerations

Submissions

333.       On whether it is in the interests of justice and fairness for HMRC to be able to raise the Rangers argument at this juncture, HMRC made the following main points:

(1)          The appellant was aware that HMRC intended to raise this argument since 25 January 2018 (at the latest).  

(2)          Whilst the appellant may wish to submit further evidence that should be limited (see below) given the appellant has already provided much documentary evidence (much of it at the request of HMRC) and a detailed witness statement for Mr Forsyth.

(3)          In these circumstances and, given the parties’ previous correspondence (as set out above), this application is not “very late”.  Allowing the proposed amendment would satisfy the overriding objective of dealing with cases fairly and justly by enabling the tribunal to apply recent developments in the law, as explained by the Supreme Court.

(4)          HMRC will rely on the Rangers argument in the many related appeals before the tribunal (alongside the other arguments relied upon in this appeal).  It would be a more effective use of the tribunal’s time for it to consider all of the arguments (including the Rangers argument), given that the reasoning of the tribunal will apply equally to the other appeals.  If appeals relating to this scheme were to proceed to the Upper Tribunal, HMRC would submit that the appeals should be case-managed so that the Upper Tribunal is able to consider all of the arguments.  If the Upper Tribunal were not willing to consider the Rangers argument in this appeal, HMRC would submit that this case would not, on its own, be a suitable lead case in the Upper Tribunal.  

334.       The appellant objected as follows:

(1)          On the basis of the principles set out in Quah, it is not in the interests of fairness and justice for HMRC to be able to raise this argument at this stage.  Mr Bremner emphasised the delay and prejudice to the appellant.

(2)          On delay, Mr Bremner said that HMRC have not provided any reason (let alone any good explanation) for their delay in raising the Rangers argument.  They have had ample time to do so given that the decisions of the Court of Session and Supreme Court in Rangers were released on 4 November 2015 and on 5 July 2017 respectively.  

(3)          He added that it is simply irrelevant that the Rangers argument was referred to in correspondence in early 2018:

(a)          HMRC are not entitled to raise this argument unless and until permission is given for them to amend their statement of case.  The function of a statement of case is to set out HMRC’s position in relation to the case.  The tribunal in BPP University College v HM Revenue & Customs Commissioners [2014] UKFTT 644 (TC) at [73] highlighted the paramount importance of HMRC articulating their case in full.  An argument does not fall within the scope of the matters to be addressed by the other side unless and until it has been properly included in a party’s pleadings (see Bourke v Favre [2015] EWHC 277 (Ch) at [20]). 

(b)          The correspondence in relation to the “follower notices” clearly indicates that the Rangers argument would be addressed by the follower notice procedure rather than in an appeal before the tribunal.

(c)          It cannot properly be contended that the appellant should have been prepared to address the Rangers argument before the tribunal from 25 January 2018 onwards in the teeth of HMRC confirming consistently (until 1 May 2018) that they would not be raising that argument at the tribunal.  Nor did they address the Rangers argument in their witness evidence. 

(d)         HMRC have failed to address the fact that when applying the Quah criteria, the tribunal must take a strict view of a party’s non-compliance with its rules and directions in light of the Supreme Court’s decision in BPP Holdings Ltd v HM Revenue & Customs Commissioners [2017].  The belated nature of the application and the total absence of any explanation as to why it has been made so late in the day, after the provision of the witness evidence, are, of themselves, fatal to the application.

(4)           The appellant would suffer serious prejudice if an application to amend the statement of case is granted at this late stage given the further evidence that the appellant would want to adduce:

(a)          The whole point of the application was, Mr Bremner said, “to fix upon a different transaction” to the one that that had been under consideration which “shifts the focus utterly from the winnings under the Bet to the payment to the counterparty”. It was obvious that, had the case been pleaded as it should have been from the start, the evidence that the appellant would have led may well look different.  He said that there was “no way that that can fairly be produced at this hearing or somehow dealt with in lengthy examination-in-chief of Mr Forsyth and it’s not fair that it should be dealt with in that way”. 

(b)          Mr Bremner said that if HMRC were permitted to run the argument, he had all sorts of points and submissions that would need to be made. In outline the argument would be that the employee’s remuneration was not redirected to Heronden but on the contrary “it is the employer taking on the risks and rewards of an option and the employer who could have made a profit, could have made a loss”.  He said that “lots of time will be needed to deal with all of that” and there would need to be further evidence for these points properly to be dealt with.

(c)            He gave by way of example of the type of additional evidence that may be required: the nature of the relationship between the appellant and its employees, the nature of the employment contracts, the remuneration policy, how much each employee received as remuneration from year to year, when they were paid and received those amounts, how those amounts were calculated, who was making the decisions, what criteria were applied, evidence about the services actually carried out and what the appellant’s intention was in taking on the CSOs.  The appellant may have wanted to call the other directors as witnesses and possibly evidence from Heronden. 

335.       In its initial objection the appellant also raised further points by reference to the follower notice regime.  However, these were not addressed at the hearing.  I note that I consider it plain that that regime has not impact on this issue; it is in effect an entirely separate regime. 

336.       HMRC responded that:

(1)          The Rangers argument was not relied upon previously simply because the Supreme Court had not yet released its decision.  HMRC took some time to consider its impact and then notified the appellant of its intention to rely on it as set out above.    

(2)          It cannot seriously be maintained that the sort of evidence the appellant refers to is not required already to the extent the appellant wishes to adduce it.  All of the factors listed, if they are relevant, would be equally relevant to HMRC’s Ramsay case, namely, that the Bet, the CSO and the novation formed a series of preordained transactions designed to deliver a cash payment to the employee as a reward for the services.

Decision on case management considerations

337.       I decided that, in all the circumstances, it was in the interests of fairness and justice for HMRC to be able to raise the Rangers argument given the importance of the tribunal considering all relevant authorities and that I could see no prejudice to the appellant.  The appellant cannot have been in any doubt since January 2018 that it was HMRC’s intention to raise the effect of this case from the correspondence in relation to the follower notices.  Even if the date the appellant should have taken this on board as a real possibility is taken to be 16 May 2018, that provides ample for the appellant’s advisers to consider and deal with the legal argument. The Rangers argument is simply an extension of HMRC’s Ramsay argument made by reference to the latest leading authority on that topic in the context of employment taxes.  I could not see that this required any major change in approach by the appellant’s legal advisers.

338.       In that context, I did not accept that the appellant would need to produce extensive further evidence to deal with the Rangers argument.  All the matters which Mr Bremner raised seemed to me to be just as relevant to the Ramsay issue as put before HMRC sought to raise the Rangers argument, yet the appellant had not sought to bring that evidence.  I note, in particular, that the appellant did not call the other directors or Heronden as witnesses but only relied on the witness evidence of Mr Forsyth.  Moreover, I considered that the timetable of the hearing could be managed to enable the appellant to  provide a further witness statement from Mr Forsyth, if the appellant considered it necessary for his evidence to be expanded to deal with any factual points arising in respect of the Rangers argument.  I, therefore, adjourned the hearing at lunchtime on the first day until the following morning to allow the appellant time to do so.

339.       I note that Mr Forsyth produced another short witness statement on the following morning in which he set out some details of the appellant’s remuneration policy as recorded in the evidence in Part B.  He said that he had contacted Heronden with a view to a representative appearing as a witness but they had not responded.  Nothing more was said by Mr Forsyth or Mr Bremner about Heronden appearing as a witness in the following days of the hearing.  I note that, in the event, Mr Bremner’s arguments as regards the Ramsay issue were not materially different as regards the taxability of the Heronden Payments or the Bet Profits and the addition of the Rangers argument did not interfere with the timetable for the hearing which was concluded before the end of the scheduled five days. 

Conclusion

340.       For all the reasons set out above, the appeal is dismissed on the basis that, at the parties’ request, this is a decision in principle only.  The parties may apply to the tribunal for the tribunal to determine the amount of the resulting income tax and NICs charges if the parties fail to reach agreement. 

341.       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.

 

 

 

HARRIET MORGAN

 

TRIBUNAL JUDGE

RELEASE DATE: 10 DECEMBER 2019