United Kingdom Special Commissioners of Income Tax Decisions

You are here: BAILII >> Databases >> United Kingdom Special Commissioners of Income Tax Decisions >> Drummond v Revenue & Customs [2007] UKSPC SPC00617 (05 July 2007)
URL: http://www.bailii.org/uk/cases/UKSPC/2007/SPC00617.html
Cite as: [2007] UKSPC SPC617, [2007] STC (SCD) 682, [2007] UKSPC SPC00617, [2007] STI 1818

[New search] [Printable RTF version] [Help]

Drummond v Revenue & Customs [2007] UKSPC SPC00617 (05 July 2007)

Spc00617

CAPITAL GAINS TAX – Computation of gain – Second hand life assurance policy – Surrender proceeds brought into computation of chargeable event gain for income tax – Whether surrender proceeds to be excluded as disposal consideration for CGT purposes – No – TCGA 1992 s37(1)

CAPITAL GAINS TAX – Acquisition consideration – Wholly and exclusively for the acquisition of the asset – Second hand life assurance policies acquired for £210,000 above surrender value as part of tax avoidance scheme – Whether entire consideration incurred for acquisition of the policies or as part of pre-ordained tax avoidance scheme – Whether alternatively the £210,000 was consideration for acquisition of the policies – No – Appeal dismissed – TCGA s.38(1)

THE SPECIAL COMMISSIONERS

JASON DRUMMOND Appellant

and –

THE COMMISSIONERS FOR HER MAJESTY'S REVENUE & CUSTOMS Respondents

Special Commissioner: SIR STEPHEN OLIVER QC

Sitting in public in London on 17, 18 and 19 April 2007

Patrick Way and Hui Ling McCarthy, counsel, instructed by McKie & Co Ltd, for the Appellant

Timothy Brennan QC and Nicola Shaw, counsel, instructed by the general counsel and solicitor for the Commissioners, for the Respondents

© CROWN COPYRIGHT 2007

DECISION

Mr Jason Drummond appeals against the amendment to a self-assessment return for the year to 5 April 2001. The effect of the amendment is to disallow the sum of £1,962,233 as an allowable loss for capital gains tax ("CGT") purposes. Mr Drummond's self-assessment return had sought to deduct that amount ("the disputed loss") in computing his chargeable gains for the year.

Mr Drummond's claim to deduct the disputed loss is based on two transactions in which he participated:

(i) On 4 April 2001 he contracted to buy five second hand "non-qualifying" life assurance policies from a company called London & Oxford Capital Markets Plc ("London & Oxford") for a stated consideration of £1,962,233. Those five policies had each been created on 23 February 2001 on payment of single premiums of £250 by the life assured (Ellen Sedgley). On 26 March 2001 Ms Sedgley assigned them to London & Oxford which then topped each of them up with further premiums of £349,750 on 30 March 2001.

(ii) On 5 April 2001, when the surrender value of the five policies was £1,751,376, Mr Drummond requested London & Oxford to surrender the five policies. London & Oxford complied and the same day notified the life assurance company of its wish to encash the policies.

Following London & Oxford's encashment of the policies at Mr Drummond's direction, it was sent a "chargeable event certificate" as required by section 552(1) of Income and Corporation Taxes Act 1988 ("ICTA"). The certificate describes London & Oxford as "policy owner" and the nature of the chargeable event as "surrender". The certificate certifies the gain in respect of each policy as £274.38 and states:

"This policy gain represents income and should, therefore, be included in your tax return for the year coinciding with the event date."

Mr Drummond's self-assessment return prior to amendment by the Revenue had been compiled, so far as it related to his disposal of his interests in the five policies, on the basis that he had an allowable loss of (in round figures) £1.96 million. Section 37(1) of Taxation of Capital Gains Act 1992 ("TCGA"), as he interpreted it, required the amount paid to London & Oxford on surrender of the five policies, i.e. £1,751,376, to be excluded from the consideration for the disposal of the policies; this was because that amount had, as he understood the words of section 37(1), been taken into account as a receipt in computing the "income or profits or gains or losses of the person making the disposal" (namely himself) for the purposes of income tax.

Two preliminary points need to be made. First, second hand policies are made chargeable assets by section 210(2) of TCGA. Second, much of the legislation referred to in this decision has been substantially altered since 2000/2001.

The Revenue's grounds for amending Mr Drummond's self-assessment can be summarized as follows:

(i) Mr Drummond is wrong about section 37(1) of TCGA. The £1,751,376 paid by the insurance company on surrender was not taken into account in computing Mr Drummond's income or profits or gains. It was brought into the earlier calculation of the gain treated as arising in connection with each of the five policies for the purpose of determining the amount deemed by section 547(1) to form part of Mr Drummond's total income for the year to 5 April 2001. (I refer to this as "the section 37(1) issue".)

(ii) The difference, of some £210,000, between what Mr Drummond agreed to pay for the five policies, i.e. £1,962,233, and the amount paid out to London & Oxford following the surrender the next day, £1,751,376, is to be excluded, by reason of section 38(1)(a) of ICTA as acquisition consideration in computing Mr Drummond's gain or loss on the policies. The £210,00 was not in the circumstances consideration given by him "wholly and exclusively for the acquisition of" the five policies. (I refer to this as "the £210,000 wholly and exclusively issue".)

(iii) No part of the £1,962,233 is, in the circumstances, to be regarded as acquisition consideration. It was not incurred wholly and exclusively for the acquisition of any asset. It was incurred for no purpose other than to facilitate a tax avoidance scheme. (I refer to this as "the wider wholly and exclusively issue".)

(iv) If the Revenue were wrong on the section 37(1) issue and the £1.75 million paid out on surrender came within the expression "taken into account as a receipt in computing income … of the person making the disposal for the purposes of the Income Tax Acts", it would, say the Revenue, follow that the premiums paid should be excluded as acquisition expenditure. This is because the expenditure incurred by Mr Drummond was so closely related to the premiums paid by London & Oxford that, viewed realistically in the context of the tax avoidance scheme, Mr Drummond's expenditure should be regarded as having been allowed for income tax purposes and should, therefore, be excluded from the computation for CGT purposes by virtue of section 39(1) of TCGA.

The section 37(1) issue

Both parties approached this issue as a "black-letter" legal exercise. For this exercise I can in broad terms simplify the scenario to the following transactions:

(i) The life assured (Ms Sedgley) effects the policy paying an initial premium of £1,250.

(ii) Ms Sedgley then assigns the policy to London & Oxford for £1,275 (producing a "chargeable event gain", see below, of £25).

(iii) London & Oxford pay a further premium of £1,748,750 bringing the total premiums paid up to £1,750,000.

(iv) London & Oxford transfer the policy to Mr Drummond in return for a consideration stated to be £1,962,233.

(v) Mr Drummond surrenders the policy and receives surrender proceeds of £1,751,378.

(Five policies were used: but for simplicity I have explained the scenario with reference to a single policy.)

Section 37(1) of TCGA provides:

"There shall be excluded from the consideration for a disposal of assets taken into account in the computation of the gain any money or money's worth charged to income tax as income of, or taken into account as a receipt in computing income or profits or gains or losses of, the person making the disposal for the purposes of the Income Tax Acts."

The critical question is this. What money or money's worth has been taken into account in computing income or profits or gains of Mr Drummond (the person making the disposal) for income tax purposes? The case for Mr Drummond is that the entire surrender proceeds of the policy will have been taken into account as a receipt in computing Mr Drummond's "chargeable event gain" and is therefore to be excluded from the disposal consideration by reason of section 37(1). This calls for consideration of the chargeable event gain provisions.

The "chargeable event gain" treated as total income by section 547(1) of ICTA was certified as £1,351.35. This figure results from the application of sections 539 to 541 found in Chapter II of Part XIII of ICTA.

The purpose of Chapter II is stated in section 539(1). This is the imposition of charges to tax "in respect of gains to be treated in accordance with this Chapter as arising in connection with" three types of insurance contract, namely "policies of life insurance, contracts for life annuities and capital redemption policies". The Chapter applies to transactions in relation to such policies after 19 March 1968. (By then the CGT code had been in operation for three years.) Section 540 defines chargeable events and section 541 provides for the computation of gain on the happening of a chargeable event in relation to policies of life assurance.

A chargeable event "in relation to a policy of life insurance", which is as here a non-qualifying policy, includes "the surrender in whole of the rights conferred by the policy" (section 540(1)(a)(iii)). Here the surrender notice was served on 5 April 2001.

Section 541(1) provides:

"On the happening of a chargeable event in relation to any policy of life insurance, there shall be treated as a gain in connection with the policy …

(b) if the event is … the surrender in whole of the rights thereby conferred, the excess (if any) of the amount or value of the sum payable … by reason of the event … over the sum of the following –

(i) the total amount previously paid under the policy by way of premiums and

(ii) the total amount treated as a gain … on the previous happening of chargeable events."

Mr Drummond's case, as noted, proceeds by the following steps. The £1,751,376, is "the amount payable by reason of the event" from which the total of the premiums is to be subtracted "in computing the gain in connection with the policy". That same amount therefore comes within the words of section 37(1) of ICTA as money taken into account as a receipt in computing the income or profits or gains of Mr Drummond for income tax purposes. It follows that the £1,751,376 must be excluded as disposal consideration for CGT purposes.

Section 547 completes the chargeable event gain provisions of ICTA. It provides the method for charging the gain to tax. Subsection (1) provides –

"Where under section 541 … a gain is to be treated as arising in connection with any policy …

(a) if, immediately before the happening of the chargeable event in question, the rights conferred by the policy … or contract were vested in an individual as beneficial owner … the amount of the gain shall be deemed to form part of that individual's total income for the year in which the event happened;"

It is then the duty of the insurer, where a chargeable event "has happened in relation to any policy", within three months to deliver a certificate specifying the name of the policyholder and the nature and date of the event. The certificate must also specify the surrender value and the amount of the premiums previously paid under the policy. In the scenario set out above, the chargeable event gain was as follows:

Surrender proceeds £1,751,376

Less: initial premium £ 1,250

Less: additional premium paid London & Oxford £1,748,750

Less previously certified gain £ 25

Chargeable event gain £ 1,351

Returning to the question of the application of section 37(1): has the sum of £1,751,376 been taken into account "as a receipt in computing the income or profits or gains or losses" (my italics) of Mr Drummond for income tax purposes? Yes, argues Mr Drummond. That amount has, by operation of section 541 of ICTA, been brought into the calculation of the gain arising in connection with the policy. And the gain, so calculated, is deemed by section 547(1) to form part of his total income for the year in which the event happened.

In common with the Revenue, I disagree with Mr Drummond's construction. The calculation required by section 541(1)(b) brings into the reckoning amounts that may have had nothing to do with the surrendering policyholder. Neither the premiums paid by, nor the chargeable event gain, of Ms Sedgley, nor the topping-up premium paid by London & Oxford nor the surrender proceeds of £1,751,376 were, in terms of section 37(1), moneys taken into account as receipts in computing Mr Drummond's income or profits or gains or losses for income tax purposes. The only amount so taken into account is the actual chargeable event gain, i.e. £1,351.35. That is a discreet amount produced from the calculation of gain "treated as arising in connection with" the policy; and that amount, as a stand-alone figure of income, is deemed by section 547(1)(a) to form part of Mr Drummond's total income.

The difference between amounts taken into account for purposes of prior calculations affecting a taxpayer and those taken into account as receipts in computing his income was recognized in the decision of Vinelott J in the decision in Hirsch v Crowthers Cloth Ltd 62 TC 759. In that case the taxpayer had sought to argue (unsuccessfully, as it turned out) that sums that had been brought into account to ascertain whether a writing down allowance or balancing charge which arose pursuant to the capital allowances legislation could be said to be taken into account for income tax purposes. Vinelott J held that the sums taken into account in relation to the capital allowances computation were too far removed from the income tax computation contemplated by what is now section 37(1) of TCGA. The link had to be "direct"; in the Crowthers' case it was nothing more than indirect. I refer to the decision of Vinelott J on page 771 where he gives examples of amounts brought directly into the computation of income for income tax purposes, such as a premium paid to a landlord on grant of the lease for less than fifty years (taxable under Schedule A or Case VI of Schedule D). Other situations that might have been in the forefront of the mind of the creators of section 37(1) are amounts taken into account in computing a taxpayer's liability to short term capital gains tax in respect of an asset disposed of after 6 April 1965. The framers of the 1968 Finance Act (which contained the provisions found in Chapter II of Part XIII of ICTA) were careful to avoid any confusion between the amounts brought into the computation of the gain to be treated as "arising in connection with the policy" and moneys brought into account "as a receipt in computing income or profits or gains … of the person making the disposal" for income tax purposes. Chapter II of Part XIII deliberately adopts a different form of words and this has the effect of requiring the prior calculation of an amount which is treated as income and deemed to be the income of the taxable person.

For those reasons I am against Mr Drummond on the section 37(1) issue. The surrender proceeds are not to be excluded from the consideration for the disposal of the policies.

Finally on the section 37(1) issue, I should add that if the provision applies at all in the present circumstances, then Mr Drummond is correct in claiming the amount of the chargeable event gain (£1,351.35, or his share of it) to be excluded.

The "wholly and exclusively" issues

So far I have approached this case without regard to the fact that all the transactions were carried out as part of what Mr Simon McKie and Mr Drummond (both of whom gave evidence) accepted to have been a tax avoidance strategy. The Revenue say that construing all the statutory provisions purposively and viewing them realistically, there was no resulting capital loss to be set against Mr Drummond's gains for 2000/2001; and in any event the amount of some £210,000 (being the difference between the £1.96 million that Mr Drummond was said to have incurred as expenditure and the £1.75 million encashment proceeds) is to be ignored as acquisition expenditure on the basis that it was not incurred wholly and exclusively for the acquisition of any asset.

The strategy in outline

I start with a brief description of the strategy. This requires Mr Drummond to acquire an interest in five policies of life assurance and to cause these to be surrendered. He acquires his interest by accepting an "Offer to Sell" by London & Oxford (described in the strategy as "the market maker") on 4 April 2001. The five policies that are the subject-matter of the Offer to Sell are part of a package of twenty that had been written on the life of Ms Sedgley in return for a small premium.

The creation and the further transactions relating to those policies are preparatory parts of the strategy which calls for London & Oxford to hold a portfolio of second-hand life assurance policies available for customers who have chargeable gains which they wish to shelter. London & Oxford, agrees to sell the requisite batch of policies to the customer. The amount payable by the customer is essentially the surrender value plus what Mr McKie described as a "required gross profit percentage"; the Offer to Sell says of that amount that part to be paid by the individual upfront on accepting the offer and part later. Whether the customer chooses to surrender is said (in the document explaining the strategy) to be up to him; but, he is advised, if he does, he will have a capital gains tax loss of an amount equal to the aggregate of the amounts payable by him to the market maker.

Each of the policies is written on a "good life". It is important to the strategy that the policy does not determine, on death for example, prior to the occasion when the individual purchaser gives instructions for its surrender so as to achieve the conclusion of the strategy. The premiums under the policy or policies are "invested" in a bond. The single premium is provided by way of loan to the life assured. The life assured then assigns the policy or policies to the market maker at a small mark-up over their surrender value. The market maker then pays £349,750 as an "additional premium" on the policy. They are then ready for the next stage of the strategy. A potential customer seeking tax shelter is introduced to Mr McKie, or to his company known as McKie & Co Ltd ("MCL"), and thence to the market maker.

Once the customer has been introduced to Mr McKie and has shown interest in the strategy, he is provided with what was referred to as independent financial advice. The customer then agrees to buy a number of the policies.

To avoid 4% stamp duty the policies are not assigned or transferred by the market maker to the individual. Instead, the seller (the market maker) makes a written offer to sell them to the customer. The offer is accepted by the act of payment of part of the total consideration.

Before he has paid in full and before the financial year has ended the customer instructs the market maker to surrender the policies. The market maker notifies the insurance company, before the end of the financial year, that it wishes to encash the policies. The insurance company pays the surrender moneys to the market maker and the market maker then deducts expenses and the outstanding balance of the consideration due to it and pays the rest to the customer.

The transactions in detail

Mr Simon McKie, director and controlling shareholder of MCL, formulated the strategy to enable individual customers with otherwise unrelieved capital gains to generate offsetable capital losses without suffering a corresponding economic loss. As this strategy related to second-hand life insurance policies, there had to be a stock of these available to meet the demands from customers. Simon McKie therefore approached Mr Paul Newton, an old friend who was chairman of London & Oxford. London & Oxford was a small corporate finance and investment company from which Mr McKie's practice rented its offices. The idea had been for London & Oxford to introduce a financial institution to act as stockholder for the second-hand policies to be provided to customers.

Simon McKie then set about identifying low risk policies. The Premier Access Bond issued by American Life Insurance Company ("AIG") was chosen. The surrender and maturity values of this type of policy are determined by reference to units of assets (i.e. units of interest-bearing deposits with financial institutions) in which AIG notionally invests the premium or premiums.

The strategy was to be marketed by MCL which would approach third party advisers (primarily tax professionals) and inform those advisers of the strategy in confidence.

In late 2000 Simon McKie drafted a detailed strategy description which gave numerical examples of how it might be implemented with an analysis of its tax effects.

By the end of 2000 Simon McKie and Mr Newton of London & Oxford had not secured funding from the financial institution as they had hoped; instead they obtained a funding facility from Lloyds TSB Bank ("Lloyds") on 12 January 2001. A letter of 22 February contains the formal offer.

Simon McKie had three meetings with KPMG (accountants). KPMG, Mr McKie said in evidence, adopted it as a preferred strategy for their clients (subject to technical approval) and signed a confidentiality agreement on 15 February 2001.

Creating the stock of life policies

Simon McKie and Mr Newton approached three individuals whom they identified as suitable life assureds. One was Simon McKie's partner (now his wife), one was her brother and the third was Ms Sedgley who worked for London & Oxford. Ms Sedgley was offered an interest-free loan by St James's Park Venture, a company in which Mr Newton had an interest. The loan was to be repayable on demand. Mr Newton explained to Ms Sedgley that the loan and her payments of the premium were part of the tax mitigation strategy. On 23 February Ms Sedgley paid the single premium of £250 on each of 20 AIG policies and they were effected with her as life assured.

Each of the potential life assureds was visited by an adviser on the staff of EFG Private Bank, said to be an independent financial adviser. EFG Private Bank had been introduced to the potential life assured by Simon McKie. The note of a meeting between Mr Andrew Britt of EFG Private Bank and Ms Sedgley records that he explained the bonds to her and that they discussed the advantages and disadvantages of buying them. Simon McKie said that EFG Private Bank had been introduced as part of the strategy. He was concerned to avoid the criticism that the purchase and capitalization of the policies formed part of a pre-ordained series. EFG Private Bank earned nothing for their advice at this stage. This, said Simon McKie was because they were aware of the strategy and they were aware that if London & Oxford purchased the policies from the individual life assureds they would still earn nothing for their advisory services. EFG Private Bank's commission (i.e. £5,000 in relation to the policies to which the Jason Drummond transaction related) would be earned when the customer agreed to buy them. On 21 February 2001 Ms Sedgley signed up to an "IFA agreement" with EFG Private Bank.

KPMG are activated

Simon McKie wrote to a Mr D Kilshaw of KPMG (on 23 February 2001) as follows:

"London & Oxford have acquired an AIG Premier Access Bond consisting of 20 policies and have arranged finance to capitalise these policies by paying additional premia of £5 million which they expect to do on Monday. I believe that gives us time to implement the strategy with one or two clients before Budget day.

Once the policies are capitalized I would ask you to consider whether you have one or two clients who are likely to benefit from acquiring £5.6 million of policies with the intention of sheltering gains of that amount before Budget day.

I would emphasize that no client should be identified before we have capitalised these particular policies and you should not review your clients' generally so you do not consider the suitability of any clients other than the ones whom you will advise to buy policies before Budget day.

Because of the very tight time constraints I am sending to you for your review documents which I list and comment on below."

An "independent financial adviser" assists Ms Sedgley

On 23 March 2001 Ms Sedgley made a telephone call to a Mr David Beck of EFG Private Bank to tell him that London & Oxford had offered to buy all 20 of her policies for £5,100. Mr Beck advised her that "from a purely commercial perspective it would make sense to accept the offer notwithstanding the small chargeable event that would be assessed on [her] following the sale". Those words are extracted from his note of the telephone conversation. Ms Sedgley and London & Oxford entered into a sale agreement on 26 March for immediate completion; the same day she assigned all 20 policies to London & Oxford.

Topping up the policies

On 28 March 2001 London & Oxford charged the policies (i.e. those purchased from Ms Sedgley and from the other two life assureds referred to above) as security for London & Oxford's overdrawings with Lloyds.

On 30 March 2001 London & Oxford drew down on its overdraft facility (then standing at £7 million) and paid £6,995,000 to AIG to top-up the 20 policies on Ms Sedgley's life.

Jason Drummond comes on the scene

Simon McKie arrived at his office on 3 April 2001 and found a message from Mr Kilshaw of KPMG. Apparently Mr Kilshaw was having a meeting with Mr Drummond. A meeting was fixed with Mr McKie for 4 April.

Mr Jason Drummond is a man of means. He had realised a gain of £4.8 million earlier in the financial year on sale of shares in a company in which he still had a large holding and that holding was, he said, "just one of my assets". He had, he explained, asked KPMG to be more proactive in his tax affairs and had been having conversations with them about different schemes or strategies that could be used to offset his tax.

The 4 April meeting

Preparatory to the meeting some steps were taken:

(i) London & Oxford set up with Lloyds a "client account" into which, Mr Newton told Lloyds in a fax, Mr Drummond was to pay £1,155,000 "this morning". The fax went on to say that Mr Drummond "then proposes to transfer these funds to London & Oxford this afternoon".

(ii) £1 million was transferred from Mr Jason Drummond's current account with Barclays Bank to the Lloyds current account.

(iii) Mr Drummond agreed with KPMG to engage KPMG to give advice on the strategy "involving the acquisition and subsequent encashment of second-hand endowment bonds" and to be introduced to MCL. KPMG also agreed to assist Mr Drummond with the reporting of the transaction to the Inland Revenue. KPMG's fee was to be "1% of the value of the bonds you purchase".

The meeting was attended by Mr Drummond, Mr Kilshaw and a Mr Jennings for KPMG, Mr Beck from EFG Private Bank (as independent financial adviser), Mr Newton and Simon McKie. Mr Drummond and MCL entered into an "advisory agreement". Simon McKie gave advice orally. This was based on a written summary that he had prepared in advance. That written summary was contained in a letter from MCL to Mr Drummond. The MCL letter, signed by Simon McKie, recited how Mr Drummond had been told by KPMG that Simon McKie "was able to provide tax advice on a strategy which would generate a capital loss of £1,962,233 so as to reduce by that amount ("the Targeted Gain") the Chargeable Gain which you have made".

The following are extracts from the part of the letter headed "The Mitigation Strategy":

"Providing tax legislation and other relevant circumstances have not changed in the intervening period, at some time after this purchase you will ask your investment adviser for their advice on assigning or surrendering the policies. In order to realise a capital loss in accordance with the Mitigation Strategy, it is likely that they will recommend a surrender of the policies. When the policies are surrendered there will be a "chargeable event" under the Income Tax Chargeable Events Legislation. It is likely that you will make a chargeable event gain on the surrender which will be subject to income tax. That chargeable event gain is unlikely to be large because it will be calculated as the difference between the surrender value of the policies and the premia previously paid under the policies. Because it is likely that at the time of the surrender the policies will only have been in existence for a short period of time, that difference is likely to be small.

The surrender of the policies will also be a disposal for capital gains tax … . Provided the Mitigation Strategy is successful, for capital gains tax purposes your disposal proceeds will be the surrender proceeds less any amount taken into account in calculating your chargeable event gain. Because the whole of your surrender proceeds will have been taken into account in calculating the chargeable event gain, your disposal consideration for capital gains tax purposes will be reduced to nil. The expenditure deductible in calculating a capital gain or loss on disposal will be the amount for which you have purchased the policies. Because that amount is not taken into account in calculating a chargeable event gain it will not be excluded from the capital gains computation. For that reason you will make a capital loss equal to your expenditure on the policies.

There is a market in second-hand policies but, I understand, that it is highly unlikely one would be able to identify suitable policies to shelter a gain of the magnitude of the targeted gain in the general market. For that reason London & Oxford has acquired policies of the requisite characteristics and paid additional premia into those policies. The policies are assigned to Lloyds Bank as security for a loan. The security will be released when you purchase these policies. London & Oxford makes a market in these policies."

Simon McKie took Mr Drummond through the figures explaining how he had quantified (to quote from his evidence) "the net benefit of the strategy to [Mr Drummond] after taking into account the investment loss which would arise on his purchase and acquisition of his policies on the assumption that he surrendered the policies rather than retaining them and making a sale to a third party." Simon McKie also explained to Mr Drummond what Mr Drummond's financial loss would be in the event if the strategy were unsuccessful. The written summary under the heading "The Mitigation Strategy" states, under the heading "costs":

"Thus the gross benefit of the strategy would be £784,893 and this is reduced to a net benefit of £573,840 by the costs of the strategy. If the strategy were to fail you would have lost £67,869 after allowing for tax relief on the excess of the purchase price of the policies over your estimated proceeds. (You should note that this does not include the cost of establishing the effectiveness of the scheme. These costs would be chargeable against the fighting fund but it is not guaranteed that that fund will be sufficient to meet all costs.)"

The written summary explained how, according to the strategy, a small chargeable event gain (the difference between the surrender value of the policies and the aggregate amount of the premiums paid on them) would have accrued.

The "independent financial adviser" participates again

When Simon McKie had completed his explanation of the procedure involved in the strategy and of the figures, he and Mr Newton left the room and Mr Beck of EFG Private Bank, independent financial adviser, provided his advice to Mr Drummond. A note of Mr Beck's advice records that the advice was about Mr Drummond's purchase of an AIG premier access bond. It goes on to say:

"We looked at the current interest rate being offered on the bond and saw that deposits in excess of £1 million currently attract an equivalent gross rate for a higher rate taxpayer (such as yourself) amounting to 6.88%. This compares with it favourably with the leading bank and building society rates currently being offered.

No tax advice was given in addition to the above. In particular, no advice was given with regard to the tax implications of the fact that I understand that you will be purchasing an assigned policy from London & Oxford rather than making an investment direct with AIG life. This course of action, and indeed any further action undertaken in respect of the bond, will be on the advice of taxation advice given to you by Simon McKie of McKie & Co."

Stamp duty

The calculations relating to the strategy prepared for Mr Drummond contain no amounts for stamp duty. This is the explanation in the advisory letter from Mr McKie to Mr Drummond of 4 April 2001:

"Under the strategy, you will purchase an interest in the policies but you will not become the registered owner of the policies with the insurance company. If you were to do so you would pay stamp duty on the purchase of 4%. Instead, the vendor will make a written offer to you to sell the policies and specify that the offer may be accepted by making a payment of the requisite amount to the vendor. It will not go on to complete an assignment to you so that after you accept the contract by making the payment, the vendor will hold the policies to your order."

Analysis of difference between Mr Drummond's "buying costs" and his selling proceeds

The Estimated Results of the Tax Mitigation Strategy prepared by MCL for Mr Drummond contain an entry headed "Analysis of difference between buying and selling costs". The difference is £211,034. In evidence Simon McKie said of the £211,034 that "essentially it is the cost to Mr Drummond of the transaction". It was also, he said, "in effect London & Oxford's gross profit from the transaction." In the analysis the £211,034 is made up of five component parts:

(i) "Fixed costs" of 4% of £1,962,233 (the aggregate of the two amounts referred to in the Third Schedule to the Offer to Sell – see paragraph 51 below). These work out at £78,489.

(ii) "Fighting fund" at ½%, i.e. £9,811.

(iii) "Contingent costs" at 5%, i.e. £98,112. (An escrow account for this has retained this amount.)

(iv) "Client adviser's cost" at 1%, i.e. £19,622. That I understand to have been KPMG's commission for introducing Mr Drummond to the strategy.

(v) "IFA's costs" are £5,000.

Mr Drummond decides to go ahead

Mr Drummond went ahead with the strategy. Asked whether he accepted that it was a tax avoidance scheme and nothing else he said – "It was a strategy to reduce my tax for that period." He indicated that he understood the costs of the strategy whether successful or unsuccessful. He acknowledged that the strategy involved his undertaking to pay out £1.96 million and being entitled to get back £1.75 million. He observed:

"… my main focus is what it would cost me if it did not work out and that was £67,000 and what was the gain if it succeeded, which was obviously over £500,000; so that was very much the way I looked at it."

In response to the suggestion that he had not been "truly looking for a suitable home for his spare cash", Mr Drummond said that his intention had been to try and offset some of his gain for that year. He had not looked at what he described as the "headline numbers" (i.e. £1.96 million cost and a recovery of £1.75 million): he said that his concern was "with the amount he could offset for tax and the costs if unsuccessful".

The steps by which Mr Drummond buys into the strategy

Mr Newton provided Mr Drummond with a written offer. Part of the document reads as follows:

"London & Oxford offers to sell the policies … to [Mr Drummond] for a consideration consisting of the Consideration Payments …

Acceptance of the offer to sell shall be effected by [Mr Drummond] paying to London & Oxford an amount equal to [£1 million] on or before 4 April 2001 on which occasion a contract for the sale and purchase of the policies shall be deemed to have come into effect and to have been completed at that time. It shall be a term of the contract that [Mr Drummond] will [pay £962,233] on [5 April 2001].

The said sale shall take effect free of all encumbrances."

The Third Schedule to the Offer to Sell defines the "Consideration Payments as:

"A payment of £1,000.000 made to London and Oxford on 4 April 2001

A payment … of £962,233 made to London and Oxford on 5 April 2001."

Mr Newton, acting on Mr Drummond's behalf, faxed Lloyds (on 4 April) instructing them to transfer £1 million from Mr Drummond's client account held by London & Oxford to London & Oxford's bank account. The same fax informed Lloyds that "We may wish to surrender the Premier Access Bonds".

Mr Drummond's position overnight

At the closing of the day Mr Drummond was left with two possibilities.

First he could abandon the strategy, pay the further £962,233 due on 5 April, require the policies to be assigned to him by London & Oxford and bear the further 4% stamp duty chargeable on the assignment (i.e. some £80,000). Total loss to him by the end of 5 April would therefore be £290,000 which included the fixed costs of £78,489, KPMG's introductory commission of £19,622 and the IFA's fee of £5,000. Abandoning the strategy and letting the Offer to Sell take its course would result in Mr Drummond being at least £290,000 out of pocket.

The other possibility was that Mr Drummond could go ahead with the strategy and improve his financial position by £574,000 if it succeeded or lose £68,000 if it were to fail.

The next day (5 April 2001): the "independent financial adviser" calls again and the strategy is completed

This was the last day of the financial year. Unless Mr Drummond activated the strategy so that the policies were surrendered, there would be no losses to set off against the gain he had made earlier in the year. Instead he would, as just observed, have thrown away some £290,000.

In evidence Mr Drummond could not suggest anything, apart from an announcement of a change in taxation legislation, that might have happened since he left the meeting in the afternoon the day before that would have prevented him from pursuing the strategy and activating the surrender of the policies. As it happened Mr Beck of EFG Private Bank, independent financial advisers, gave Mr Drummond a telephone call on 5 April and advised him to surrender the policies. It will be recalled that the Mitigation Strategy says of the "investment adviser" that "it is likely that they will recommend a surrender of the policies". Mr Drummond, in evidence, could not recall the conversation with Mr Beck because it was "quite a long time ago". At all events the strategy was fulfilled and Mr Drummond went ahead and telephoned Mr Newton of London & Oxford. Mr Newton delivered a withdrawal form in respect of the policies to AIG. Lloyds, to whom the policies had been charged by London & Oxford, notified AIG that they were happy for AIG to act on the instructions of Mr Newton to surrender the policies and to release the charge once the proceeds were received. Lloyds then notified AIG of the release of the charge.

The same day, according to a letter from London & Oxford to Lloyds, instructions were given to pay £98,112 into the contingency (escrow) account and to pay £789,143 to Mr Drummond on receipt of the surrender funds from AIG. (The surrender proceeds of the policies to which the appeal relates came to £1,751,376 and this amount was in London & Oxford's account on 6 April 2001.)

The Chargeable Event Certificate

AIG issued the chargeable event certificate in respect of the surrender on 5 April 2001 as required by section 552(1) of ICTA. This identifies London & Oxford as the policy owner and specifies the single premiums invested (£250 for each policy) and the additional premiums (£349,750 for each policy). The previous chargeable event gain of 89p per policy (i.e. the gains deemed to form part of Ms Sedgley's total income) is shown. A chargeable event gain of £274 per policy is certified in relation to the surrender. The certificate ends with the words – "This policy gain represents income and should, therefore, be included in your tax return for the year …".

I have no information about whether London & Oxford received a separate chargeable event certificate in relation to their own transfer of beneficial interest in the policies to Mr Drummond on 4 April 2001.

The "wholly and exclusively issues"

Section 38(1) of TCGA provides, so far as is relevant, that:

"(1) Except as otherwise expressly provided, the sums allowable as a deduction from the consideration in the computation of the gain accruing to a person on the disposal of an asset shall be restricted to –

(a) the amount or value of the consideration, in money or money's worth, given by him or on his behalf wholly and exclusively for the acquisition of the asset … ,

(b) the amount of any expenditure wholly and exclusively incurred on the asset …".

The case for the Revenue on the "wider wholly and exclusively issue" is that no part of the £1.96 million, described as "the Consideration Payments" in the Offer to Sell of 4 April was consideration in money given by Mr Drummond wholly and exclusively for the acquisition of the asset, i.e. the five policies. To the extent that he paid out money in response to the Offer to Sell, his motivation had been to go through the steps of the tax avoidance scheme; the expenditure could not therefore be regarded as wholly and exclusively for the acquisition of the asset. On the "£210,000 wholly and exclusively issue", the case for the Revenue was that this amount was incurred as the cost of the scheme and not for or on the asset. Mr Drummond disagreed. A single consideration of £1.96 million was incurred; without incurring that amount he did not and could not acquire the assets.

The starting point in the analysis is to note that there was no purpose in London & Oxford's sale and in Mr Drummond's purchase of the rights to the five policies other than the facilitation of the tax avoidance strategy. This is the clear message from the documentation and from the oral evidence. The "strategy" depended on Mr Drummond acquiring those rights for £1.96 million and causing them to be turned into cash by the end of 5 April 2001. From Mr Drummond's point of view the acquisition of the five policies as an investment would, for reasons given below, have been unthinkable; and to have completed the terms of the Offer to Sell and to have obtained an assignment of the five policies with all the attendant loss and additional expense would have defied reality. The Offer to Sell does not tell the whole story. What therefore was the relevant transaction for tax purposes? The Ramsay approach as that was explained in Barclays Mercantile Bank Finance v Mawson ("BMBF") 76 TC 448 and [2005] STC 1 at paragraphs [32] and [36] is relevant in answering this question.

The approach, as prescribed in BMBF in paragraph [36], is:

"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."

I start with the first step of the BMBF approach. Transactions that answer the statutory description "consideration … given … wholly and exclusively for the acquisition of the asset" in section 38(1) of TCGA possess two essential characteristics. The whole of the consideration must be for the acquisition of the asset. And the acquisition of the particular asset must be the sole purpose for which the consideration is given. See the judgment of Romer LJ in Bentleys Stokes & Lowless v Beeson 33 TC 491 at 503. The expression "for the acquisition of the asset" calls for an analysis of what the person in question gets in return for the consideration.

The second step in the BMBF approach is to decide whether the particular transaction answers the statutory description, adopting "an unblinkered approach to the analysis of the facts": see paragraph 36 of BMBF and the quotation from the judgment of Ribeiro PJ in Collector of Stamp Revenue v Arrowtown Assets [2003] HK CFA 46 [35].

To the extent that the strategy was designed to avoid the conclusion that the steps were preordained in the Furniss v Dawson [1984] STC 153, [1984] AC 474, [55] TC 324 sense (a point I shall come to when dealing with the role of EFG Private Bank) the following passage from the decision of the House of Lords in IRC v Scottish Provident Institution 76 TC 538, [2005] STC 15 at [23] is in point:

"We think that it would destroy the value of the Ramsay principle of construing provisions … 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."

What then does an unblinkered approach reveal as the intended and expected transaction? The only scheme document was the Offer to Sell. But the common objective of all the participators was to generate a CGT loss by exploiting the statutory mismatch between the CGT and the "chargeable event gain" provisions. This follows from the findings of fact to which I now turn.

The purpose and effect of the strategy was to achieve the purported CGT loss without Mr Drummond, win or lose, incurring more than the £210,000 difference between the so-called acquisition cost of £1.96 million and the so-called sale proceeds of £1.75 million. To achieve this result Mr Drummond had to notify London & Oxford on 5 April 2001 of his instructions to surrender the policies in time for London & Oxford to take the necessary encashment steps with AIG. The strategy did not work if all the steps envisaged in the Offer to Sell were taken. Besides, there were no arrangements to provide the £962,233 due on 5 April and the strategy was costed on the basis that no stamp duty (some £80,000) would become payable. London & Oxford were never going to transfer the policies to Mr Drummond. All the fees borne by Mr Drummond were calculated on the basis that the strategy would be carried through. Those features indicate the true intention of Mr Drummond and London & Oxford which was not to allow the Offer to Sell to run its course.

EFG Private Bank and its employees who "advised" Ms Sedgley and Mr Drummond were, in my view, acting out a charade for which EFG Private Bank were paid a single fee of £5,000 by London & Oxford. Ms Sedgley had been introduced to EFG Private Bank by Simon McKie and Mr Newton. At that stage EFG Private Bank was paid nothing. EFG Private Bank's commission was to come. As Simon McKie explained, EFG Private Bank was "aware of the strategy" and they would earn their commission if London & Oxford "were able to go on and sell the policies to clients under the strategy". And once the clients were found, EFG Private Bank's interest was in advancing the strategy. EFG Private Bank's participation in the strategy was window-dressing. It in no way evidenced any freely made decisions by Ms Sedgley or Mr Drummond. There was in my view no realistic likelihood of either Ms Sedgley or Mr Drummond failing to carry out the steps required by the strategy. Its part had been written for it and inserted in the strategy to include "a commercially irrelevant contingency, creating an acceptable risk that the scheme might not work as planned": see the passage from IRC v Scottish Provident Institution set out above.

Next, it is relevant to observe that the strategy could only achieve its tax avoidance objective if the rights to the five policies acquired by Mr Drummond on 4 April were turned into cash on 5 April. There was no other practicable course. To have completed the process and got the policies transferred into Mr Drummond's name would, as already indicated, have required him to find the second payment of cash of £962,233 on 5 April plus (according to The Mitigation Strategy, see paragraph 47 above) stamp duty of some £80,000; and Mr Drummond would then have foregone the hoped-for benefit of £575,000 of tax shelter. Thus, for an outlay of £1.96 million plus stamp duty of £80,000, he would have been left with five policies whose surrender value was £1.75 million. Furthermore, had Mr Drummond really been interested in obtaining AIG Premier Access Bonds worth £1.75 million, he could (I infer) have obtained them for that amount any day direct from AIG. Those features call into question two matters. First, was the £1.96 million really consideration for the acquisition of the five policies; or was it, adopting the "unblinkered analysis of the facts", the nominal price of the tax shelter? Second, was the £1.96 million incurred "wholly" for the acquisition of the five policies, or was £210,000 of it (including the contingency payment held in the escrow account) for participating in the strategy.

The five policies were in every respect real. The insurance company was a major institution. The underlying investments were genuine and potentially long-term and the rights of the policyholders were in all respects of an arms' length nature. But each policy was effected, topped up and encashed in preparation for the strategy. Encashment at the instance of Mr Drummond, as London & Oxford's customer, by the end of the tax year 2000/2001 was an essential step if the scheme was to achieve its purpose. That, as I have already indicated, was the express understanding of the parties (see the summary of the evidence and, in particular, the extracts from "The Mitigation Strategy" set out above). The Offer to Sell of 4 April 2001 was used to enable that result to be achieved. The rights to the policies were vested in Mr Drummond without the delay that might have been occasioned by going through the steps of formal assignments and without the 4% stamp duty charge. The only real function of the independent financial adviser on 5 April must have been to make sure that Mr Drummond did not oversleep and miss the one act that he had to perform by the year end, namely to call Mr Newton and tell him to direct the surrender of the policies.

Mr Drummond did not participate in the strategy until 4 April 2001. The issue of whether the £1.96 million was incurred wholly and exclusively for the acquisition of the five policies can be expressed in terms of this question: were the Consideration Payments of £1 million on 4 April and £962,233 (which it was a term of the contract should be paid on 5 April 2001), viewed realistically and taking an unblinkered BMBF approach to the facts, the consideration given wholly and exclusively for Mr Drummond's acquisition and London & Oxford's disposal of the five policies? The answer in my view is – No. The Offer to Sell does not stand alone. The scheme was designed to achieve the tax shelter, not Mr Drummond's acquisition of the five policies. To achieve the tax shelter by the end of 2000/2001 (and without incurring stamp duty) Mr Drummond's direction to encash or surrender the policies was the essential step and the only one that he had to take of his own initiative. The alternative, of keeping the policies, was (as I have explained) unthinkable. For Mr Drummond to have paid for, taken transfer of and kept the policies would have defied the expectations and intentions of everyone involved. I am therefore against Mr Drummond on the wider section 38(1) issue.

If I were wrong on that, it could, arguably, follow that Mr Drummond had for CGT purposes made an acquisition of the five policies. This leaves the question whether the whole £1.96 million was consideration given wholly and exclusively for their (assumed) acquisition or whether £210,000 should be left out of account as being the cost of the shelter. In the circumstances summarised above it would, I think, be unreal to view the transaction as one in which Mr Drummond acquired assets known to have a value of £1.75 million for £1.96 million. There was no evidence that he wanted to acquire and hold the policies. The entire weight of the evidence was to the contrary. As Mr Drummond put it (see paragraph 50 above) his concern was with the amount he could offset for tax and the costs if unsuccessful. The only possible inference, viewing the transactions realistically, is that the £210,000 was not incurred "exclusively" (let alone "wholly") for the acquisition of the five policies. It was in reality money spent for the services of Simon McKie, London & Oxford and KPMG. I am therefore against Mr Drummond on "the £210,000 wholly and exclusively issue".

It follows from the conclusions reached above that, because I am against Mr Drummond on the section 37(1) issue and on the wider wholly and exclusively issue, my conclusions are immaterial to and incompatible with the final argument of the Revenue set out in paragraph 6(iv) above.

For all those reasons I dismiss the appeal.

SIR STEPHEN OLIVER QC

SPECIAL COMMISSIONER

RELEASED: 5 July 2007

SC 3158/2006