Centrica Overseas Holdings Limited v The Commissioners for HM Revenue and Customs [2021] UKUT 0200 (TCC) [2021] UKUT 0200 (TCC)
[1]This is an appeal against a decision of the First-tier Tribunal (“the FTT”) released on 23 April 2020 (“the Decision”). The FTT dismissed the appellant’s appeal against a decision of the respondents that it was not entitled to claim relief for corporation tax purposes for certain expenditure which it claimed was expenses of management of its investment business. We shall refer to the appellant as “COHL” and the respondents as “HMRC”. The FTT gave permission to appeal in a decision dated 22 July 2020.[2]COHL is an intermediate holding company in the Centrica Plc group of companies. At all material times it owned 100% of the share capital of Oxxio BV (“Oxxio”), a Dutch holding company which itself had four subsidiaries registered in the Netherlands. Assets of two of those subsidiaries and the shares in a third subsidiary were sold to a company called Eneco Group NV (“Eneco”) in a deal which was completed in March 2011. Professional fees were paid in connection with the transaction relating to services from Deutsche Bank AG London (“Deutsche Bank”), Pricewaterhouse Coopers (“PwC”) and De Brauw Blackstone Westbroek (“De Brauw”) over the period from July 2009 to March 2011. The fees which COHL claims are expenses of management of its investment business totalled some £2.5m (“the Disputed Expenditure”) and COHL claimed relief in its company tax return for the period ending 31 December 2011.[3]COHL’s claim for relief was made pursuant to s1219 Corporation Tax Act 2009 (“CTA 2009”) which provides so far as relevant as follows:(1) In calculating the corporation tax to which a company with investment business is liable for an accounting period, expenses of management of the company's investment business which are referable to that period are allowed as a deduction from the company's total profits. (1A) …(2) For the purposes of this section expenses of management are expenses of management of a company's investment business so far as — (a) they are in respect of so much of the company's investment business as consists of making investments, and (b) the investments concerned are not held for an unallowable purpose during the accounting period to which the expenses are referable.(b) … 2
[4]A company with investment business is defined by s 1218(1) CTA 2009 as a company whose business consists wholly or partly of making investments. It thus encompasses holding companies and intermediate holding companies.[5]There are certain issues between the parties as to exactly what the FTT decided in relation to various issues which we shall address in detail below. Overall, the FTTJudge dismissed the appeal in the following terms:
[235]I have decided that COHL did not itself carry out the management activities in relation to which the disputed expenditure was incurred. The conditions for relief in section 1219 are not therefore satisfied as the expenditure was not expenses of management of COHL’s investment business.[6]The FTT’s finding, so stated, was sufficient to determine the appeal against COHL but in case it was wrong on that the FTT properly considered other issues which had been raised by HMRC to justify their decision to refuse relief. The FTT dealt with those issues as follows:(1) It rejected an argument of HMRC that the fees of Deutsche Bank and PwC were not expenses of management but were expenses of the disposal of Oxxio assets. However, it appeared to accept that argument in relation to some of the fees of De Brauw.(2) It rejected an argument of HMRC that the Deutsche Bank fees could not be expenses of management because they amounted to a “success fee” and for that reason must be treated as an expense of the disposal of Oxxio assets.(3) It rejected an argument of HMRC that the fees of Deutsche Bank and PwC were capital expenditure such that s1219(3)(a) prevented relief. However, it accepted that argument in relation to all the fees of De Brauw. THE GROUNDS OF APPEAL[7]The FTT granted COHL permission to appeal on each of three grounds on which it sought permission. HMRC have also served a response pursuant to Rule 24 of the Tribunal Procedure (Upper Tribunal) Rules 2008, relying on grounds on which they were partly or wholly unsuccessful before the FTT.[8]In relation to the FTT’s finding at [235], COHL argues that the FTT determined its appeal on the basis that the management referred to in s1219 CTA 2009 must be undertaken by the company whose investment business it is, and that management must be undertaken by the directors acting formally in their capacity as such. COHL argues that the FTT wrongly found that the decisions of COHL were not taken by the directors of COHL acting as such and so s1219(1) was not satisfied.[9]HMRC acknowledge that they did not make any submission to the FTT that the only people who can manage a company’s investment business are the directors of the company. However, they contend that the FTT’s decision is properly to be read as a decision that COHL had not established evidentially that the Disputed Expenditure constituted expenses of management of COHL’s investment business. In particular, COHL had not established a sufficient link between the disputed expenditure and any investment business of COHL.[10]Subject to that, the parties were largely agreed as to the issues of law which arise on this appeal. We can therefore characterise the issues arising on this appeal as follows:(1) What was the basis on which the FTT determined that the Disputed Expenditure was not expenses of management of COHL’s investment business, and was it entitled to do so as a matter of law? (“Issue 1”) 3 If we find in favour of HMRC on this issue, then COHL’s appeal must be dismissed. If we find in favour of COHL on this issue, then the following issues arise:(2) Did the Disputed Expenditure amount to expenses of management, or was it (or some of it) expenses of the disposal of Oxxio assets? (“Issue 2”)(3) Could the Deutsche Bank fees amount to expenses of management if they were in the nature of success fees? (“Issue 3”)(4) Was the Disputed Expenditure capital in nature such that it fell to be disallowed by virtue of s1219(3)(a)? (“Issue 4”)[11]There are various authorities as to the distinction between the expenses of management of an investment business and the costs of acquisition or disposal of an asset. We consider those authorities in detail below. The FTT made detailed findings of fact in relation to the sale process and the involvement of Deutsche Bank, PwC and De Brauw in that process. THE FTT’S FINDINGS OF FACT[12]The FTT’s decision runs to 82 pages and 441 paragraphs. It contains a detailed analysis of the evidence, its findings of fact based on that evidence and a discussion of the issues. For present purposes we can summarise the FTT’s findings of fact by reference to the headings used by the FTT, namely:(1) Background facts;(2) Decision making in the Centrica group;(3) Services provided by Deutsche Bank, PwC and De Brauw; and(4) The transaction timeline. Background facts[13]COHL is an investment holding company. It is a subsidiary of GB Gas Holdings Limited, which is itself a subsidiary of Centrica Plc.[14]The Centrica Group is well known as a major participant in the energy market. It operates in the UK, North America and elsewhere and its activities include energy supply, gas and oil exploration, development and production and the operation of power stations in the UK. It is the biggest supplier of energy to the UK domestic market.[15]COHL was and is a substantial intermediate holding company holding the Group’s investments outside the USA. As at 31 December 2009, the book value of assets in COHL’s balance sheet was £1,264,917,000. Its investments included shares in and loans to companies in the UK, Spain, Canada, Norway, India, Germany and other European countries including the Netherlands.[16]COHL had acquired Oxxio on 1 July 2005 for some £88m. The investment was not a success. From the time of its acquisition, the group generated losses on a year-by-year basis. In 2008 alone, it made a loss of over €100m. COHL provided substantial financial support to Oxxio through a revolving loan facility and injections of equity finance.[17]Oxxio’s four Dutch subsidiaries were Oxxio Nederland BV (“Oxxio Nederland”), Centrica Energy Netherland BV (“CEN”), Oxxio Tolling BV (“Oxxio Tolling”) and Oxxio Metering BV (“Oxxio Metering”).[18]Oxxio Nederland had gas and power purchasing contracts with 180 employees and some 620,000 residential and business customers. Oxxio Tolling had a long-term tolling agreement relating to a power plant at Rijnmond in Holland. Oxxio Metering had a smart meter energy 4 business. CEN had a supply agreement involving the supply of gas and electricity to some 140,000 residential customers.[19]In 2009 Centrica was reviewing its involvement in the European energy market generally. The future of the Oxxio businesses were considered as part of this general review. Centrica’s annual report and accounts for 2009 included the following statements: In June 2009…management approved and initiated a plan to sell Oxxio BV in the Netherlands. … It is anticipated that the sale of Oxxio…will complete by 30 June 2010. Oxxio…was classified as a discontinued operation from 30 June 2009.[20]A meeting was held on 10 July 2009 at which the disposal of the Oxxio Group was discussed. The objective was to achieve a “going concern sale” of the Oxxio Group and the project was codenamed “Project Erasmus”. The agenda for the meeting also included consideration of different possible transaction structures. Representatives of Deutsche Bank, PwC and De Brauw attended the meeting although they had not yet been formally engaged at that time.[21]A Centrica Plc board minute dated 28 July 2009 recorded that Oxxio and two other European businesses “would be treated as discontinued businesses held for sale”. The FTT records as follows:[23]“Held for sale” is an accounting term of art and is defined by IFRS5. A “disposal group” is to be classified as “held for sale” if its value is to be recovered principally through a sale transaction. The disposal group must be “available for immediate sale in its current condition” and a sale must be “highly probable”. For this to apply, management must be committed to a plan to sell the asset and must have initiated an active programme to locate a buyer and complete the plan. It must be actively marketed for sale at a price which is reasonable in relation to its current fair value. It should be expected that the sale will complete within a year. The FTT found at [24] as follows:[24]Clearly, the board of Centrica had decided, in June 2009, that it wanted to sell the Oxxio business and it was taking steps to do so. The sale process proved to be difficult and lengthy. The FTT found at [218] that the initial intention was for COHL to sell the shares in Oxxio. However, from an early stage the Centrica group was considering other ways of realising value from the Oxxio business. Options included selling the assets of the business, rectifying problems in the business with a view to a future sale of shares and not selling the business at all but winding it down. Disposal of the business was not something to be done at any price.[25]The eventual disposal in March 2011 was not the straightforward sale of shares in Oxxio which the Centrica group had initially hoped it would be.[26]The problem for Centrica was that a prospective purchaser would not necessarily want to buy all of the businesses together. Also, there were problems with management and financial controls which made a share purchase unattractive from the buyer’s point of view. Oxxio had been making losses for a number of years and the scale of the loss in 2008 was unexpectedly large. The sale was therefore difficult.[27]During the sale process, Centrica gained a greater understanding of problems in the Oxxio group. There had been accounting irregularities and poor risk management practices. This meant that a share sale began to look more difficult and it looked like an asset sale would be more attractive to a purchaser. 5[28]Eneco first made an indicative offer in September 2010 which was rejected. There may have been another offer from Eneco in December 2010 which was also rejected. Eneco made a final offer in January 2011 which the board of Centrica Plc approved in principle at a meeting on 22 February 2011.[29]In the event, it was not possible to sell the whole group and the eventual transaction involved an asset sale involving partial de-mergers. Effectively the assets of Oxxio Nederland and CEN were demerged and purchased by Eneco. Eneco also acquired the entire shareholding in Oxxio Metering from Oxxio.[30]Following these transactions, COHL continued to own Oxxio and to provide financial support to Oxxio Tolling which remained part of the Centrica Group. COHL received value from the transaction not through the sale of Oxxio shares, but by way of repayment of loans from the proceeds of sale.[31]Deutsche Bank, PwC and De Brauw were involved throughout the process. Each firm was engaged by Centrica Plc to provide certain services. In the first instance, the fees were paid by Centrica Plc as COHL did not have a bank account. This was usual practice within the group and costs incurred in this way would then be charged to what was considered the appropriate entity by means of book entries. The Disputed Expenditure was included as an accrual in the financial statements of COHL for the period ended 31 December 2011. It was accepted by HMRC that COHL had borne the cost of the fees. The fees paid were as follows: PwC - £172,423 De Brauw - €766,328 Deutsche Bank - €3,550,515.32[32]Deutsche Bank’s fees consisted of a fixed fee of €2.5m which was to be payable “in the event the Oxxio Transaction [a defined term] is completed” and “at the time of completion”. An “additional incentivisation fee” was payable to Deutsche Bank in Centrica’s sole discretion, in the event that a transaction was completed. Deutsche Bank was paid an incentivisation fee of €1m, but COHL did not claim any deduction for that fee. This structure was typical for this type of transaction and gave Deutsche Bank an incentive to get the deal over the line. Decision making in the Centrica group[33]Strategic decisions within the group were made by Centrica Plc, which had various central teams providing advice and support for the group. The teams covered areas such as legal, tax, accountancy, regulatory matters and mergers and acquisitions. Many subsidiaries such as COHL had no employees of their own. The directors of COHL occupied senior positions within this centralised group structure. The directors of COHL included the group Head of Tax and key members of the group financing team and they took what were described as “operational decisions”. At the point where a formal decision had to be taken, a COHL board minute would be produced, although the FTT said that it was unclear what decisions merited a formal board decision. The evidence included three formal board minutes of COHL:(1) A board minute dated 19 March 2009 which stated that after due and careful consideration the board of COHL had resolved that it would inject €110m of equity capital into Oxxio. Funds came from the group treasury function.(2) A board minute dated 30 September 2009 which stated that after due and careful consideration it was resolved that a loan should be made by COHL to Oxxio Tolling to provide working capital on the terms of a draft loan agreement prepared by Centrica Group Treasury and Tax departments. 6(3) A board minute dated 1 July 2010 which recorded that Oxxio had asked COHL for a replacement and additional loan facility of €250m which had been reviewed and approved by the Group Legal and Group Treasury departments. After due and careful consideration, the board resolved that COHL should enter into the agreement.[34]A board minute for Centrica Plc dated 7 May 2010 stated that “the sale process for Oxxio had been put on hold to allow the local team to focus on the work required for completing the 2008 and 2009 statutory audits”. There were no board minutes of COHL referencing the decision to sell Oxxio or the progress of the transactions. However, at [39] the FTT found as follows:[39]There does not seem to have been any clear delineation between the various roles in which the people who were the directors of COHL were acting. They were all senior people carrying out group roles, they worked together, many in the same building and they would receive information and discuss matters relevant to COHL in their group capacities. There was, accordingly, no need to brief the directors of COHL or inform them how the transaction was progressing as they (as directors of COHL) would already know these things in their respective group capacities. Services provided[35]The Centrica group routinely procured services for entities within the group. In the case of Oxxio, the advisers were engaged by Centrica Plc, and in the first instance Centrica Plc paid for the services. The services were then re-charged to COHL on the basis that it was considered to be the “main beneficiary” of those services.[36]Deutsche Bank’s role was to advise, provide information and make recommendations in relation to the disposal of the Oxxio businesses. This included wide-ranging advice and assistance in relation to strategic alternatives for the various businesses, for example the possibilities of share or asset sales and asset swaps, structuring, negotiating, planning and managing the disposal process and identifying and evaluating potential purchasers. The FTT found as follows at [71]:[71]An important part of Deutsche Bank’s role was to evaluate the options available in relation to Oxxio. The strategic decision to divest had already been taken, but the options as to how to achieve this were wider than in normal transactions because of the diversity of the businesses within the Oxxio sub-group and the problems which emerged in the finances and management of the main customer business. It was not clear what appetite the potential purchasers in the market might have for the different elements in the business. There was even the possibility that the business would prove unsellable and would have to be closed down.[37]By June 2010, Eneco had been identified as a potential purchaser and by October 2010 a virtual data room had been set up, co-ordinated by Deutsche Bank to provide information to Eneco. Eneco made an initial offer in September 2010 which was rejected, and it was not until 24 March 2011 that the transaction with Eneco completed. Deutsche Bank continued to be involved with the transaction until that time, advising in relation to other options for the disposal in case the deal with Eneco fell through.[38]PwC’s role was principally in the preparation of a Vendor Due Diligence Report (“the VDD Report”) to be made available to potential purchasers. VDD Reports are generally obtained where there are difficulties in the business being sold. PwC were given full access to Oxxio’s management and financial records. The VDD Report was available to Centrica but was primarily intended for the preferred bidder on the basis that PwC would assume a duty of care to the preferred bidder in relation to the report. In fact, the VDD Report helped Centrica understand what was going on in the Oxxio business and how best to proceed with the transaction. 7[39]In July 2010 PwC produced what was described as a “deep dive” report (“the Deep Dive”). The purpose of the Deep Dive was to enable Centrica to understand the extent of the problems in Oxxio and inform Centrica as to the options available, including whether the problems were so bad that the business was unsaleable. In its half-year accounts to 30 June 2010 Centrica stated that it intended to dispose of Oxxio as soon as practicable and continued to report it within discontinued operations. However, there had been discussions about whether the sale process should be put on hold whilst the problems were resolved or abandoned altogether in light of the issues.[40]PwC submitted their final VDD Report on 28 January 2011, which marked the completion of their work.[41]De Brauw acted as legal advisers, advising Centrica on matters of Dutch law including employment law, competition law, tax, material contracts, the preparation of draft and final sale and purchase agreements. This included advice on the structure of the sale and generally in relation to completion of the transaction. They were involved in preparation of the virtual data room. A draft sale and purchase agreement had been prepared as early as November 2009. The FTT describes some of this work as general advice on legal issues in evaluating potential structures for the sale and some of it as “‘nitty gritty’ transaction specific work”.[42]In November 2010, De Brauw prepared a document setting out a detailed explanation of the steps to be taken to achieve a demerger of the Oxxio businesses, although the transaction eventually took a slightly different form. At the same time, De Brauw provided advice on competition law issues. The transaction timeline[43]The FTT identified four key events in the timeline of the transaction as follows:(1) The strategic decision to sell the Oxxio business in June 2009. The preferred option was a share sale, although from an early stage it was recognised that a share sale might prove difficult and at times it was thought that it might not be possible to sell the business at all.(2) An offer made by Eneco in January 2011 which formed the basis of subsequent negotiations and the eventual sale. At that stage many parts of the deal remained to be negotiated and there was still a high risk that the deal would fail.(3) A Centrica Plc board meeting on 22 February 2011 at which the board approved in principle the specific proposed transaction with Eneco and delegated authority to negotiate the detailed terms to Plc board members including Mr Grant Dawson, who was Centrica group general counsel and a director of COHL. COHL accepted that expenditure incurred after this date was not expenses of management but were expenses of implementing a decision which had been taken to sell the Oxxio businesses. There followed a period of due diligence and intense negotiations and there was still a high risk the deal might fail. Centrica continued to explore other possibilities if the sale to Eneco did not go ahead.(4) The sale and purchase agreement was signed on 24 March 2011, effectively marking the end of the sale process. We have previously described the broad structure of the deal. DISCUSSION[44]We shall now consider the four issues raised by the grounds of appeal and HMRC’s Rule 24 notice. In doing so, we consider the basis on which the FTT reached its decision on each issue. 8 Issue 1 – the basis on which the FTT determined the appeal[45]This issue concerns the FTT’s finding at [235] quoted at [5] above, and the basis on which the FTT determined that the Disputed Expenditure was not an expense of management of COHL’s investment business. The FTT considered this issue at [190] – [237].[46]COHL contends that the FTT wrongly found that the decisions of COHL in relation to the Oxxio disposal were not taken by the directors of COHL acting as such, and therefore s1219(1) was not satisfied. HMRC contend that the decision was in substance that COHL had not established evidentially that the Disputed Expenditure constituted expenses of management of COHL’s investment business. In particular, COHL had not established a sufficient link between the Disputed Expenditure and any investment business of COHL.[47]Clearly, it is a condition of the relief that the expenditure for which it is claimed must be expenses of management of the investment business of the company claiming relief. In that sense there must be a link between the expenditure and the investment business of the company claiming relief. HMRC say that COHL sought to establish the requisite link with evidence from directors of COHL and Centrica, namely Ms Charlotte Redcliffe who was Head of Tax of Centrica group and a director of COHL, Mr Richard McCord the Finance Director of Centrica Plc and Mr Christopher Defert, who we understand was a senior employee of Centrica Plc working in mergers and acquisitions.[48]Both parties acknowledge and accept that HMRC did not make any submission to the FTT that any lack of formality in the decision-making should lead to a conclusion that COHL was not managing its investment business.[49]The FTT devoted a section of its decision to the construction of s1219 at [160] – [189]. COHL’s argument is recorded at [162] as follows:[162]Mr Rivett took the view that all that section 1219 requires is that the management expenses are “in respect of” the company’s investment business. He submits that, as a matter of construction, section 1219(2)(a) makes it clear that for the purposes of section 1219(1) only some connection (however limited) is required between the relevant expenses and the relevant investment business for an expense to constitute an expense of management of a company’s investment business. There is no requirement that the expenses be wholly and exclusively incurred on management. The business of the company need not be wholly that of investment. It does not even require the management to be by the company itself (although he submits that in this case it was). All that is required is that there is an investment business and the expenses must relate, in some way, to the management of that business.[50]HMRC’s argument was described as placing a “more restrictive” construction on the section. The argument is recorded at [161] and [174] as follows:[161][Mr Henderson] stated that tax law works on an entity basis and in order to be deductible, expenses must relate only to the business of a specific entity. He described section 1219 as “a closely targeted relief aimed at expenses incurred in managing the investments held as part of a specific investment business”.[174]Mr Henderson submitted that … expenses of management were deductible as expenses of management of a company’s investment business only to the extent that they were “in respect of so much of the company’s business as consists in the making of investments”. He argued that in this context, the expression “in respect of” was to be construed as a restriction (to expenses incurred by reference to a particular part of the company’s business) and he denied that the draftsman intended to be “conspicuously expansive” and require only a loose connection between the expenses and the investment business. 9[51]The FTT’s conclusion on this issue of construction was as follows:[185]… it seems to me that the plain words of sub-section (2) of section 1219 indicate that it is intended to limit, rather than expand, the meaning of “expenses of management of a company’s investment business”. Sub-section (1) makes clear that where a company has investment business, what is deductible is the expenses of management of that particular company’s investment business. It does not include the expenses of management of any other business which the company may carry on.[186]Sub-section (2) goes on to say that for these purposes, expenses of management are expenses of management of the company’s investment business “so far as (a) they are in respect of so much of the company’s investments business as consists of making investments…”. The words “so far as” indicate that not all expenses of management of the company’s investment business are deductible. Only those that satisfy the provisions of paragraphs (a) and (b) are included. Paragraph (b) relates to unallowable purposes and it is not suggested that that is relevant here. Paragraph (a) requires the expenses to be “in respect of so much of the company’s investment business as consists of making investments”. I construe “in respect of” to mean that there must be a close link between the expenses in question and that part of the company’s business which consists of making investments. That is reinforced by the words “so much of” which focusses on the part of the business which consist of making investments.[187]In the context of the section and the relief as a whole, I am unable to interpret section 1219(2)(a) as requiring only some sort of loose connection between the expenses and the investment business.[52]It was against that background that the FTT went on to consider whether the Disputed Expenditure was incurred in respect of COHL’s investment business. Pausing there, it is not clear to us why it is necessary to examine the strength or weakness of any link between the expenditure incurred and the company’s investment business, at least on the facts of this case. To use neutral terms, the Disputed Expenditure was clearly directed towards dealing with the Oxxio business. The issue between the parties was whether it was incurred by COHL in the management of its investment business. Either there is a link between the expenditure and the management of COHL’s investment business or there is no link.[53]HMRC’s submissions to the FTT are recorded in various paragraphs as follows:[191]HMRC accepts that COHL is an investment company. Mr Henderson does not seek to argue that COHL did not incur the expenditure and he does not seek to argue that the mere fact there was no agreement between Centrica and COHL that COHL would bear the cost of the advice means, of itself, that the expenses are not deductible. He does take the point that the expenses were not expenses of managing COHL’s investment business. He submits that the expenses have to be linked to the relevant investment business.[202]Whilst recognising that the facts of these cases are very different from ours, Mr Henderson draws from them the proposition that in order to be deductible the expenses of management have to be in relation to the management of a particular company’s investment business and in this case the company is COHL.[203]He submitted that it was clear that the expenditure was arranged by Centrica and that the investments were disposed of by Oxxio and not by COHL. He recognised that there was no wholly or exclusively test in section 1219 but the expenses must still be expenses of COHL’s investment business.[205]Mr Henderson acknowledged that in large groups such as Centrica, management tends to focus on businesses and does not necessarily approach things strictly on a legal entity by legal entity basis. He submits that the tax code does focus on legal entities and in the present case it is 10 necessary to show that the expenses were incurred on the management of the investment business of the specific legal entity concerned, namely, COHL.[206]He had two main points on this. First, that COHL did not actually do any management at all as everything was done at Plc level and secondly that from an early stage it was clear that a sale by COHL of the shares in Oxxio was unlikely and COHL did not own the businesses which were ultimately sold so that the deal which was done could not be regarded as part of the management of COHL’s own investment business.[54]The FTT was not persuaded by the second argument relied on by Mr Henderson. At [219] it stated as follows:[219]Although the Eneco transaction did not involve the full disposal of the Oxxio businesses, it did result in Oxxio BV selling, via the partial demerger, its customer business and other assets. This benefitted COHL’s investment business in two ways. First COHL’s costs were reduced as it no longer needed to provide the financial support by way of equity and debt investment which it had previously done in order to try and stem the large losses. Secondly, COHL received a substantial amount of money out of the proceeds of the transaction by way of repayment of its loans.[55]The FTT held at [220] that the fact COHL itself did not dispose of anything was not a bar in itself to the Disputed Expenditure being expenses of management of COHL’s investment business. The FTT then went on to deal with the first argument at [222] to [235]. It is necessary to quote extensively from the reasoning of the FTT which led to its conclusion at [235]:[221]Returning to Mr Henderson’s first point: he submitted that the requirements of section 1219 were not satisfied because COHL did not manage anything at all. All the relevant decisions were taken by Plc.[222]It is clear that the Board of Plc took the strategic decisions in relation to the businesses of the group. That is hardly surprising. It is what ultimate holding companies do. It is also not surprising that functions like legal, finance and tax were provided by teams at Plc level which then provided their services in relation to the various subsidiaries in the group which needed them…[224]There is no reason why a subsidiary should not pay for services provided or procured by its “Topco”, but in order to determine if those payments are deductible, we must return to the words of section 1219…[225]As I concluded above, and as indicated in Dawson and Howden, there is a requirement that the expenses incurred by a specific company must be linked to the management of that company’s investment business in order for that company to deduct the expenses from its profits.[226]Mr Rivett contends that the section does not require the management to be carried out by the company itself. It is true that section 1219 does not say so in so many words, but it is implicit that it must be the company itself which manages its own business. One has to focus on the investment business of a specific company and on the management of that business. The only body which is authorised to carry on the management of a company’s business is the board of directors of that company. The shareholders of a company may have influence over a company’s actions and may be able to procure a company to do certain things by exercising their voting rights as shareholders, but the day-to-day management of a company’s business is, legally, the province, I would say the exclusive province, of the board of directors of the company (or people to whom the board has delegated authority). 11[227]In order for expenses incurred by a company to be deductible therefore, they must be expenses linked to the management of the company’s investment business and the management must be carried out by the company, acting by its directors, as they are the only people who are authorised to manage the company’s business.[229]The real difficulty, as Mr Henderson pointed out, is that COHL was not actually managing anything. Taking a realistic view of the facts it was Plc which made all the decisions, strategic and otherwise. The various group functions did not think of themselves as providing services to COHL, they were working to give effect to Plc’s strategic decision to divest itself of the Oxxio businesses and exit the Dutch energy market. Although Ms Redcliffe and Mr Dawson were directors of COHL as well as senior employees within the group functions, the evidence indicates that throughout the Eneco transaction they received information, provided advice and made decisions with their Plc Head of Tax/General Counsel hats on. It was admitted that there was no real distinction between the capacities in which they were acting and there is no evidence to show that they changed hats and took decisions in their capacity as directors of COHL.[232]As Mr Henderson submitted, tax law operates at the level of the legal entity. If COHL wishes to deduct expenditure as expenses of management of its investment business, it has to show that it, acting by its board of directors, carried out the management of the investment business as a result of which it incurred the expenditure. COHL has not done this. To the extent that the people who were directors of COHL participated in managing the process of the disposal of the Oxxio business they did so in their group capacities and there is no evidence to show that they took any relevant decisions as directors of COHL or that the advice from Deutsche Bank, PwC and De Brauw, the cost of which they incurred, was used by them as directors of COHL in conducting COHL’s investment business.[233]Mr Rivett submitted that section 1219 must be capable of working in the context of groups and if a deduction was denied in this case it would mean that it would be virtually impossible for an intermediate holding company to obtain a deduction for its expenses of management and that the object of the legislation, to tax investment companies on their true economic profit, would be defeated. I disagree. Mr Rivett tried to explain the lack of minutes for COHL by saying that they were only produced where there was a need for corporate decision making, for example where there was a need for a formal audit trail. It seems to me that a decision to dispose of a significant investment, especially such a difficult one, is more than worthy of a corporate decision. There were a number of minutes of Plc and committees of its Board dealing with the progress of the Oxxio transaction and making decisions about it. The few minutes of COHL that were in my bundles acknowledged the roles of the group functions in recommending/approving the proposed actions and preparing the relevant documents. Had the directors of COHL, in that capacity, considered the actions recommended by Plc and made decisions, as directors of COHL, about the Oxxio transaction recorded in minutes of the meetings of COHL’s board, it might have been possible to say that COHL was managing its investment business.[234]…It is not the case that section 1219 does not allow relief in group situations, but in order to obtain relief, group companies must be conscious of the need for the “corporate plumbing” to be properly installed and must ensure that the relevant investment company manages its own investment business, even if strategic decisions are taken elsewhere in the group.[235]I have decided that COHL did not itself carry out the management activities in relation to which the disputed expenditure was incurred. The conditions for relief in section 1219 are not therefore satisfied as the expenditure was not expenses of management of COHL’s investment business.[236]That disposes of the appeal, but in case I am wrong, I will now consider the other issues raised by HMRC. 12[237]I am assuming, for the purposes of considering the remaining issues that, contrary to my finding above, the disputed expenditure was an expense of COHL’s investment business. In this context I will sometimes refer to COHL below as doing things which were in fact done by Centrica. (emphasis added)[56]Before us, Mr Henderson submitted that there was relatively little evidence before the FTT as to what COHL’s investment business comprised, which it was necessary to understand in order to decide whether the Disputed Expenditure was for the purposes of that business. We do not accept that submission. The FTT described COHL as a “substantial intermediate holding company” and described the extent of its overseas investments, focussing as one would expect on its holding in Oxxio and the nature of the Oxxio subsidiaries.[57]Mr Henderson’s primary submission on Issue 1 was that the FTT’s reasoning was in the context of determining whether there was, on the evidence, a sufficient link between the Disputed Expenditure and any investment business of COHL. In circumstances where COHL did not engage the advisers and did not make any disposals of any of its investments, where there was only limited evidence of what its investment business involved and where there was no evidence of the directors of COHL engaging in the Oxxio transaction in their capacity as such, COHL had failed to demonstrate how the disputed expenditure could be treated as expenses of management.[58]We do not accept Mr Henderson’s analysis of the FTT’s decision. It is clear to us on a reading of the FTT’s decision as a whole that the refence to “corporate plumbing” at [234] was to a perceived need for board minutes or resolutions of COHL relating to the management of COHL’s affairs. In our view the FTT did view the absence of formality in relation to the decision-making of COHL as fatal to its claim for relief. That is why at [233] the FTT states that if the directors of COHL had made decisions in their capacity as such, recorded in board minutes of COHL, then it might have been possible for COHL to say that it was managing its investment business. With respect, in our view the FTT was wrong to consider that it was necessary to have a minimum degree of formality as to decision-making in relation to the Oxxio transaction before the Disputed Expenditure could be treated as expenses of management of COHL.[59]It appears to us that the FTT was looking for an outward expression that the directors of COHL were making decisions in relation to Oxxio with their COHL “hats” on rather than their Centrica group “hats” on. But the FTT acknowledges at [229] that Ms Redcliffe and Mr Dawson “received information, provided advice and made decisions with their Plc Head of Tax/General Counsel hats on”. However, it says that there was no evidence to show that they “changed hats and took decisions in their capacity as directors of COHL”. In our view, no such formality was required.[60]Mr Rivett made various submissions as to why the FTT was wrong to conclude that management of an investment business must be undertaken by directors acting formally in their capacity as such. Essentially, those submissions were to the effect that there is nothing in s1219 which requires such a degree of formality. We agree with that overarching submission.[61]There was no dispute that the Disputed Expenditure was incurred; that COHL ultimately bore the cost of that expenditure; that the expenditure was properly recorded in the accounts of COHL; and that the disposal of the Oxxio business was of benefit to COHL. Mr Henderson also accepted in his oral submissions that if COHL had an investment business then someone must be managing that investment business. The FTT rightly pointed out that the directors of 13 a company may delegate management of the company’s business to third parties, subject to the provisions of its articles of association. In the present context, article 5 of COHL’s articles of association provides in wide terms that the directors may delegate any of the powers conferred on them to any other person or committee as they see fit.[62]The directors of COHL were clearly participating in the strategic decision-making in relation to Oxxio and in the circumstances that was sufficient to evidence that the Disputed Expenditure was incurred in relation to COHL’s investment business. In our view the FTT was wrong to hold otherwise.[63]There is no reason why the investment business of an intermediate holding company should not be managed on behalf of that company by individuals employed by the ultimate holding company, especially where those individuals include, as they did here, directors of the intermediate holding company who fully participated in the decision-making process. It may well be desirable for decision making and any delegations of authority in that context to be recorded in board minutes or correspondence. However, such formality is not necessary in terms of delegation of authority for the purposes of s1219.[64]The FTT accepted that COHL had an investment business which included the Oxxio group. In oral submissions, Mr Henderson sought to suggest that there was no evidence that this was COHL’s investment business. However, that was not the basis on which the FTT decided the appeal and it was not an issue raised in HMRC’s Rule 24 notice. The three sets of board minutes referred to by the FTT evidence the authorisation of three large finance transactions undertaken by COHL in relation to the Oxxio group. There is no doubt that the decisions of COHL in relation to those transactions amounted to the management of COHL’s investment business.[65]Mr Henderson argued that a group cannot just choose which holding company or intermediate holding company incurs the expense and thereby obtains relief for the expenses of management. He suggested that the most likely candidate for who was managing the investment in Oxxio group was Oxxio itself. It seems to us that all holding companies and intermediate holding companies will have an investment business. But in the circumstances of this case it was COHL which owned and financed the Oxxio group and it was that investment which was being managed, even if in the end there was no disposal of the Oxxio shares. COHL would therefore be the likely candidate and that is the basis on which the expenditure was charged to the accounts of COHL. HMRC accepts that it was properly so charged, or at least they have not challenged COHL’s accounting.[66]The FTT found that Centrica Plc through its group functions took strategic decisions in relation to the group, including COHL. The FTT’s decision appears to be that it was Centrica Plc that was managing COHL’s investment business, rather than COHL. However, in circumstances where the directors of COHL participated in that decision-making we consider that this was management of the investment business on behalf of COHL with the informal approval of its directors. We are satisfied on the basis of the FTT’s findings of fact that COHL was managing its investment business and the FTT was wrong to conclude otherwise.[67]We therefore allow the appeal in relation to Issue 1, which means that we must go on to consider Issues 2 – 4. Issue 2 – expenses of management[68]This issue concerns whether the Disputed Expenditure amounted to expenses of management or expenses of the disposal of Oxxio assets. It involves a close consideration of the following authorities: a decision of the House of Lords in Sun Life v Davidson [1958] AC 14 184 (“Sun Life”) and decisions of Patten J, as he then was and of the Court of Appeal in Camas Plc v Atkinson [2003] EWHC 1600; [2004] EWCA 541 (“Camas”).[69]Sun Life concerned the taxation of life assurance companies. Sun Life carried on the trade of an insurance business, and also an investment business which involved the investment of life insurance premiums to meet its obligations to policyholders. The Inland Revenue, as it was, was entitled to elect to charge Sun Life to tax on its trading profits or on the income from its investments. It elected to charge tax on Sun Life’s investment profits which meant that Sun Life was entitled to relief for expenses of management pursuant to s33 Income Tax Act 1918, a predecessor to s1219. In the ordinary course of its activities, Sun Life held investments which were described as “part of its circulating capital” and which did not constitute part of its fixed capital assets. Those investments included investments in quoted and unquoted shares. The question was whether Sun Life was entitled to relief as expenses of management for stamp duty and brokerage fees on the purchase and sale of investments. The House of Lords held unanimously that stamp duties were not expenses of management, and by majority (Lord Reid dissenting) that brokerage costs were not expenses of management.[70]Counsel for Sun Life had conceded that not all expenses of conducting the business would be expenses of management, and in particular that expenses of management would not include the costs of purchase of investments themselves.[71]Viscount Simonds regarded the concession as “illogical”. However, on the basis of the concession he found that stamp duty and brokerage could not be expenses of management. His reasoning was as follows: … if the expense of purchasing an investment is not an expense of management, I can see no valid ground of distinction between the price of the stock which is purchased and the stamp duty paid upon contract or transfer and the brokerage paid to the broker.[72]The speeches in Sun Life all proceed explicitly or implicitly on the basis of Sun Life’s concession. They adopt slightly different descriptions of the test as to what amount to expenses of management in the context of the acquisition of an asset as follows: Each item is an integral part of the cost of acquisition or, as the Commissioners put it, a part of the expenses of the particular purchase, not of the expenses of management. Viscount Simonds These expenses are … so closely linked with the transaction of purchase that they may naturally be considered as items in the total cost of a purchase which has already been resolved upon by the management of the company, and not as expenses of management. Lord Morton The brokerage and stamp duty, though not … an integral part of the purchase price, are a direct and necessary part of the cost of a normal method of purchase. Lord Somervell … looking to the purpose and content of the section it appears to me that the phrase has a fairly wide meaning so that, for example, expenses of investigation and consideration whether to pay out money either in settlement of a claim or in acquisition of an investment must be held to be expenses of management … It seems to me more reasonable to ask, with regard to a payment, whether it should be regarded as part of the cost of acquisition, on the one hand or, on the other hand, something severable from the cost of acquisition which can properly be regarded as an expense of management. Lord Reid 15[73]The speeches in Sun Life were considered by Patten J and then by the Court of Appeal in Camas. At the time of those decisions in 2003 and 2004, the predecessor to s1219 did not have any exclusion for expenses of a capital nature. Camas Plc was the holding company of a trading group and was listed on the London Stock Exchange. It identified the possibility of merging with or otherwise acquiring Bardon Group Plc (“Bardon”), which was also a quoted company and approximately the same size as Camas. Camas incurred fees to professional advisers in connection with making an offer for Bardon. An offer was made but rejected by Bardon and the bid was abandoned. The Special Commissioners held that the expenditure was not expenses of management within what was then s75 Income and Corporation Taxes Act 1988. Camas appealed to the High Court and the appeal was allowed by Patten J who held that the expenses were expenses of management. The expenditure in issue in Camas was similar to the expenditure in the present appeal. It was described by Patten J at [14] as follows:[14]Schroders' fees relate to their work as the financial adviser to Camas in respect of the takeover. This includes advice on strategy and tactics throughout the process. It includes the formulation of alternative strategies in relation to a possible merger or a takeover, advice on the methods of financing any offer, and an appraisal of the financial impact of successful offers for Bardon over a range of prices. I have already referred to their attendance at board and other meetings and to the papers which they prepared. Warburgs as brokers also gave advice on strategy, as well as carrying out an analysis of the register of Bardon shareholders. The fees of KPMG relate to a comparison of the financial performance and accounting policies of the Camas and Bardon groups, a review of the benefits of integrating the two businesses, and a review of the profit forecast for the Bardon group. They also carried out an assessment of the borrowing requirements in the event that the takeover offer succeeded. Clifford Chance advised on the impact on the proposed transaction of competition laws and on what information would need to be submitted to the Office of Fair Trading. They also advised on the information that would require to be included in any circular, and gave advice to the directors of Camas on their responsibilities under the City Takeover Code and the Stock Exchange Yellow Book. Together with Schroders and Warburgs they attended a number of board meetings. Shearman & Sterling advised on US anti-trust issues and provided an analysis of the impact of US securities laws. FPC Greenaway are printers and their services related to printing the offer documentation, listing particulars, circulars to shareholders and some press releases. All of this expenditure was charged by Camas to the profit and loss account in accordance with commercial accounting practice.[74]There were also findings of fact based on expert evidence from an investment banker to the effect that the consequence of undertaking a public offer was that an extensive level of preparation was required when no decision had yet been taken to make an offer. The act of working up a potential offer was itself part of the decision-making process. Patten J summarised the findings of the Special Commissioners in this regard at [17]:(a) that all the expenditure was relevant and necessary to Project Bardon;(b) that all the costs (apart from the printing costs) related to advice given to assist the board of Camas in making decisions about a possible bid; and(c) that most of the expenditure also had the dual function of providing the necessary starting-point for a bid and would have to be incurred by a potential bidder in mounting a bid.[75]Patten J conducted an extensive review of the history of the legislation giving relief to investment companies for their expenses of management, and the interaction of that relief with capital gains tax on the introduction of that tax in 1965.[76]The Revenue relied on two principal arguments that Camas should not have relief for the professional fees it had incurred. First, that the phrase "expenses of management" does not include expenses which are wholly attributable to the acquisition of a particular investment and 16 are therefore part of the costs of that acquisition. Second, and more fundamental that s75 did not permit relief for expenditure of a capital nature.[77]Patten J rejected the second argument and for present purposes it is not necessary for us to examine his reasons for doing so. He went on to consider what he described as the main issue on the appeal, which was whether the fees and other expenses incurred were expenses of management. He considered the speeches in Sun Life in detail and at [36] he said:[36]I have quoted at some length from the speeches in Sun Life [1958] AC 184 because both parties accept that they contain the law which I have to apply. Mr. Prosser Q. C., for Camas, invites me to accept as the correct test the question posed by Lord Reid as to whether the costs under review should be regarded as part of the costs of acquisition or something severable from it, which can properly be regarded as an expense of management, and the Crown has accepted in its skeleton argument that this is the true test, or at least the most helpful formulation of the test in the present context. I agree with that, but the real difficulties of course arise in its application to the facts. In a sense Sun Life [1958] AC 184 was the easy case to decide. Both stamp duty and brokerage fees only arise and become payable as part of the costs of sale and purchase of the investments. In the case of stamp duty the tax is payable by reference to, and as a charge on, the purchase price. Neither was any less consequential on the exercise of a management investment decision than the costs of purchase themselves, and it was therefore relatively easy to draw the line.[78]The Revenue in Camas conceded, in the context of their argument that the expenditure was not expenses of management, that if the expenses were not part of the purchase costs then they were expenses of management. As in Sun Life therefore, the issue was whether or not the expenditure for which relief was being claimed fell to be treated as part of the costs of the projected acquisition of the Bardon group.[79]At [43], Patten J held that the Special Commissioners had wrongly adopted a “causative test” giving rise to a result that any expenditure relating to a management decision to invest would be disallowed on the basis that it was an essential part of the process leading up to the acquisition. He expressed his conclusion at [44] as follows:[44]It seems to me that the correct answer to Lord Reid's question is that the expenses under consideration are not part of the costs of acquisition. On the Special Commissioners' own findings of fact, the services of Schroders and the other professionals were needed and were used in order to obtain advice on a possible investment in the form of the acquisition of the Bardon group, and to decide whether to go ahead. The work stopped when, on advice, the decision was taken to abort any possible acquisition. But even if the acquisition had gone ahead, the nature of the services would have been the same. Although one element of the professional services involved the working up of the bid, Mr Reed's evidence indicated that this was part of the decision-making process, and the Commissioners accepted that. I am unable to see how the cost of any of this can fairly be described as part of the cost of acquisition in the sense that brokerage fees, payments for financing and stamp duty obviously are, and the Special Commissioners have reached their conclusion, in my judgment, by asking themselves the wrong question.[80]The Court of Appeal upheld Patten J. In relation to the test in Sun Life, Carnwath LJ stated at [21]:[21]… expenditure is not excluded merely because it relates to activities carried out in contemplation of acquisition. Lord Reid said that the expenses of ‘investigation and consideration’ whether to pay out money in acquisition of an investment should be treated as expenses of management… Nothing in the Sun Life case supports the exclusion of the cost of 17 investigations and other activities which are part of the process leading to the decision to purchase. As Mr Prosser says, what is excluded by Sun Life is expenditure on ‘the mechanics of implementation’.[81]The Court of Appeal also considered a decision of the Irish Supreme Court in Hibernian Insurance Co Ltd v MacUimis [2000] 2 IR 263 which concerned equivalent provisions in the Irish Corporation Tax Act 1976. In that case it was held that expenses similar to that in Camas on an acquisition which did not proceed were not expenses of management. Hibernian’s argument that the costs were deductible when incurred before the decision to purchase but not if incurred after it was rejected. Costs incurred of exploration, evaluation and investigation of a particular acquisition were held not to be expenses of management.[82]The Court of Appeal rejected the Irish Supreme Court’s application of the principles in Sun Life. Carnwath LJ said as follows:[26]… It has to be borne in mind that the costs of purchase came into the debate in Sun Life, not as part of a statutory test, but simply because, by common agreement, it was accepted that the act of purchase as such was not within the ordinary meaning of “management”. It therefore became necessary to determine what expenses were an integral part of such a purchase. By contrast, it was common ground that the process of reaching a decision to purchase was management in the ordinary sense. There is nothing in the speeches which supports the view that an activity which is part of that decision-making process ceases to be management, merely because it may also assist in the purchase if that is decided upon—still less if it is not. Unlike the provisions relating to Schedule D expenses, there is no requirement that the expense should be “wholly and exclusively” related to management. …[30]Before us, Mr Henderson [now Henderson LJ] supports the commissioners' reasoning. He accepts that expenditure in deciding on what sort of company to acquire is not excluded. However, expenditure ceases to be on “management” once a particular company has been identified: “The starting point is when a particular target has been identified and where expenditure is directed to acquisition of that particular target.” (This formulation is similar to that of Barron J in the Hibernian Insurance case.)[31]On this issue I agree, respectfully, with Patten J. It is a short point. If my analysis of the speeches in Sun Life Assurance Society v Davidson [1958] AC 184 is correct, the activities in this case were all part of the process of managerial decision-making. I see nothing in Sun Life, or in the ordinary meaning of “management”, which provides any support for Mr Henderson's suggested distinction between the process of deciding on the sort of company to acquire, and that of deciding on the acquisition of a particular company.[32]On the facts of this case, unlike the Sun Life case, no final decision to purchase was ever made. As Asquith LJ put it in another context, the project never “moved out of the zone of contemplation—out of the sphere of the tentative, the provisional and the exploratory—into the valley of decision”: Cunliffe v Goodman [1950] 2 KB 237, 254. The revenue's argument might have been stronger if the stage had been reached of a “firm intention to make an offer”, triggering the “strict timetable” described by Mr Reed. Even then, I would not necessarily conclude that any expenditure thereafter, even if the purchase proceeded, would have to be treated as costs of acquisition, rather than management. It must depend on the circumstances. Between such a triggering event and a final purchase there may be many chances and changes, requiring what can properly be regarded as “managerial” consideration. How one should categorise particular expenses in any such case must depend on the particular facts.[33]Like the judge, I do not see this conclusion as involving any disagreement with the Commissioners on their findings of fact. They held that, on the evidence relating to acquisitions of this scale, "the act of working up a potential offer is part of the decision-making process", and that in this case the work was "wholly directed at the projected acquisition". They saw these findings as leading to the conclusion that it was "a direct and necessary part" of the proposed 18 acquisition in this case. That formula reflects the words of Lord Somervell in Sun Life, adopted by May LJ in Hoechst. With respect to the Commissioners, however, I think they misapplied the formula. Lord Somervell made clear that he regarded that expression as a very narrow one, not intended to be wider than the particular items in issue in that case. The fact that the work was part of "the decision-making process" supports its categorisation as managerial. That is not affected by the fact that it was also a "necessary" prerequisite to acquisition, and directed to that possibility. It was preparatory to the making of a decision to purchase, not part of the implementation of a purchase already decided upon.[83]It is clear therefore that there is a distinction between expenses incurred in deciding whether to acquire or dispose of an asset, and expenses incurred on the “mechanics of implementation” once that decision has been taken. The former will be expenses of management and the latter will not be expenses of management. The categorisation of particular expenses will be a fact-sensitive enquiry. Further, in our view expenditure incurred in assessing how to make an acquisition or a disposal may fall on either side of the line. It may be part of the decision whether to proceed or part of the implementation of a decision. The answer will be part of the factual enquiry and may involve a value judgment.[84]Mr Rivett submitted that HMRC’s case was in essence an attempt to apply the approach of the Irish Supreme Court in Hibernian, which had been rejected by the Court of Appeal in Camas. Mr Henderson disavowed any such intention. He accepted on the basis of Camas that expenditure on deciding whether to buy or sell an investment may be an expense of management. It was what he described as “make up your mind expenditure”. However, he submitted that expenditure on implementing a decision to sell an investment is on the other side of the line and not deductible.[85]It does not seem to us there is any fundamental difference in approach between the parties as to the nature of the test. The dispute is really whether the FTT properly applied the test to the facts it had found.[86]The FTT concluded at [334] that 22 February 2011 was the date on which Centrica took the “necessary type of decision” to sell the Oxxio business. This was the date on which the board of Centrica Plc approved the transactions in line with the terms presented to it. The board delegated authority to Mr Dawson and others to negotiate final terms and authority to sign an agreement. The basis on which the FTT identified that date appears at [329] and [330]:[329]I agree with Mr Rivett that in order to have the kind of decision to sell which marks the transition from managing the investments to implementing the decision already made:(1) There must be an identified purchaser who the seller has concluded is satisfactory. That is the seller must be content that the purchaser is credible and has the funds available or can obtain the funds to go through with the purchase(2) The price must be acceptable(3) The broad structure of the transaction must be agreed(4) The other terms of the deal must be broadly satisfactory.[330]In other words, the seller must have decided to go ahead with a specific transaction with a specific purchaser on specific terms. That is not to say that every last detail must be agreed or that subsequent changes to the terms would vitiate the nature of the decision. It is inevitable in an M&A transaction that negotiations continue, important matters are agreed and changes are made even after detailed Heads of Terms have been agreed. What is important is that a decision to enter into a specific transaction is taken. Even that is not an absolute cut-off point. Undoubtedly most of the work will then become implementation, but there may still be decisions made which constitute management… 19[87]Mr Henderson submitted that the FTT erred as a matter of principle in fixing a date to mark the transition between managing the investment and implementing a decision to sell the investment. Instead, he said it is necessary in all cases to identify what the expenditure was for. Mr Rivett acknowledged that an artificial cut off point is not appropriate, and that what is important is the nature of the expenditure either side of the date identified by the FTT. He submitted that the FTT had in fact taken that approach, notwithstanding that it had identified the date.[88]It was common ground before the FTT that everything after 22 February 2011 was implementation expenditure and would not be relieved.[89]If Mr Henderson is right in saying that the FTT adopted a cut-off date, then we agree with him that the FTT was wrong to do so. It is the nature of the expenditure that is important, rather than when it was incurred, although we accept that the timing of expenditure may give an indication as to whether it is “make your mind up” expenditure or implementation expenditure. However, it does not appear to us that the FTT did use the date of 22 February 2011 as marking the date of a transition. Despite what the FTT said at [329] and [330], the FTT said this at [335]:[335]In order to decide whether the expenses which predated this were expenses of management, I now need to consider whether the work which was actually carried out by each of the advisors was to assist Centrica in getting to the point where it could make that decision, or whether it was something else such as general expenses of the business.[90]The FTT then went on to consider the nature of the Deutsche Bank and PwC expenditure and gives reasons why the expenses incurred before 22 February 2011 are expenses of management. For example, at [340] and [348] it summarised its findings of fact as follows:[340]In summary [Deutsche Bank’s] advice was directed to the “how” to get the most value out of the Oxxio business. Its advice informed Centrica as to the options available to them and assisted the company in deciding how best to divest Oxxio and whether to enter into that particular transaction with Eneco. Accordingly the expenditure incurred on the Deutsche Bank fees were expenses of management.[348]The VDD report, although prepared for a prospective purchase did, along with the Deep Dive report assist Centrica with its decision making process and I find that PwC’s fees were also expenses of management.[91]In our view the FTT was not simply applying a bright line test by reference to the date of 22 February 2011. It was applying the test outlined in Camas to the disputed expenditure. That approach is confirmed when the FTT considers the fees of De Brauw. The FTT said this at [315]:[315]The advisors’ fees in the present case will be expenses of management if the work carried out by the advisors was directed at helping COHL to evaluate the Oxxio businesses and/or make decisions, not just about whether to divest itself of the Oxxio group, but also about how best to realise value from it. That is, whether they were expenses of “investigation and consideration”.[93]Mr Henderson submitted that the FTT interpreted expenses of management too broadly. In particular, he criticised the reference to “how best to realise value” from the investment. We do not accept that submission. The question of how best to realise value from an investment may, depending on the facts, be tied up with the question of whether to realise an investment. In our view the FTT simply concluded on the facts of this case that the question of how to 20 realise the Oxxio investment was inextricably linked to the question of whether to sell the business.[94]In the alternative, HMRC say that the date chosen by the FTT at [329] was too advanced. They rely on the decision in June 2009 as the decision to sell the Oxxio business as marking a transition from decision-making to implementation. We have found that the FTT did not use the 22 February 2011 as a transition date, nor is it appropriate to use June 2009 as marking a bright line transition from decision-making to implementation. In this context it is important to recall the FTT’s findings of fact:(1) Centrica was not prepared to sell at any price;(2) A prospective purchaser might not have wanted to buy all the Oxxio businesses and in particular a share purchase might not be attractive.(3) In July 2010, there were concerns that the problems in Oxxio may have been so bad that the business was unsaleable.(4) The decision in June 2009 was only the start of a long and difficult process and the that decision was not in fact implemented – Oxxio BV and Tolling remained in the ownership of COHL. What followed was a process of “investigation and consideration” (per Lord Reid) and cannot fairly be described as the “mechanics of implementation” mentioned in Camas.[95]Mr Henderson also submitted that much of the work of Deutsche Bank involved identifying possible buyers and negotiating the deal, which he said must be viewed as part of the implementation of a decision to sell the Oxxio business.[96]Mr Henderson pointed to some of the work of Deutsche Bank in identifying and dealing with potential buyers, including preparing scripts for conversations with potentially interested parties. We acknowledge that in isolation that looks like work in connection with the implementation of a decision to sell the businesses. Indeed, Mr Rivett accepted that finding a buyer was a major part of Deutsche Bank’s role. However, we are not satisfied from this that the FTT made any error in its approach to applying the test set out in Camas.[97]We accept that Centrica decided that it wanted to sell the Oxxio business, publicised that fact and drew up its accounts on the basis that the Oxxio group would be treated as discontinued businesses held for sale. However, the way in which it was going to achieve a sale and whether it was willing to proceed with a sale depended on what was possible and on the advice it received. The FTT found that the business would not be sold at any price and that it may be necessary to close down the business. Centrica or COHL needed to appraise the business, understand what was going wrong and identify how a beneficial sale could take place. We view the decision in June 2009 as a decision in principle, where many issues still remained to be decided upon. The FTT found that the purpose of PwC’s Deep Dive report in July 2010 was to enable Centrica to understand the extent of the problems in Oxxio and inform Centrica as to the options available.[98]In our view the FTT was entitled to conclude that the Deutsche Bank fees and the PwC fees prior to 22 February 2011 were expenses of management.[99]The FTT dealt with the De Brauw fees at [350] – [365]. It is clear to us that the FTT regarded some aspects of the work done by De Brauw as helping to inform Centrica’s decision making in relation to the disposal. For example, the work on competition law. However, it also regarded other aspects of De Brauw’s work as “nitty gritty” transaction specific in preparation for implementing a disposal decision. For example, the costs of preparing the data room and drafting deal documentation such as the sale and purchase agreement. The FTT described the 21 costs of preparing the data room as part of the general expenses of the business and not expenses of management.[100]The FTT’s findings in relation to the De Brauw fees are not entirely clear. It seems to have taken a view that fees in connection with competition law advice at least might be treated as expenses of management, but later it expresses doubt as to whether any of the De Brauw fees are properly to be treated as expenses of management.[101]At [364] the FTT stated that a time apportionment method to identify deductible expenditure was not appropriate and then said at [365]:[365]Should it be required, the Appellant would need to provide a list of the actual items of expenditure which it contends are deductible, applying the above principles, and agree it with HMRC.[102]The FTT had already decided that the appeal was going to be dismissed following its finding in relation to Issue 1. On that basis, the FTT did not make any attempt to identify what element of the De Brauw fees were properly to be treated as expenses of management. It appears to have left the point open for further argument were it to be successfully appealed on Issue 1 and the parties were unable to agree the element which was to be treated as expenses of management.[103]HMRC say that the FTT should have found that none of the De Brauw fees were expenses of management. COHL say that all the De Brauw fees prior to 22 February 2011 were a necessary part of investment decision-making and as such were expenses of management.[104]We are not in a position to determine what element of the De Brauw fees might properly be regarded as expenses of management. Given our decision on Issue 1, and the way in which the FTT dealt with Issue 2 in relation to the De Brauw fees, we consider it appropriate to remit the appeal to FTT to determine that question. It should do so on the basis of the evidence already adduced and in light of further submissions from the parties. We leave it to the FTT to give further directions for that issue to be determined. Issue 3 – Success fee[105]HMRC contend that the fee payable to Deutsche Bank was a success fee only payable if the transaction was completed and as such it fell to be disallowed as part of the cost of disposal.[106]The FTT referred to what had been said in Camas about success fees. As mentioned above, Patten J in Camas rejected a causative test applied by the Special Commissioners in that case. He found that the expenditure was an expense of management and was not to be regarded as part of the costs of acquisition. At [43], he contrasted that expenditure with a “success fee” which would only become payable in the event that the acquisition proceeded to completion. None of the expenditure in Camas involved success fees, but their treatment was canvassed in submissions because one of the advisers was entitled to a success fee, albeit never paid because the acquisition did not proceed. It was said that the costs would have been “significantly increased” if the bid had been successful. Patten J said this in relation to success fees:[43]… Although not a matter for decision on this appeal, I accept Mr. Prosser's submission that success fees would not be expenses of management, but would fall into the category of severed expenses (like brokerage fees) which cannot be severed from the costs of the acquisition itself.[107]The comments in Camas were obiter and the FTT in this appeal considered that they concerned fees payable over and above the cost of work done by the advisers in Camas. The FTT’s conclusions as to the nature of Deutsche Bank’s fee and that it amounted to expenses of management appears as follows: 22[372]The fixed fee in this case represented the compensation paid for the services rendered over the two year period leading to the disposal, and completion was merely the trigger for payment. The “services” to be provided and set out in the engagement letter were wide in scope consisted in various types of advice and assistance (see paragraph 64 above). The fixed fee was payment for the advice, not for achieving completion. …[391]Deutsche Bank’s fixed fee was payable for the work done, even though it was contingent on an Oxxio Transaction taking place. That it was for the work and not for achieving completion is suggested by the fact that Centrica would sometimes pay some of the fees or promise the next deal to an investment bank even if a transaction fell through. This type of fee is not something necessitated by a sale or an inherent part of the process of sale in the same way that stamp duty and brokerage fees are necessitated by a purchase of shares. And Lord Reid in Sun Life did not even exclude brokerage fees. The parties could have agreed that Deutsche Bank was to be paid for its services on an hourly basis, or by way of a fixed fee payable come what may, but they did not. In accordance with the normal practice in the sector, the fee was contingent.[108]HMRC challenge that conclusion and say that Deutsche Bank was being rewarded for achieving the sale. As such their fees were inextricably linked and inseverable from the disposal of the Oxxio business.[109]COHL support the reasoning of the FTT, which found that the fees were for the advice provided and the work done and as such were not part of the costs of disposal.[110]In a sense, any success fee or contingent fee payable to advisers acting in relation to a transaction is payable for the services. It is only because the adviser provides services that the transaction can proceed. Without the services the adviser would earn no fee, whether it is described as a success fee or otherwise.[111]In this case, Deutsche Bank agreed to provide services and in the event the transaction completed it would receive what was a “fixed fee” of €2.5m. COHL paid that fee, and also paid the incentivisation fee of €1m. The FTT described the incentivisation fee as a “true incentive payment” which we take to mean it was a payment purely to get the transaction over the line and not directly referable to the services provided.[112]It seems to us that a distinction can be made between a fee payable as a reward for achieving an outcome, referable to the value of the outcome, and a fee payable as a reward for services referable to the value of the services. Each may be contingent on the transaction completing. Whilst the observations of Patten J were obiter, it seems likely that he had in mind the former rather than the latter when he referring to the “success fee” in that case.[113]The question is whether the fees of Deutsche Bank, assuming they were otherwise management expenses, become part of the costs of disposal because they were only payable on successful completion of the disposal. In our view the FTT was right to find that in substance the fees were for services which enabled COHL to decide whether and how to dispose of the Oxxio business. The fact that the fixed fee was only payable on completion of the transaction does not change the nature of the expense so as to make it part of the cost of disposal, applying the test set out in Camas. In our view the fixed fee can be severed from the costs of disposal, to use the language of Lord Reid in Sun Life. Issue 4 – were the expenses capital in nature?[114]This issue concerns whether the expenses of management fell to be disallowed as being capital in nature. We shall consider this issue firstly in relation to the fees of Deutsche Bank and PwC. 23[115]The FTT said at [426] that the advice of Deutsche Bank and PwC was to be used to enable COHL to make decisions about its investment in Oxxio. It held at [428] that the advice was used by COHL to conduct its investment business and was revenue in nature. Having said that, the FTT appears to find at [434] that only expenditure up to an initial offer by Eneco in January 2011 was revenue in nature. By implication, expenditure after that point was to be treated as capital in nature.[116]HMRC contend that all the Disputed Expenditure was capital in nature and the FTT was wrong to find that any relief was available.[117]In response to the Revenue’s argument in Camas that capital expenditure could not be an expense of management, Carnwath LJ observed at [38] that the relevant provision contained no reference to a distinction between capital and revenue expenditure. He then referred to the notorious difficulty in drawing a line between the two and at [39] stated:[39]… Unless forced to do so, I would be reluctant to assume that the legislature intended such an uncertain distinction to be introduced into the present debate. (I note in passing that the current Finance Bill apparently seeks to introduce such a distinction into section 75. That may pose problems for the future, but it casts no light on the issues before us.)[118]The Court of Appeal rejected the Revenue’s argument that capital expenditure could not be an expense of management. In doing so, it rejected the Revenue’s reasoning that otherwise investment companies would be more favourably treated than trading companies. Carnwath LJ adopted a submission of the Attorney General in Sun Life: If it were possible to have a capital expense of management (which is doubtful), there would be nothing to exclude it … If the expression ‘expenses of management’ is given its natural meaning, it is unnecessary to consider whether any particular expense is capital or not. The words must be given their ordinary meaning without a gloss.[119]The introduction of s1219(3)(a) CTA 2009 was by way of amendment following the decision of Patten J in Camas. HMRC did not go so far as to say that the amendment was intended to reverse the decision of Patten J. Mr Henderson submitted that it was simply intended to reinforce HMRC’s understanding that capital expenditure could not be an expense of management.[120]Mr Rivett submitted that the amendment may have been intended to capture the cost of investments themselves, following doubts expressed by the House of Lords in Sun Life as to whether the concession was rightly made. Hence, an investment business might otherwise seek to argue that the cost of a new headquarters to house its investment team was an expense of management.[121]We do not consider that the amendment was intended to deal with the concerns expressed in Sun Life as to whether the concession in that case was rightly made. It has not been suggested that following Sun Life taxpayers have ever sought to argue that such expenditure might be expenses of management. There is nothing in Camas which would give encouragement to such an argument.[122]Mr Henderson submitted that the main effect of the amendment was to ensure that the deal costs of implementing a decision to acquire or dispose of a capital asset cannot be treated as expenses of management where they are capital in nature. But such deal costs would not be relieved using the test applied by Patten J or the Court of Appeal in Camas. The costs of implementing a decision to acquire a capital asset are not expenses of management. It is therefore unclear what effect the amendment may have. 24[123]In their skeleton, HMRC rely on various well-known authorities on the distinction between revenue and capital expenditure and describe the principle to be applied to distinguish the two types of expenditure in the present context as follows: The effect of the Disputed Expenditure was to bring about a disposal of the Oxxio businesses. Looking to its effect, it is therefore submitted that the Disputed Expenditure was capital in nature. Further, it is submitted expenditure incidental to a capital transaction is itself capital.[124]If that is right, no kind of decision-making expenditure as to the acquisition or disposal of a capital asset in the context of an investment business would qualify for relief. However, Mr Henderson acknowledged that ongoing monitoring of investments was likely to have a “revenue feel to it” and the test for expenses of management might often give the same result as the test for capital expenditure.[125]We do not need to refer to all the authorities cited to us by HMRC as to the distinction between capital and revenue expenditure. The point is illustrated by one of the authorities relied upon by HMRC, namely Sargent v Eayrs [1973] STC 50. In that case, Goff J held that expenses incurred by a farmer in visiting Australia with a view to buying a farm there were capital in nature. HMRC submitted that the same principle should be applied in the present case.[126]In our view, HMRC’s test as articulated in its skeleton would clearly suggest that Camas would be decided differently today as a result of s1219(3)(a). However, Mr Henderson was reluctant to go so far. In oral submissions he accepted what was said in Bramwell on The Taxation of Companies and Company Reconstructions (2018) at A8.2.3: The disallowance of capital expenditure requires a purposive interpretation if it is not to preclude all expenditure from qualifying. That is to say, "capital expenditure" could be said to encompass all the expenditure of an investment business given that all the assets are held on capital account. However, the contrast intended to be drawn appears to be between routine care and maintenance (allowable) and expenditure on actual or intended acquisitions and disposals (not allowable). The difficulty is. where to draw the line, as where an investment company is considering expansion by acquisition but has not decided on any particular target.[127]In our view, Parliament must have intended that some expenses of management would not be relieved where they were also capital in nature. On the basis of what was said in Sun Life and Camas, such expenditure is likely to be very limited in nature and it may be that Parliament was acting out of an abundance of caution and because HMRC had long considered that capital expenditure was excluded from relief under s 1219. The meaning of capital expenditure in the context of expenses of management is necessarily more limited than the meaning in the context of trading businesses. In our view it is aimed at expenses which do not normally recur, but which have the effect of creating, enhancing or disposing of a capital investment. It does not exclude expenditure which informs decision-making and the exercise of managerial discretion.[128]In a case such as this, expenses of management are likely to be revenue expenses because the test is similar. The expenditure was not one-off in nature because COHL had many capital investments apart from Oxxio Group, which might involve management from time to time including appraising an acquisition, disposal or restructuring, and because Oxxio Group would not necessarily be sold. We consider that the FTT was right to conclude that the Deutsche Bank and PwC expenses were not capital in nature.[129]We now turn to consider the fees of De Brauw. At [429] the FTT stated that the advice and work of De Brauw was different in nature from that of Deutsche Bank and PwC: 25[429]… it was directed, not to enabling the company to conduct its business of managing investments, but to preparing to carry out a transaction, once the nature of the transaction had been determined, and once a decision had been made, carrying out the transaction.[130]In relation to the fees of De Brauw, the FTT found at [437] as follows:[437]It is difficult to see that the De Brauw fees were revenue expenses of management for work done in assisting COHL in considering how to deal with its Oxxio investment, in the same way that the fees of Deutsche Bank and PwC were, at least up to the point where an “in principle” transaction was being pursued. Except to the extent it may have provided generic legal advice, the effect of De Brauw’s work was to bring about a disposal of the Oxxio business. Accordingly, I find that, De Brauw’s fees were capital expenses, even to the extent they could be considered expenses of management. Accordingly, they are not deductible under section 1219.[131]COHL contends that the FTT was wrong to find that no relief was available for the De Brauw fees.[132]It is difficult to see how the FTT concluded that some of the fees of De Brauw were expenses of management but fell to be disallowed as being capital in nature. The explanation is that the FTT appears to have considered that if there comes a point where a business has decided in principle to sell an asset, how to sell the asset and who to sell it to, so that there is a “potential transaction”, the subsequent expenditure is more closely connected with the capital transaction than with the decision-making process. The FTT’s reasoning appears in [433]:[433]However, if there comes a point at which those questions have been answered, at least in principle, so that there is a potential transaction, with a credible offeror, having made an acceptable offer to do an acceptable kind of deal, the subsequent expenditure on bringing that potential deal (being a capital transaction) to fruition is more closely connected with the capital transaction than with the decision making process and will itself be capital in nature. Clearly, many further decisions will be needed as a result of the ongoing negotiations. The price may be refined. The precise nature of the transaction may change. How exactly should the asset sale be effected, should there be one demerged company or two and so on? Many of these may be management decisions, but the expenditure in relation to them will be capital in nature.[133]On the basis of this reasoning, the FTT considered that the “cut off point” to be applied was the initial Eneco offer in January 2011. It is not clear why the FTT chose this as the relevant cut off point for the purposes of the capital/revenue issue and 22 February 2011 as a relevant date in relation to the expenses of management issue.[134]To the extent that the De Brauw fees were expenses of management because they informed decision-making in relation to the disposal of the Oxxio business it seems to us that they were revenue in nature and the FTT was wrong to hold otherwise. They did not amount to one-off, non-recurring expenditure because decision-making in relation to COHL’s investment business was an on-going activity. The investment business included a number of significant investments and from time to time it would need to consider whether to hold or dispose of such investments. CONCLUSION[135]For the reasons we have given we allow the appeal. We re-make the decision of the FTT in relation to the expenses of Deutsche Bank and PwC. That expenditure qualifies for relief under s1219 CTA 2009 and COHL’s underlying appeal against HMRC’s decision to refuse relief is allowed. COHL’s appeal in relation to the expenses of De Brauw is remitted to the FTT to be determined by the same Judge as indicated above. 26 Signed on Original MR JUSTICE FANCOURT UPPER TRIBUNAL JUDGE CANNAN Release date: 18 August 2021
Cited in 1 later judgment