|[Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]|
First-tier Tribunal (Tax)
You are here: BAILII >> Databases >> First-tier Tribunal (Tax) >> Marathon Oil UK LLC v Revenue and Customs (INCOME TAX/CORPORATION TAX : Capital allowances)  UKFTT 822 (TC) (13 November 2017)
Cite as:  UKFTT 822 (TC)
[New search] [Context] [View without highlighting] [Printable PDF version] [Help]
 UKFTT 822 (TC)
Appeal number: TC/2016/01228
CAPITAL ALLOWANCES—whether payment to subsidiary gave rise to “special allowance” for general decommissioning expenditure— meaning of “ incurred on decommissioning plant or machinery”— appeal dismissed
- and -
THE COMMISSIONERS FOR HER MAJESTY’S
REVENUE & CUSTOMS
JUDGE THOMAS SCOTT
Sitting in public at the Royal Courts of Justice, Strand, London WC2 on 4, 5 and 6 July 2017
Jonathan Peacock QC of Counsel for the Appellant
Akash Nawbatt QC of Counsel, instructed by the General Counsel and Solicitor to HM Revenue and Customs, for the Respondents
© CROWN COPYRIGHT 2017
1. This appeal concerns the entitlement of Marathon Oil U.K., LLC (“MOUK”) to a “special allowance” for capital allowances purposes in respect if its accounting period for the year ended 31 December 2008.
2. The only issue is whether a payment of $300 million made by MOUK to its subsidiary on 18 December 2008 was “expenditure…incurred on decommissioning plant or machinery” within the terms of section 163 of the Capital Allowances Act 2001 as they applied at that time.
3. The commercialisation of an oil or gas field in the North Sea is a long-term undertaking. In relation to “upstream” activities (namely those related to the exploration for and production of crude oil or natural gas) the four typical phases in the life of a field would be exploration and appraisal, development, production and decommissioning.
4. In the decommissioning phase, with which this appeal is concerned, the infrastructure which has been installed must be removed, shut down and made safe. At the appropriate stage, the Secretary of State issues a statutory notice requiring (typically) the field operator to submit to it a “Decommissioning Programme”. Once the Secretary of State has approved the Programme, the operator can, and indeed must, implement it.
5. There are detailed guidelines regarding decommissioning published by various bodies, including the UK Government and the United Nations. In general, those who are licensed to operate in the North Sea must leave the relevant site “as was” when production has ceased. The cost and complexity involved in discharging that obligation varies with the precise installations and items to be decommissioned.
6. The following summary is based on the statement of agreed facts produced by the parties. It records the position at the relevant period, namely the financial year ended 31 March 2009.
7. Marathon Oil Corporation (“MOC”) was the parent company of the Marathon Oil group, and was resident in the US for all tax purposes. The Marathon Oil group was engaged in the exploration for and production of petroleum and natural gas around the world.
8. The Appellant, MOUK, was the main operating company of the group in the UK, and was a wholly-owned indirect subsidiary of MOC. MOUK was incorporated in Delaware but resident in the UK for UK tax purposes.
9. MOUK carried on activities qualifying as a “ring fence trade” in the UK for UK tax purposes. MOUK beneficially owned interests in Block 16/7a and the East Brae filed in the North Sea (“the Brae fields”). In industry parlance, MOUK was both a “participator” in and “operator” of the Brae fields under the terms of the agreements between the various participators in those fields.
10. Under the Petroleum Act 1998, the participators in the Brae fields, including MOUK, were jointly and severally liable for the decommissioning of all installations at the Brae fields.
11. MOUK had a year-end for accounting purposes of 31 December.
13. MODS was US incorporated and UK resident for UK tax purposes, and filed its first UK corporation tax computations in respect of the period 18 December 2008 to 31 December 2009.
14. Marathon International Petroleum (G.B.) Limited (“MIPGB”) was a wholly-owned indirect subsidiary of MOC which was UK resident for tax purposes. MIPGB provided services to other companies in the Marathon Oil group relating to the exploration for and production of oil and gas, the decommissioning of oil and gas facilities, and associated technical and administrative services.
15. Marathon Service (G.B.) Limited (“MSGB”) was a wholly-owned indirect subsidiary of MOC which was UK resident for tax purposes. MSGB carried out the business of recruiting and training administrative, technical, professional and clerical personnel who were qualified to work in the oil and gas industry, and it supplied the services of those personnel to other companies in the Marathon Oil group.
16. On 18 December 2008 the following transactions took place (all references to dollars being to US dollars):
(a) MOUK subscribed for one share of common stock in MODS, with a par value of $1, for a total consideration of $30 million.
(b) MOUK and MODS entered into an agreement relating to the provision of decommissioning services (the Decommissioning Services Agreement or “DSA”).
(c) MOUK paid $300 million to MODS pursuant to Clause 3.2 of the DSA as “the Initial Brae Consideration”.
(d) MODS entered into a loan agreement with MOC under which MODS loaned $329,750,000 to MOC on the terms set out in the loan agreement. The loan was a “permitted investment” by MODS for the purposes of Clause 8 of the DSA.
(e) MODS entered into a services agreement with MIPGB (“the MIPGB Services Agreement”) pursuant to which MIPGB undertook to provide suitably qualified employees and associated technical and administrative services to MODS in relation to MODS decommissioning activities.
(f) MIPGB entered into an employee services agreement with MSGB (“the Employee Services Agreement”) under which MSGB agreed to provide MIPGB with suitably qualified personnel to carry out or supervise decommissioning activities, and with associated technical and administrative services.
17. On 31 December 2009 MOUK filed its company tax return for its accounting period ended 31 December 2008. In that return, MOUK made an election to have a “special allowance” made to it under section 164 of the Capital Allowances Act 2001 (“CAA 2001”) in respect of the payment to MODS of the IBC.
18. The effect of that election was to reduce by $300 million the amount of MOUK’s taxable profits for its accounting period ended 31 December 2008, both for the purposes of ring fence corporation tax and the “supplementary charge” on ring fence profits.
19. In March 2009 MOUK made disclosure to HMRC that it had entered into the DSA, and provided HMRC with copies of the relevant documentation for the transactions.
20. On 5 July 2010 HMRC opened an enquiry into MOUK’s return for its accounting period ended 31 December 2008. The enquiry was opened under paragraph 24(1) of Schedule 18 to the Finance Act 1998 (“Schedule 18”).
21. On 28 September 2015 HMRC issued a closure notice under paragraph 32 of Schedule 18. HMRC’s conclusion stated as follows:
“ … the amount of US $300m paid to [MODS] in the period does not constitute expenditure on which capital allowances are due in computing profits for the purposes of the charge to either Corporation Tax or the Supplementary Charge.”
22. The effect of that closure notice in respect of MOUK’s accounting period ended 31 December 2008 was to:
(a) increase MOUK’s profits for the purposes of ring fence corporation tax and the supplementary charge by £158,980,826 (being $300 million converted at the applicable rate);
(b) increase MOUK’s liability to ring fence corporation tax (charged at a rate of 30%) by £47,694,248; and
(c) increase MOUK’s liability to the supplementary charge (charged at a rate of 20%) by £31,796,165.
23. Accordingly, the total amount of tax at issue in the appeal for the accounting period ended 31 December 2008 is £79,490,143 excluding interest.
24. On 23 October 2015 MOUK appealed against HMRC’s closure notice. On 29 January 2016 HMRC upheld its decision following a statutory review. On 26 February 2016 MOUK filed a notice of appeal against the review decision with the Tribunal.
25. MOUK maintains that the payment of $300 million under the DSA entitled it to the “special allowance” for the year ended 31 December 2008. HMRC argues to the contrary.
26. The only issue is whether the payment made by MOUK to MODS was expenditure “incurred on decommissioning plant or machinery”.
27. HMRC’s primary argument is that, as a matter of statutory construction and on the facts, that payment was expenditure incurred not on decommissioning but on setting aside funds for future costs. HMRC also makes a subsidiary argument that the payment by MOUK to MODS gave rise to a Quistclose trust, with the result that for the relevant accounting period MOUK retained the beneficial interest in the funds paid.
28. There are a number of issues which are not in dispute in this appeal.
29. First, it is agreed that at 18 December 2008 MOUK was carrying out a ring fence trade in the UK as defined in the legislation.
30. Secondly, if (which HMRC do not accept) the IBC payment was incurred on decommissioning, HMRC accept that the election made by MOUK was otherwise valid and satisfied the relevant statutory requirements.
31. Thirdly, it is agreed that expenditure incurred on decommissioning oil and gas installations is capital expenditure for capital allowance purposes in this case.
32. Finally, HMRC do not seek to argue that MOUK’s ability to claim the special allowance for its accounting period ended 31 December 2008 is affected by subsections (5) or (6) of section 5 CAA 2001. The reasons for this are explained at  and  below.
33. The Quistclose argument is dealt with separately, from  onwards.
34. The central issue—whether the payment by MOUK to MODS was expenditure incurred on decommissioning—is a mixed question of law and fact.
35. In determining that issue I have been guided by the two-step approach suggested by Lord Nicholls ( on behalf of the Judicial Committee of the House of Lords) in Barclays Mercantile Finance Ltd v Mawson  STC 1, at :
“… the two steps which are necessary in the application of any statutory provision: first, to decide, on a purposive construction, exactly what transactions will answer to the statutory description and secondly, to decide whether the transaction in question does so. As Ribeiro PJ said in Collector of Stamp Revenue v Arrowtown Assets Ltd  HKCFA 46, at :
“[T]he driving principle in the Ramsay line of cases continues to involve a general rule of statutory construction and an unblinkered approach to the analysis of the facts. The ultimate question is whether the relevant statutory provisions, construed purposively, were intended to apply to the transaction, viewed realistically.” ”
36. As I stated in White v HMRC  UKFTT 802(TC) in endorsing Lord Nicholls’ two-step approach ( at ):
“That approach can in my respectful opinion be summarised as “what does the statute mean, and does it apply in this case?” ”
37. In addition to the agreed facts set out above, I have made various findings of fact relevant to the issue in this appeal. Those findings may conveniently be grouped under three headings ( though they were not so grouped by Counsel for either party), namely:
(a) why were the transactions entered into?
(b) what were the contractual obligations? and
(c) how did the arrangements work in practice?
38. In terms of the relevant evidence, I have reviewed a number of documents, discussed below, which shed light on the reasons for the arrangements, as well as the manner in which they were documented. I was also shown documentation illustrating how the arrangements had operated in practice following their implementation.
39. In relation to witness evidence, the fact-finding exercise was made more difficult by MOUK’s decision in that regard.
40. One of Mr Peacock’s central submissions was that “incurred on decommissioning” means “incurred for the purpose of decommissioning”. In support of that submission one might have expected to have the benefit of witness evidence from those individuals within the Marathon group who could best speak to that issue. In particular, why did MOUK enter into the arrangements and pay the IBC to MODS in December 2008? Further, how did the bare bones of the DSA operate in practice by reference to the statutory tests regarding the special allowance?
41. Clearly, the most helpful evidence in these areas could have been provided by those individuals within the Marathon group or its advisers with direct contemporary knowledge of the relevant transactions.
42. However, the only evidence put forward by MOUK was from Catherine Lee Krajicek. Ms Krajicek gave written and oral evidence.
43. I have a number of concerns regarding the proper weight to be attached to several aspects of Ms Krajicek’s evidence. Ms Krajicek joined the Marathon group in 2007 and held many positions of increasing responsibility thereafter. At the time of her witness statement (25 November 2016) and at the time of the hearing her position was “Marathon Oil Group Vice President, Conventional Assets”. Ms Krajicek stated that in this role she was responsible for the group’s operations in several regions, including the UK, as well as for corporate health, environment, safety and security.
44. In my judgment, Ms Krajicek was suitably qualified and well positioned to give evidence regarding oil and gas industry practice; petroleum engineering (in which she holds a degree); the history of the group’s assets and operations, and the engineering aspects of the group’s operational activities, including the process of decommissioning.
45. These areas were dealt with helpfully in Ms Krajicek’s written and oral evidence.
46. The problem is that Ms Krajicek had no involvement with the transactions with which this appeal is concerned. Although her witness statement stated that she was not “directly” involved, Ms Krajicek clarified the position in cross-examination by Mr Nawbatt as follows:
“… You used the words “directly involved” and I just wanted to establish whether you had any involvement in those transactions?
A. No, I did not. At the time of these transactions I was working both on our Bakken assets in North Dakota or the Gulf of Mexico assets.”
47. Ms Krajicek was therefore unable to speak with any direct knowledge to the purpose of the transactions, or their implementation, or indeed their operation in practice in the years immediately following their implementation.
48. Nor did Ms Krajicek profess to have any specialist knowledge regarding UK tax or legal matters. Yet of her 30 page witness statement, slightly more than half purported to give her evidence regarding the detailed legislative position governing decommissioning in the North Sea; the detail of the Brae decommissioning project from 2008 onwards; the UK tax position; the US foreign tax credit position; a summary of the relevant transactions in the appeal, and the accounting, legal and tax effect of the transactions.
49. Ms Krajicek stated that “where necessary I have gathered the information in [my statement] concerning those transactions from colleagues at Marathon who had the closest involvement with them”, and that she had been “assisted” in drafting the evidence regarding tax by her taxation colleagues within the group.
50. Mr Peacock submitted that Ms Krajicek’s evidence regarding the matters I refer to at  should be accepted in full by the Tribunal in the absence of direct challenge by HMRC. It was clear, however, that HMRC did not accept that Ms Krajicek’s evidence in this areas should carry any material weight. I agree. It was apparent in cross-examination that Ms Krajicek’s evidence in those areas simply consisted in repeating what she had been told by various colleagues.
51. In relation to Ms Krajicek’s evidence regarding the decommissioning of Brae at an operational level, there was also a problem, albeit of less magnitude. Ms Krajicek possessed, in my judgment, direct knowledge and experience regarding decommissioning in respect of the group’s operations at a general level. But as regards the specific issue of the decommissioning in this appeal, Ms Krajicek only took over her current role in September 2016, some two months before her witness statement. There must be a limit on the degree and granularity of information which can be reliably obtained and critically evaluated regarding the transactions in this appeal during the 8 years between December 2008 and the date of the appeal.
52. Although not dealt with in her witness statement, the extent of Ms Krajicek’s involvement prior to September 2016 was clarified in cross-examination by Mr Nawbatt as follows:
“… Did you have any involvement with MOUK or MODS between 2008 and taking up your current appointment in September 2016?
A. No, I did not.”
53. In my judgment, the weight to be given to Ms Krajicek’s evidence regarding the detail of the Brae decommissioning project must take into account her lack of any involvement with MOUK or MODS prior to September 2016. Indeed, even since September 2016 Ms Krajicek had had overall rather than day-to-day responsibility for UK operations.
54. Mr Peacock and Ms Krajicek pointed out that in preparing her evidence Ms Krajicek had consulted various colleagues. That is an unsatisfactory response for two reasons. First, some of those colleagues were not themselves involved in the 2008 transactions. Secondly, MOUK could have chosen to call those colleagues, or relevant advisers, as witnesses. Some of them were present during the proceedings before me. MOUK decided, however, not to do so, thereby preventing any opportunity for the Tribunal to hear their evidence or for them to be cross-examined.
55. For all these reasons, in making the findings of fact set out below, as regards items (a) and (b) referred to at , I have given relatively little weight to Ms Krajicek’s evidence as compared to the written contemporaneous evidence. As regards item (c)—how the arrangements worked in practice—I have taken account of Ms Krajicek’s experience in relation to operational matters and decommissioning generally, but have weighed against that the recent nature of her involvement with MOUK and MODS.
56. The question in this appeal is not why MOUK has incurred the expenditure on decommissioning which it has during the last 8 years, and continues to do so. The question is why it incurred some $300 million of expenditure in December 2008, many years in advance of monies leaving the Marathon group to incur decommissioning costs.
57. I find that the reasons for accelerating expenditure in this way are accurately recorded in a number of documents which preceded the transaction.
“I thought it would be helpful to briefly set out the approach I think we should adopt in connection with the UK decommissioning project.
The background to this proposal is that UK tax law does not currently provide effective tax relief for the cost of decommissioning Marathon’s North Sea oil platform. That is because relief is available when incurred, and, at that time, the forecast revenues will have declined significantly…
I understand that the expenditure is likely to be very significant (hundreds of millions).
Marathon Oil have been lobbying for a change in UK tax law to remove the consequence of the current regime which is perceived to be unfair and creates a fiscal barrier to exploiting a field over its pre-tax economic life. It is uncertain whether the desired law will be achieved.
The proposal (in broad terms) is to subcontract the decommissioning project to a group company established specifically for the purpose of designing and implementing the decommissioning project.
The contract would be prepaid to crystallise the tax relief early.
The “cash” would not leave the Marathon group. Effectively, the transactions are on intercompany account.”
59. The next relevant point in the chronology was 12 March 2008. That was when, in response to industry lobbying, changes were introduced by the Finance Act 2008. Those changes assisted effective tax relief for decommissioning costs incurred during the life of a field rather than at its end—in industry terms, “mid-life costs”. The changes removed the condition attached to the tax relief that the decommissioning costs must have been incurred for the purposes of or in connection with the closing down of an oil field, and removed the requirement that the decommissioning must have been carried out to comply with an approved abandonment programme.
60. However, UK tax was not the only concern of the Marathon group. The changes introduced by the Finance Act 2008 had reduced, though not eliminated, the concerns relating to effective UK tax relief. However, as a US-owned group, Marathon was naturally concerned to manage its foreign (non-US) tax credit position in respect of Brae decommissioning costs.
61. By the time the board of directors of MOUK came to consider the detailed Ernst & Young proposal—named Project Fawkes—on 11 December 2008, the focus was on US tax concerns to a greater extent than UK issues per se. A note prepared by Ernst & Young titled “Directors’ Briefing Note—Project Fawkes” was presented to and considered by the MOUK board at their meeting on 11 December 2008. So far as relevant, that note stated as follows:
“MOUK as a US corporation is subject to Federal Income Tax (FIT) and is included within the MOC consolidated FIT return. The PRT and RFCT paid by MOUK can be credited against the FIT liability of MOC as a foreign tax credit (FTC) and, due to the pooling of non-US liabilities within the MOC return, results in a lowering of the effective tax rate. However, SCT is not credited in the FIT calculation which results in an increase in the overall MOC tax burden.
In addition to the year-by-year FIT calculations there are rules that permit the carry-forward and carry-back of FTC incurred in a year against the FIT liability in other years (subject to limitation); this feature provides opportunities to “optimize” the MOC tax charge. However, should a FTC be refunded at a later date, for whatever reason, this requires the recalculation of the FIT liability for the year in which the FTC was included; this feature makes planning very difficult and particularly where substantial decommissioning liabilities are forecast to be incurred.
The PRT, RFCT and SCT rules provide for a cost deduction to be taken when decommissioning cost are “incurred”. Since these costs are normally incurred at or towards the end of field life, there is normally little income that can be sheltered. Under current rules for all of the taxes, the losses generated can be carried back and set off against prior years’ income, resulting in a repayment of PRT, RFCT and SCT. These refunds of FTC result in the necessity to recalculate the MOC FIT liability for the years affected often undermining “optimization” in those earlier years and thereby driving the effective rate of MOC upwards…
Decommissioning costs are forecast to commence in 2015 with work on Brae “B” platform decommissioning studies. The main offshore campaign is forecast to run from 2018 through to 2024, with the final work being the onshore dismemberment and disposal of the Brae “A” and East Brae platform jackets.
The above costs will result in substantial refunds of PRT, RFCT and SCT which will require the recalculation of those prior years’ FIT liabilities resulting in an adverse MOC tax effect.
Management of Tax
With a view to managing the above problem, Marathon has consulted with Ernst & Young (E & Y).
Marathon has been advised of a method by which the deduction for RFCT and SCT for the decommissioning costs can be accelerated, thereby eliminating the majority of the FIT FTC refund problems outlined above. In addition, this would reduce the commercial risk of not obtaining the expected tax refund due to possible tax law change that either abolished or reduced the rate that would apply to the loss carry-back.
The acceleration technique involves the establishment of a MOUK owned decommissioning services company that would contract with MOUK to undertake its decommissioning work. The work would be performed under a service contract that would provide for the upfront payment of a substantial portion of MOUK’s forecast inflated decommissioning liability, discounted to the time of payment. The upfront payment would never be repayable by the service company, but it would accrete over time at an agreed guaranteed rate; the services company would invest the money in an appropriate security to cover its accretion exposure and would earn an arms’ length margin, currently 9%, on the services that it provides to MOUK.
Analysis of the above fact pattern indicates that the upfront payment would meet the “incurred” requirement for the purposes of RFCT and SCT deductions; this would not apply for PRT purposes due to certain provisions in PRT law.
Bond Pearce, a law firm, has drafted a services contract which has been reviewed with Queens’ Counsel who has supported the E&Y analysis. It should be noted that the implementation of such an arrangement is not without risk of challenge by HM Revenue & Customs; indeed this is highly likely. However, following much discussion and consideration, we believe that this arrangement is appropriate for the directors of MOUK to consider.”
62. Were there any reasons other than taxation why MODS paid MOUK in December 2008? I find as a fact that there were not. Ms Krajicek confirmed in her evidence that there was no “operational” reason to accelerate expenditure in this way. Indeed, in response to comments from Mr Nawbatt regarding Ms Krajicek’s evidence, Mr Peacock confirmed to the Tribunal as follows:
“Purpose of the transactions
MOUK has always accepted that there was no operational reason for entering into the relevant transactions. MOUK has also stated from the moment that it voluntarily disclosed the relevant transactions to HMRC in February 2009 that the reason for entering into the relevant transactions was to gain certainty for the Marathon Oil group as to its foreign tax credit position for the purpose of US federal income tax. Marathon Oil group considers that this was a commercial reason for entering into the relevant transactions.”
63. I have considered the relative weight given by MOUK to the UK and US tax consequences in its decision to implement the transactions. Of course, the intended UK tax consequence of the transactions—namely to accelerate $300 million of tax relief to 2008—was necessary in order to achieve the US tax objectives described by Ernst & Young in the MOUK board briefing paper. To that extent, the UK and US tax objectives operated hand in hand.
64. The submission from Mr Peacock which I have quoted at  asserts that the US tax objective was “the reason” for the transactions. Ms Krajicek’s evidence was consistent with this, though given her lack of involvement with MOUK until 2016 that evidence is of negligible weight.
65. However, I am not persuaded that the standalone UK tax consequences of the arrangements—regardless of their impact on managing US tax—were not also part of MOUK’s reasons for entering into the transactions. The significant acceleration of the capital allowances relief, coupled with the closing off of any risk of future adverse changes to the UK fiscal regime, must sensibly have been part of MOUK’s rationale. The latter point is explicitly drawn out in the Ernst & Young briefing note (described as a “commercial” risk, though it is difficult to see why the US risk is a tax risk while the UK risk apparently is not). I also note that MOUK, and Ernst & Young on its behalf, continued to acknowledge the importance of the UK tax consequences during correspondence with HMRC subsequent to implementation. To take one example, in the note prepared by Ernst & Young of a meeting held between Ernst & Young and HMRC on 30 November 2012, having referred to the 2008 legislative relaxations, it is stated as follows:
“Despite these legislative changes, considerable uncertainty remained, in particular whether the changes would be restricted or even repealed in the future, the rate at which the expenditure would ultimately be relieved and whether deductions for the purpose of PRT would be abolished. When Marathon considered this uncertainty together with the US tax position, and in particular the interaction with the foreign tax credit position, they decided to proceed with the implementation of the transaction in December 2008. The purpose of this was to ensure certainty on the RFCT and SCT position.”
66. I find as a fact that while the main reason for MOUK entering into the arrangements was to control the group’s tax credit position in the United States, an important element of its reasoning was to obtain the desired UK tax consequences regardless of the US position. There was no reason other than taxation why MOUK entered into the transactions.
67. The agreed facts set out above summarise the relevant transactions. In determining the facts it is necessary to consider the contractual arrangements in more detail.
68. The key document was the Decommissioning Services Agreement or DSA, under which MOUK made the payment to MODS which is the subject of this appeal. The most important provisions of the DSA were as follows:
(a) Under Clause 2.1 MODS agreed to provide “Services” to MOUK. Services were defined by Clause 1.1 to include “Decommissioning Services” and “Other Services”.
(b) Under Clauses 2.1 and 3.1 MOUK was liable to pay consideration for the Services consisting of the Initial Brae Consideration or IBC and certain further sums.
(c) The IBC was defined by Clause 1.1 as an amount of $300 million.
(d) Under Clause 3.2 MOUK was liable to pay the IBC to MODS on execution of the DSA.
(e) Clause 3.3 expressed the IBC to be payable “for and in respect of the Brae Decommissioning Services”. Brae Decommissioning Services were defined by Clause 1.1 as “Decommissioning Services in respect of the Brae Specified Plant and Machinery”.
(f) Brae Specified Plant and Machinery was defined as “the plant and machinery listed in Schedule 6, Part A, as such Schedule is updated from time to time on a basis to be agreed between MOUK and MODS”. I discuss at [ ] onwards whether Schedule 6 indeed identifies specific plant and machinery.
(g) Clause 1.1 stated that Decommissioning Services means:
(i) demolishing plant and machinery;
(ii) preserving plant and machinery pending its reuse or demolition;
(iii) preparing plant and machinery for reuse; and
(iv) arranging for the reuse of plant and machinery where appropriate,
where such plant and machinery is, or forms part of, an offshore installation or submarine pipeline and in connection with the Decommissioning of a Field and to the extent that such plant or machinery has been used for a ring fence trade.
(h) Clause 3.4 stated that the IBC was non-refundable.
(i) The IBC was expressed to form part of a contractual balance called the Brae First Payment (“BFP”). For every month that the BFP continued to be in positive balance, an amount was added to it, termed the Brae Accretion Amount, as detailed in Schedule 4.
(j) The DSA set out detailed provisions for the billing of the Services supplied by MODS to MOUK. The costs of any Brae Decommissioning Services supplied by MOUK were deducted from the balance of the BFP, and once the balance of the BFP reached zero MOUK became liable to pay additional amounts in respect of any further such services.
(k) Under Clause 8.1 MODS was obliged to invest the BFP in and only in certain defined “Permitted Investments”.
(l) Under Clause 2.2 MODS was obliged to provide Services “in accordance with the applicable abandonment programme, from time to time...” The then abandonment programme for the Brae fields was in Schedule 7, and Clause 2.4 envisaged revisions to that programme.
69. On 18 December 2008 MODS used the IBC to part finance a loan to MOC of $329,750,000. The loan agreement was a “Permitted Investment” for the purposes of the DSA.
70. Again on 18 December 2008 MODS entered into the MIPGB Services Agreement and MIPGB entered into the Employee Services Agreement with MSGB, as described more fully at .
71. MOUK submits that on 18 December 2008 when it paid the IBC to MODS it thereby incurred $300 million on decommissioning plant or machinery.
72. In viewing the facts realistically, it is necessary not only to address the documents executed on 18 December 2008, but also to consider how those arrangements operated in practice following their implementation.
73. I have therefore considered the nature of the legal relationship between MOUK and MODS; the movements of funds; accounting issues, and when decommissioning costs actually arose and were invoiced under the DSA.
74. The DSA is silent as to the legal relationship between MODS and MOUK, and between MOUK and the Brae joint venture partners. Ernst & Young initially described the relationship between MOUK and MODS as a proposal to “subcontract” the decommissioning project: see . In its Project Fawkes tax opinion dated 4 December 2008, however, Ernst & Young set out a number of possible options. Mr Peacock submitted that the relationship was that MODS acted as agent for MOUK as a participator, rather than operator, in Brae.
75. Further, it appears that the consent of MOUK’s joint venture partners was not sought for the arrangements, and they were not even informed of them until several months following their implementation.
76. It is unsatisfactory that I was not presented with sufficient evidence properly to determine the precise nature of the relationship between MODS and MOUK. Having said that, HMRC do not seek to challenge that the arrangements were legally effective-—whatever their tax consequence—so the relevance of that issue in this appeal is confined to its impact on the question of what MOUK incurred expenditure on, or, expressed another way, when it incurred it, to which I return in considering that question below.
77. As to the movement of funds, I find that, as Ernst & Young made clear in their advice in 2007 and again in the MOUK board briefing paper, in December 2008 the monies did not leave the Marathon group. The IBC was received by MODS and the vast majority of those funds plus the $30 million subscribed by MOUK for share capital was promptly lent by MODS to its ultimate US parent. Unless and until MODS billed MOUK for Services, the balance effectively accreted to the benefit of MODS, and therefore the Marathon group.
78. I have reviewed various documents, including the accounts of MOUK and MODS, in determining the accounting treatment. I find as follows.
79. As regards MODS, although (as a US incorporated company) its accounts were drawn up under US GAAP, they were also drawn up under UK GAAP for UK tax purposes. On receipt the IBC was not recognised as a profit but credited to the balance sheet as a liability, described and noted under the heading “ Performance Obligation”. Accretion to the creditor balance was recognised as an expense in the profit and loss account, offsetting the interest earned by MODS on the on-loan to MOC. As and when MODS incurred expenditure covered by the IBC in subsequent periods, the balance for the Performance Obligation was amended, and an appropriate proportion of the IBC balance was transferred to profit and loss account.
80. As regards MOUK, a balance sheet debit was made to reflect payment of the IBC. In MOUK’s accounts for the period ended 31 December 2008 under the heading “Accounts receivable and prepaid expense” there is a note describing the DSA arrangements. Subsequent entries reflected the accretion of the IBC as provided for in the DSA, resulting in annual debits to MOUK’s balance sheet and credits to its profit and loss account.
81. The intercompany accounting entries between MOUK and MODS effectively netted off on consolidation.
82. In determining how the arrangements worked in practice, I have considered the written evidence and, with the caveat mentioned above regarding prior involvement, the evidence of Ms Krajicek.
83. I find that the contractual obligations as described above were respected in practice. The only exception to this is that the obligation in the DSA on MODS to execute a floating charge in favour of MOUK was not carried through.
84. When and how did the actual costs of decommissioning Brae arise? When did MODS invoice MOUK for decommissioning costs under the DSA?
85. In practice, the first decommissioning costs for a field often relate to feasibility studies and planning. That was the case here, with some studies costs incurred by MODS between 2009 and 2012. However, those costs, while invoiced by MODS to MOUK, were not allocated to the IBC. That was because the DSA definitions partly tracked the statutory requirements for costs to qualify as eligible for tax purposes, and in those earlier years HMRC did not accept costs on studies as eligible for tax relief. By 2016, when HMRC practice had changed, the parties amended the DSA by side letter to state that thereafter such costs were to be allocated to the IBC.
86. The first costs of decommissioning which were set against the IBC and invoiced under the DSA arose in 2012. To give an idea of how the total billed costs compared to the $300 million paid in December 2008, including both those allocated to the IBC and the studies costs not so allocated, the amounts were as follows. Approximately $5 million was billed in 2012; $14 million in 2013; $4 million in 2014; $7 million in 2015, and $14 million in 2016. The forecast prepared by MOUK projects approximately $29 million of costs for 2107.
87. Thus, between December 2008 and the end of 2016, slightly less than $50 million of actual decommissioning costs ( regardless of eligibility for tax relief) had been incurred by MODS on behalf of MOUK.
88. Unless stated otherwise, references below and in this judgment are to the CAA 2001 and to the provisions as they were in force at the relevant time, namely 18 December 2008.
89. Section 162 provides that a “ring fence trade” is a separate qualifying activity for the purposes of the provisions of CAA 2001 which deal with plant and machinery. It states as follows:
“162 Ring fence trade a separate qualifying activity
(1) If a person carries on a ring fence trade, it is a separate qualifying activity for the purposes of this Part.
(2) In this Chapter “ring fence trade” means activities which-
(a) fall within the definition of “ oil-related activities” in section 16(2) of ITTOIA 2005 or within any of paragraphs (a) to (c) of section 492(1) of ICTA ( oil extraction activities, the acquisition, enjoyment or exploitation of oil rights, etc.), and
(b) constitute a separate trade (whether as a result of section 16(1) of ITTOIA 2005 or section 492(1) of ICTA or otherwise).”
90. Section 163 defines “general decommissioning expenditure” and the meaning of “decommissioning”. It provides as follows:
“163 Meaning of “general decommissioning expenditure”
(1) Expenditure is “general decommissioning expenditure” for the purposes of sections 164 and 165 if the conditions in subsections (3) and (4) are met.
(2) But that is subject to subsections (4ZA) to (4ZC).
(3) The expenditure must have been incurred on decommissioning plant or machinery-
(a) which has been brought into use for the purposes of a ring fence trade, and
(i) is, or forms part of, an offshore installation or a submarine pipeline, or
(ii) when last in use for the purposes of a ring fence trade, was, or formed part of, such an installation or pipeline.
(4) The plant or machinery must not be replaced.
(4ZA) An amount of general decommissioning expenditure determined in accordance with subsection (1) is to be reduced under subsection (4ZB) if it appears that the decommissioned plant and machinery-
(a) was brought into use partly for the purposes of the ring fence trade and partly for the purposes of another trade, or
(b) was brought into use wholly for the purposes of the ring fence trade, but has, at any time since, not been used wholly for those purposes.
(4ZB) The amount determined in accordance with subsection (1) is to be reduced to an amount which is just and reasonable having regard to the relevant circumstances.
(4ZC) The relevant circumstances include, in particular, the extent to which the decommissioned plant and machinery has not been used for the purposes of the ring fence trade.
(4A) In this section “decommissioning”, in relation to any plant or machinery, means-
(a) demolishing the plant or machinery,
(b) preserving the plant or machinery pending its reuse or demolition,
(c) preparing the plant or machinery for reuse, or
(d) arranging for the reuse of the plant or machinery.
(4B) In determining whether expenditure is incurred on preserving plant or machinery pending its reuse or demolition, it is immaterial whether the plant or machinery is reused, is demolished or is partly reused and partly demolished.
(4C) In determining whether expenditure is incurred on preparing plant or machinery for reuse, or on arranging for the reuse of plant or machinery, it is immaterial whether the plant or machinery is in fact reused.
(5) In this section-
(a) “oil field” has the same meaning as in Part 1 of OTA 1975, and
(b) “offshore installation” and “submarine pipeline” have the same meaning as in Part IV of the Petroleum Act 1998.”
91. Section 164 deals with the “special allowance” for decommissioning expenditure which is relevant in this appeal. It provides, so far as relevant, as follows:
“164 General decommissioning expenditure incurred before cessation of ring fence trade
(1) If a person carrying on a ring fence trade incurs general decommissioning expenditure, and the plant or machinery concerned has been brought into use for the purposes of that trade, he may elect to have a special allowance made to him.
(2) The election-
(a) must be made by notice to an officer of Revenue and Customs no later than 2 years after the end of the chargeable period in which the general decommissioning expenditure is incurred, and
(b) is irrevocable.
(3) The election must specify-
(a) the general decommissioning expenditure to which it relates, and
(b) where the plant or machinery concerned has been or is to be demolished, any amounts received for its remains.
(4) If a person makes an election under this section-
(a) he is entitled to a special allowance for the chargeable period in which the general decommissioning expenditure is incurred…
(5) The amount of the special allowance for a chargeable period is equal to so much of the general decommissioning expenditure to which the election relates as is incurred in that period.
(6) If plant or machinery is demolished, the total of any special allowances in respect in respect of expenditure on decommissioning the plant or machinery is reduced by any amount received for the remains of the plant or machinery.
Here “decommissioning” has the meaning given by section 163(4A).”
92. Section 5 deals with when capital expenditure is incurred, and, so far as relevant, provides as follows:
“5 When capital expenditure is incurred
(1) For the purposes of this Act, the general rule is that an amount of capital expenditure is to be treated as incurred as soon as there is an unconditional obligation to pay it.
(2) The general rule applies even if the whole or a part of the expenditure is not required to be paid until a later date.
(3) There are the following exceptions to the general rule…
(5) If under an agreement an amount of capital expenditure is not required to be paid until a date more than 4 months after the unconditional obligation to pay has come into being, the amount is to be treated as incurred on that date.
(6) If under an agreement-
(a) there is an unconditional obligation to pay an amount of capital expenditure on a date earlier than accords with normal commercial usage, and
(b) the sole or main benefit which might have been expected to be obtained thereby is that the amount would be treated, under the general rule, as incurred in an earlier chargeable period,
the amount is to be treated as incurred on the date on or before which it is required to be paid.
(7) This section-
(a) is subject to any provision of this Act which has the effect that expenditure is to be treated as incurred on a date later than would result from the application of this section…”
93. The point of construction in this appeal is a narrow one. In determining whether the payment of the IBC by MOUK was “expenditure…incurred on decommissioning plant or machinery” within section 163, HMRC accept that the payment did give rise to the incurring of expenditure on 18 December 2008. At that point, accept HMRC, the obligation to pay the $300 million became unconditional, meaning that expenditure was incurred at that time within section 5 CAA 2001.
94. HMRC also accept that section 5(5) did not operate to defer the date on which the expenditure was incurred, because under the DSA the unconditional obligation to pay and the requirement to pay both arose on 18 December 2008.
95. HMRC further accept that section 5(6) did not defer the date on which the expenditure was incurred, because, even if the subsection was otherwise engaged, the date on which the expenditure was “required to be paid” (in the closing words of section 5(6)) was again 18 December 2008.
96. So, in construing section 163 the question is solely whether the IBC payment was incurred “on decommissioning”. In determining that question, it is first necessary to construe the statute purposively, before applying that construction to the facts.
97. In my respectful opinion, the most helpful description of the modern approach to statutory construction remains that contained in UBS AG v Revenue and Customs Commissioners  1 WLR 1005. Lord Reed, delivering the judgment of the court, described the approach as follows, at  :
“61. As the House of Lords explained in Barclays Mercantile Business Finance Ltd v Mawson  1 AC 684, in a single opinion of the Appellate Committee delivered by Lord Nicholls of Birkenhead, the modern approach to statutory construction is to have regard to the purpose of a particular provision and interpret its language, so far as possible, in the way which best gives effect to that purpose. Until WT Ramsay Ltd v Inland Revenue Comrs  AC 300, however, the interpretation of fiscal legislation was based predominantly on a linguistic analysis. Furthermore, the courts treated every element of a composite transaction which had an individual legal identity (such as a payment of money, transfer of property, or creation of a debt) as having its own separate tax consequences, whatever might be the terms of the statute. As Lord Steyn said in Inland Revenue Comrs v McGuckian  1 WLR 991, 999, in combination those two features—a literal interpretation of tax statutes, and an insistence on applying the legislation separately to the individual steps in composite schemes—allowed tax avoidance schemes to flourish to the detriment of the general body of taxpayers.
62. The significance of the Ramsay case was to do away with both these features. First, it extended to tax cases the purposive approach to statutory construction which was orthodox in other areas of the law. Secondly, and equally significantly, it established that the analysis of the facts depended on that purposive construction of the statute. Thus, in Ramsay itself, the terms “loss” and “gain”, as used in capital gains legislation, were purposively construed as referring to losses and gains having a commercial reality. Since the facts concerned a composite transaction forming a commercial unity, with the consequence that the commercial significance of what had occurred could only be determined by considering the transaction as a whole, the statute was construed as referring to the effect of that composite transaction. As Lord Wilberforce said, at p 326:
“The capital gains tax was created to operate in the real world, not that of make-belief…”
98. The Supreme Court cited the summary of the Committee in Barclays Mercantile, at :
“63…As Lord Nicholls of Birkenhead said in MacNiven v Westmoreland Investments Ltd  1 AC 311, 320, para 8: ‘The paramount question always is one of the interpretation of the particular statutory provision and its application to the facts of the case.’
As the Committee commented, this is a simple question, however difficult it may be to answer on the facts of a particular case.”
99. The Supreme Court stated, at  and :
“66. The position was summarised by Ribeiro PJ in Arrowtown Assets 6 ITLR 454, para 35, in a passage cited in Barclays Mercantile  1 AC 684, para 36: “The ultimate question is whether the relevant statutory provisions, construed purposively, were intended to apply to the transaction, viewed realistically.”
67. References to “reality” should not, however, be misunderstood. In the first place, the approach described in Barclays Mercantile and the earlier cases in this line of authority has nothing to do with the concept of a sham, as explained in Snook  2 QB 786. On the contrary, as Lord Steyn observed in McGuckian  1 WLR 991, 1001, tax avoidance is the spur to executing genuine documents and entering into genuine arrangements.”
100. During this hearing, the judgment of the Supreme Court was handed down in RFC 2012 plc v Advocate General for Scotland  UKSC 45 (“Rangers”).
101. Mr Peacock and Mr Nawbatt both expressed the view that Rangers did not add anything to the existing body of law on purposive construction. However, while it is true that Lord Hodge, with whom the other Justices agreed, approved (at ) the approach of Lord Nicholls in UBS AG, I regard certain passages in Rangers as relevant in this appeal.
102. First, in approving the approach of Lord Nicholls in Barclays Mercantile as marking “ the definitive move from a generally literalist interpretation to a more purposive approach” ( at ), Lord Hodge refers to the “ real world” context of tax legislation as follows ( at ):
“13. Lord Nicholls (para 34) recognised two features which were characteristic of tax law. First, tax is generally imposed by reference to economic activities or transactions which exist, as Lord Wilberforce said (in WT Ramsay, 326) “in the real world”. In the Court of Appeal in Barclays Mercantile  STC 66, para 66, Carnwath LJ made the same point: taxing statutes generally “draw their life-blood from real world transactions with real world economic effects”.
103. Secondly, Lord Hodge makes three points in relation to the statutory interpretation of charging provisions which, in my judgment, are relevant, with suitable adaptation, to the statutory interpretation of relieving provisions. The Upper Tribunal found those three points of assistance in Farnborough Airport Properties Company v HMRC  UKUT 394 (TCC), at , as do I. Lord Hodge stated, at  :
“15. In summary, three aspects of statutory interpretation are important in determining this appeal. First, the tax code is not a seamless garment. As a result provisions imposing specific tax charges do not necessarily militate against the existence of a more general charge to tax which may have priority over and supersede or qualify the specific charge…Secondly, it is necessary to pay close attention to the statutory wording and not be distracted by judicial glosses which have enabled the court properly to apply the statutory words in other factual contexts. Thirdly, the courts must now adopt a purposive approach to the interpretation of the taxing provisions and identify and analyse the relevant facts accordingly.”
104. I should mention the decision of the Upper Tribunal in HMRC v Trigg  UKUT 165 (TCC), which concerned the interpretation of the capital gains rules relating to qualifying corporate bonds. I understand that the appeal in that case is due to be heard by the Court of Appeal.
105. In my judgment, it would be quite wrong to read the decision in Trigg as meaning that a purposive construction is somehow less appropriate where the relevant statutory language is “prescriptive” or “closely articulated”. Such a reading would confuse the process of construction with its result. A process of purposive construction may well result in an interpretation which tallies with a literal construction. But the task of the court remains to start with and work through the construction purposively. This is made clear by the following passage from the judgment, at  and :
“33. We do not consider that it is possible to identify a principle that merely because legislation is closely-articulated, or prescriptive in nature, it is as a general matter somehow less susceptible to a purposive construction. That may be the conclusion that follows from construing a particular provision purposively, but it is not in itself an inhibition on such construction. There may, as Lewison J described in Berry, be less room for purposive construction to give a different answer from a literal construction, but that can only be discerned by applying a purposive construction. The principle of purposive construction applies to all legislation, whatever its nature or character. The task for the courts or tribunals, in all cases, is to construe the statutory language of a particular provision in its context and having regard to the scheme of the legislation as a whole in order to ascertain and give effect to its purpose. Even within closely-articulated or prescriptive legislation there may be individual provisions which fall to be construed purposively in a way which would be different from a literal construction. The judgment of the Supreme Court in UBS is the most recent example.
34. That is, however, no more than an exercise of construction. Whatever underlying purpose may be identified, it is not the task of the courts to import a different meaning to the provision in question than can properly be attributed to it, merely because of a perception that such a meaning would better suit the purpose so identified. That, to adopt the words of Lord Hoffman in his British Tax Review article in 2005, referred to by Lewison J in Berry, would be an exercise in rectification and not construction.”
106. Mr Peacock submitted that incurring expenditure “on” decommissioning means incurring expenditure for the purpose of decommissioning.
107. He cited in support of this proposition three decisions—IRC v Guthrie (1952) 33 TC 327 (CS), Tower McCashback LLP 1 v HMRC  2 AC 457, and Samarkand Film Partnership No 3 v HMRC  STC 926. I deal with these authorities below.
108. In this appeal, Mr Peacock argued, the clear purpose of paying the IBC was to meet MOUK’s statutory obligation to decommission the plant and machinery in Brae. That purpose was obvious from the contractual terms of the DSA. All of the Brae Specified Plant and Machinery met the statutory conditions attached to the special allowance.
109. There was no requirement in the legislation, explicit or implicit, for the type of temporal proximity between expenditure and the performance of services argued for by HMRC in Mr Peacock’s submission. It was the purpose for which the expenditure was incurred which was determinative, and the extent to which services had at that point been performed was quite irrelevant.
110. Mr Peacock submitted that HMRC’s interpretation of section 164 effectively required one to read words into the section, with no purposive construction to support such an approach. That would be an exercise in “rectification”.
111. This reasoning was supported, Mr Peacock argued, by the changes introduced to the special allowance code in the Finance Act 2009. Those changes restricted the amount of the allowance to the costs of an accounting period which related to decommissioning work actually carried out in that period. The changes were in fact prompted by MOUK’s disclosure of the arrangements in this appeal to HMRC. The changes were operative and demonstrated that as at 18 December 2008 the provisions were no so restricted.
112. The arrangements did not, argued Mr Peacock, have the avoidance of UK taxation as their object. Their purpose was to manage the US foreign tax credit position of the Marathon group, and that was a business or commercial purpose.
113. Mr Peacock submitted that a close analysis of the code as a cohesive whole supported MOUK’s analysis.
114. There was nothing surprising in Parliament permitting expenditure on decommissioning to attract relief well in advance of the relevant activities. Decommissioning was a process, and a very long one. If Parliament had wanted to link relief to the services performed, they could easily have chosen to do so, as they had for several years previously in relation to relief for Petroleum Revenue Tax.
115. For HMRC, Mr Nawbatt emphasised the need to view the facts realistically, and to look beyond the contractual arrangements in determining what the IBC was incurred “on”. It was necessary for the Tribunal, in the language of the authorities, to determine what took place in the “real world” and in a “meaningful sense”.
116. In the real world and in a meaningful sense, the IBC was not incurred on decommissioning but on putting funds aside within the Marathon group to meet future costs of decommissioning.
117. HMRC’s view was that in order to qualify for the special allowance the legislation required that specific amounts of expenditure must be identified and spent on specific decommissioning actions in relation to specific items of plant and machinery. Those requirements were not met under the DSA.
118. Critically, submitted Mr Nawbatt, the legislation granted the special allowance only if various conditions were satisfied in respect of the plant and machinery. In this case, it simply was not possible for MOUK or HMRC to know whether those conditions would in fact be satisfied in relation to the actual decommissioning work. There were too many uncertainties and unknowns.
119. Mr Nawbatt submitted that a purposive construction of section 164 indicated that an arrangement such as the DSA was not intended by Parliament to give rise to “expenditure on decommissioning”. Looking at all the facts, and not just the contractual arrangements, in the language of UBS AG the DSA was a provision having no business or commercial purpose other than acceleration of the special allowance.
120. Mr Nawbatt distinguished Guthrie on its facts, noting also that it was distinguished in Ben-Odeco Ltd v Powlson (1978) 52 TC 459. He argued that although Samarkand referred to Guthrie, it emphasised the bona fides of the taxpayer in that case, which contrasted with MOUK’s motives in entering into the DSA.
121. Mr Nawbatt argued that Tower McCashback in fact supported HMRC’s position and not that of MOUK.
122. Mr Nawbatt also referred to the passages in Ben-Odeco indicating that in the absence of a clear indication to the contrary taxpayers in objectively situations should receive similar treatment under the tax legislation. In this appeal, MOUK’s interpretation of section 164 would result in MOUK receiving tax relief for decommissioning many years in advance of its co-venturers in Brae.
123. I will now summarise the relevant case law in so far as it may be relevant to the meaning of incurred “on” for capital allowances purposes.
124. Mr Peacock placed considerable reliance on the decision of the Court of Session in IRC v Guthrie (1952) 33 TC 327 (CS). In that case, Mr Guthrie, a butcher, paid a company for a motor car to use in his business, but the company fraudulently sold the car to someone else so it was never delivered to Mr Guthrie. Adopting a purposive construction of the legislation, which was intended to encourage businesses to invest in plant following the end of the war, the Court held that Mr Guthrie was nevertheless entitled to an initial allowance for expenditure “ on the provision” of machinery or plant. The Lord President (Cooper) stated as follows ( at 330):
“When, as in this case, there has been a bona fide expenditure of capital for an approved purpose, I consider that the Special Commissioners were justified in concluding that their concern was with the fact and object of the expenditure and not with the subsidiary question of whether the money was well spent or ill spent, or whether (bona fides being always assumed) the intended object was or was not actually realised”.
125. In Tower McCashback LLP 1 v HMRC  2 AC 457, the Supreme Court considered whether expenditure qualified as “expenditure on information and communication technology”. The Court confirmed the relevance to that issue of the decision in Ensign Tankers (Leasing) Ltd v Stokes  1 AC 655: see . It held, however, that the Court of Appeal had erred in concentrating on the central issue in Ensign Tankers, namely whether there had been “real expenditure”. The more pertinent question in Tower McCashback was “for what the expenditure was incurred”: . In the words of Lord Walker (at ):
“76. I respectfully consider that Moses LJ was right in deriving assistance from the Ensign Tankers case (paras 78 and 79 of his judgment, quoted in para 62 above) as to the relevance of the terms of the borrowing (here interest-free and non-recourse). But I respectfully think that he was wrong to concentrate on the terms as an indication of whether there was “real expenditure”. That was the issue in the Ensign Tankers case (no real loan, no real expenditure). Here the issue was whether there was real expenditure on the acquisition of software rights. I think that Moses LJ gave the right answer to the wrong question. The transfer of ownership (or at least of rights) indicated the reality of some expenditure on acquiring those rights, but was not conclusive as to the whole of the expenditure having been for that purpose…”
126. Lord Walker concluded his decision that allowances were not available in full by stating as follows (at ):
“80. … The composite transactions in this case, like that in the Ensign Tankers case (and unlike that in the Barclays Mercantile case) did not, on a realistic appraisal of the facts, meet the test laid down by the Capital Allowances Act 2001, which requires real expenditure for the real purpose of acquiring plant for use in a trade…”
127. The Supreme Court in Tower McCashback did not refer to Guthrie. The decision in Guthrie, but not that in Tower McCashback, was referred to by the Court of Appeal in the recent decision in Samarkand Film Partnership No 3 v HMRC  EWCA Civ 77.
128. Samarkand concerned an arrangement for the sale and leaseback of films. One issue related to whether the result of a payment by Samarkand was that it had incurred “acquisition expenditure in respect of the original master version of a film”. The film rights acquired were found to be virtually worthless: see . The Upper Tribunal was divided both in relation to this issue and the relevance to it of Guthrie. The unanimous decision of the Court of Appeal was delivered by Henderson LJ. He determined that although the question was a difficult one, on balance the full amount of the expenditure was eligible. His substantive conclusion on the issue stated as follows (at ):
“110. I also find Guthrie of assistance, because it shows that in answering the question what expenditure is incurred on, in a statutory context designed to provide relief for the expenditure, the focus should be on the fact and the object of the expenditure, rather than on whether the money was well spent…”
129. I mention for completeness two other cases which have dealt with the meaning of incurred “on”. First, The Vaccine Research Limited Partnership v HMRC  UKUT 0389 (TCC) concerned, among many issues, the proper quantum of capital expenditure incurred by a partnership “on” research and development. The Upper Tribunal held (at ) that the question of what, if any, monies were expended “on” research and development was a “factual enquiry”, which must be carried out by reference to a realistic appraisal of the facts. The Tribunal referred in particular to Tower McCashback and Lord Walker’s observations in that case that expenditure which “produces no economic activity” but rather “goes into a loop as part of a tax avoidance scheme” is not expenditure “on” the acquisition of software rights. The Tribunal emphasised that the issue was not what had been incurred, but what it had been incurred on: see .
130. Another decision concerning inflated expenditure, delivered before the release of the judgment in Samakand, was The Brain Disorders Research Ltd Partnership v HMRC  UKUT 176 (TCC). In that case, the Upper Tribunal upheld the decision of the FTT that parts of the relevant arrangement were a sham. However, it considered the issue of the quantum of expenditure incurred “on” research and development on the assumption that there was no sham: . The taxpayer relied in particular on Barclays Mercantile and Tower McCashback. The Tribunal held that in undertaking its “ factual enquiry” ( per Vaccine Research) the movements of money were not outside the scope of that enquiry, even though Tower McCashback made it clear that circular movements of money were not enough in themselves for HMRC to succeed: .
131. The final authority which is of material relevance is the decision of the House of Lords in Ben-Odeco Ltd v Powlson  2 All ER 1111. The issue in that case was whether interest and commitment fees arising on a borrowing to fund the acquisition of plant qualified for capital allowances as “expenditure on the provision of…plant”.
132. The House of Lords held, with Lord Salmon dissenting, that the costs did not so qualify. A factor mentioned by Lord Wilberforce in his judgment, and relied on Mr Nawbatt, was that allowing the financing costs would have been contrary to the general principle that, in the absence of clear contrary direction, taxpayers in objectively similar situations should receive similar tax treatment: . Lord Hailsham made a similar observation, at . The only judge who referred to Guthrie was Lord Hailsham, at , stating that he did not find the case analogous.
133. What is the discernible purpose of the provisions relating to the special allowance for decommissioning expenditure?
134. In my judgment, their clear purpose is to provide a special capital allowance to persons carrying on a ring fence trade for expenditure incurred on one or more of four specified activities relating to plant or machinery, where the plant or machinery satisfies various conditions. The activities are those set out in section 164(4A). The conditions attached to the plant or machinery are those set out at sections 163(3) and (4).
135. The overall purpose of the code is to provide a right to a special allowance for expenditure on certain specified activities relating to plant or machinery, such activities being likely to take place as part of the process of decommissioning the relevant field. Following changes made in 2008 (summarised at ) it was not a policy requirement of the legislation that the activities should occur at the end of the field life; they might take place “mid-life”.
136. In determining whether it is the purpose of the code, or within the purpose of the code, to grant the allowance for a payment such as that under the DSA, it is helpful to consider the legislation in greater detail.
137. Breaking down the language of sections 163 and 164, the special allowance can be seen to be available only where six conditions are satisfied, namely that:
(1) a person is carrying on a ring fence trade (section 164(1));
(2) that person has brought plant or machinery into use for the purposes of that trade ( section 163(3)(a), section 164(1));
(3) that person has incurred expenditure on:
(a) demolishing the plant or machinery,
(b) preserving the plant or machinery pending its reuse or demolition,
(c) preparing the plant or machinery for reuse, or
(d) arranging for the reuse of the plant or machinery ( section 163(4A), section 163(3), section 164(1);
(4) the plant or machinery is or forms part of, or was or formed part of, an offshore installation or submarine pipeline ( section 163(3)(b));
(5) the plant or machinery is not replaced ( section 163(4)), and
(6) that person makes an irrevocable election to HMRC for a special allowance equal to that expenditure ( section 164).
138. I heard many thoughtful and clearly articulated arguments from Mr Peacock and Mr Nawbatt on the detailed workings of the code. In so far as relevant to the issue in this appeal I have reached the following conclusions.
139. First, in terms of the point in time at which the conditions for the allowance are tested and any relief given, the provisions fix this by reference to the time when expenditure on one of the prescribed activities is incurred. It is this point and not, for example, the time when the activity is completed, or when the election is submitted, which is determinative. Section 164(5) states that the amount of the allowance “is equal to so much of the general decommissioning expenditure to which the election relates as is incurred in that period”. Under section 164(4) the entitlement to the allowance arises “for the chargeable period in which the general decommissioning expenditure is incurred”.
140. Secondly, in relation to what the expenditure must be incurred “on”, it must be incurred on one of four specified activities. Those four activities must be carried out in respect of plant or machinery which satisfies the tests I have summarised at . The activities are (broadly) demolishing, preserving, preparing for reuse or arranging for reuse.
141. Thirdly, the relevant connection between the expenditure and the activity is that the former must be incurred “on” the latter. I discuss below what that word imports, but it is significant in my judgment that the draftsman has chosen this rather than any other causal connector. Mr Peacock submitted that there was no difference in terms of effect between “on” and “for the purposes of” or “in connection with”. I disagree. As a general proposition, I would start from the assumption that the draftsman’s choice of causal connector was deliberate. In the context of this code, there are clear contrasts indicating that this was in fact the case. Two examples illustrate this.
142. The first example is the definition of “decommissioning expenditure” contained in section 161B. That is defined as expenditure “in connection with” one of three prescribed activities. The second example is even clearer. The pre-2008 legislation contained a similar definition of “decommissioning” to that applying in this appeal, but at that time the relief was effectively confined to “end of life” fields. The provisions, so far as relevant, began as follows:
“163 Meaning of “abandonment expenditure”
(1) In sections 164 and 165 “abandonment expenditure” means expenditure which meets the requirements in subsections (2) to (4).
(2) The expenditure must have been incurred-
(a) for the purposes of, or in connection with, the closing down of an
(b) on decommissioning plant or machinery…”
143. So, in the pre-2008 formulation, the draftsman has deliberately chosen different causal connectors for sub-paragraphs (a) and (b) of section 163(2). It is also clear that the draftsman perceived a difference in effect between “for the purposes of” and “in connection with”, or he would not have used both terms in sub-paragraph (a).
144. The fourth conclusion I have reached in relation to the code is that it contains no explicit restriction on the special allowance by reference to the decommissioning activity actually carried out in a particular period. Such an explicit restriction was introduced in the Finance Act 2009 for expenditure incurred after 22 April 2009, and is now found in Section 164(5B) to (5D). A similar explicit restriction has existed since 1993 for Petroleum Revenue Tax purposes: section 191 Finance Act 1993 restricts the eligibility of expenditure which is “disproportionate to the extent to which [the other party to the contract] has, at or before that time, performed his obligations under the contract…” to expenditure which is proportionate to the obligations performed.
145. Fifthly, although the code is described as dealing with “general decommissioning expenditure” it clearly operates by reference to specific items of plant or machinery, each of which must satisfy the requirements as to use and non-replacement. There are consistent references to “the” plant or machinery: sections 163(4), 163(4ZA), 163(4B), 163(4C) and section 164(1).
146. Finally, as to whether the provisions contemplate the possibility that the expenditure might precede the decommissioning activity, I conclude that the code lacks consistency as to the chronology which is intended to apply.
147. The opening words of section 163(3) (the expenditure “must have been incurred” on decommissioning plant which “has been brought into” ring fence use) appear backward-looking, contemplating that both the expenditure and the decommissioning must have taken place. However, section 163(3) (b) (i) then switches to the present tense. Section 163(4) then can be read as being forward-looking, in requiring that the plant or machinery “must not be replaced”, but it is silent as to the point or points in time at which this must be the case. Section 164(3) (b) refers to plant or machinery which “has been or is to be” demolished, which does seem to contemplate the possibility of the expenditure preceding the demolition, although again there is no guidance as to when the test is to be assessed.
148. Overall, I do not discern any cohesive structure to the timing envisaged by the draftsman for the various criteria to be satisfied or to continue to be satisfied. It must be recognised, as Lord Hodge stated in Rangers (at ), that “the tax code is not a seamless garment”.
149. I will now consider what guidance can be drawn from the authorities, and how the statute, construed purposively, applies to the facts, viewed realistically.
150. First, in my judgment the authorities support the proposition that in determining what expenditure is incurred “on”, it is necessary to determine the purpose or object of that expenditure.
151. In my opinion, this follows from Tower McCashback, in particular the judgment of Lord Walker (quoted at ). It is also supported by Henderson LJ’s conclusion in Samarkand that “the focus should be on the fact and the object of the expenditure…” (quoted at ).
152. I prefer to rely on those authorities rather than on Guthrie. The facts in Guthrie were unusual, the decision was not referred to in Tower McCashback, and it was described as “not analogous” by Lord Hailsham in Ben-Odeco.
153. The second proposition to be drawn from the authorities is that in determining what expenditure has been incurred on, the focus should be on the purpose or object rather than whether the money was “well spent”. That can be derived not only from Guthrie but more recently from Samarkand.
154. The third point found in the case law is that in an appeal such as this the issue is not whether expenditure, or “real expenditure” has been incurred, but on what that expenditure has been incurred. That is made clear in Tower McCashback and Vaccine Research.
155. The determination of the purpose or object of expenditure is essentially a factual enquiry for the court or tribunal--- I would add, after first having construed the applicable statute purposively. That enquiry takes account of all factors, including movements of money. Although Vaccine Research and Brain Disorders Research are concerned with very different situations to this appeal, in my judgment those observations apply by analogy to situations such as this appeal.
156. In some situations, the purpose or object of expenditure is straightforward. An example would be Guthrie, where the taxpayer was found to be acting bona fides and in effect to have only one purpose or object—to buy a particular car for his business. There is no indication in Guthrie that tax or tax planning considerations formed any part of the taxpayer’s purpose in incurring the expenditure.
157. But what if a taxpayer’s purposes or objects comprise a mixture of tax and non-tax purposes or objects? Where the point at issue is, for instance, whether tax avoidance was one of the taxpayer’s main objects, or whether expenditure was incurred for the purposes of a trade, then there is some guidance in the case law as to the effect of a tax object or motivation. However, we are not concerned with such issues in this appeal.
158. In this appeal, we have a situation where MOUK’s purpose or object in incurring expenditure on decommissioning Brae over the last 7 years was clearly to meet its statutory obligations. But that is not the issue. The issue is MOUK’s purpose or object in December 2008 when it made a pre-payment to its subsidiary of $300 million.
159. I have set out at  to  my findings of fact as to why the transactions in this appeal were entered into. There was no reason in terms of the operation of MOUK’s business, or in terms of its statutory decommissioning obligations, to set up a subsidiary and pay it $300 million in December 2008. The only reason for incurring that expenditure at that time was to accelerate the special allowance under section 164.
160. As set out at  Mr Peacock submitted to the Tribunal as follows:
“Purpose of the transactions
MOUK has always accepted that there was no operational reason for entering into the relevant transactions. MOUK has also stated from the moment that it voluntarily disclosed the relevant transactions to HMRC in December 2009 that the reason for entering into the relevant transactions was to gain certainty for the Marathon Oil group as to its foreign tax credit position for the purpose of US federal income tax. Marathon Oil group considers that this was a commercial reason for entering into the relevant transactions.”
161. I have set out above my findings as to the relative weight given by MOUK to the UK and US tax effects of the transactions. The important point, however, is that, in the formulation of Lord Walker in Tower McCashback, Mr Peacock’s approach runs the risk of giving “the right answer to the wrong question”. The question is not whether the avoidance of “tax” is defined by reference solely to UK tax, or whether achieving the US tax objective which depended on the acceleration of the special allowance was a “business or commercial reason”, or whether the main object of the arrangements was the avoidance of tax. The question is the purpose or object of MOUK in paying $300 million to MODS in December 2008.
162. The nub of the problem is this. The only purpose of incurring $300 million of expenditure in December 2008 was to accelerate the special allowance. The arrangements were described by their architect Ernst & Young as an “acceleration technique” (see ), and were considered on this basis by the board of MOUK in deciding to proceed with them. But, argue MOUK, because the $300 million could only be used under the DSA for services which satisfied the statutory code, the purpose or object of the payment must properly be viewed as qualifying decommissioning services. That, say MOUK, is what the expenditure was incurred “on”.
163. I deal below with whether or not the conditions of the statutory code were indeed satisfied, but let us assume for the moment that they were. Which “purpose” is relevant in determining what the $300 million was incurred on? Is it the acceleration of the special allowance, or is it the ultimate provision, over time, of decommissioning services which have been paid for in advance? In the “factual enquiry” which purpose matters under the code?
164. In considering this question, I have taken account of the case law, the code construed purposively, and the facts viewed realistically. I will deal with each in turn.
165. In relation to the authorities, the propositions which I set out at  to  do not directly answer the question of multiple or conflicting purposes or objects. However, I have been guided in my deliberations on that question by certain passages in Ben-Odeco and Tower McCashback.
166. There is no Court of Appeal judgment in Ben-Odeco, because the appeal proceeded directly from the High Court to the House of Lords under the “leapfrog” procedure. I note two points of interest in the present context in Brightman J’s judgment in the High Court.
167. First, Counsel for Ben-Odeco conceded that in looking at the object of the expenditure ( in that case financing costs on a loan to acquire plant) it must be established that the expenditure not be too remote. See  1 All ER 913 at 920:
“It would often be necessary to apply a remoteness test to a particular item.”
168. Secondly, in relation to my observations regarding the draftsman’s choice of causal connector in sections 163 and 164, Counsel for the Crown raised the following argument, at 920:
“It was submitted that the taxpayer company’s proposition amounted to a rewriting of section 41 of the 1971 Act so as to permit a first- year allowance not merely on capital expenditure incurred “ on the provision” of machinery or plant but also on capital expenditure incurred “ in connection with” the machinery or plant, or the provision thereof.”
169. Brightman J preferred the arguments of the Crown.
170. In the House of Lords, Lord Wilberforce, having observed that the authorities provided no decisive assistance, dealt with remoteness in his conclusion that the financing costs were not incurred on the provision of plant or machinery, at  2 All ER 1111, 1115:
“[The words “expenditure on the provision of”] …focus attention on the plant and the expenditure on the plant—not limiting it necessarily to the bare purchase price, but including such items as transport and installation, in any event not extending to expenditure more remote in purpose. In the end the issue remains whether it is correct to say that the interest and commitment fees were expenditure on the provision of money to be used on the provision of plant, but not expenditure on the provision of plant and so not within the subsection.”
171. Lord Hailsham rejected the Appellant’s reliance on Guthrie and Barclay, Curle & Co Ltd 45 TC 221 as follows, at 1117:
“Neither of these cases really touches the question whether the words “expenditure on the provision of machinery or plant” are wide enough to include money spent on the acquisition of money the main purpose of which was to pay for machinery or plant, as distinct from money actually expended in order to pay for the construction (or purchase) transport and installation of the machinery or plant itself.”
172. Lord Russell expressed the question more pithily, at 1123:
“In my view the question to be asked is, what is the effect of particular capital expenditure? Is it the provision of finance to the taxpayer, or is it the provision of plant to the taxpayer? In my opinion the effect of the expenditure was the provision of finance and not the provision of plant.”
173. Of course, it must be recognised that in this appeal the payment of the IBC did not have the purpose or (in Lord Russell’s terminology) the effect of raising finance for the decommissioning expenditure which was eventually incurred. It is not therefore on all fours with Ben-Odeco. However, it is in my judgment informative to frame the question in this appeal in a similar fashion: was the IBC spent on decommissioning activity or was it spent on putting monies aside in a purpose-designed fund to pay for decommissioning activity at a future date?
174. Ben-Odeco indicates that a more remote purpose or object may in effect rank behind a less remote purpose, at least where the statute in question, as here, adopts a binary test—is the expenditure incurred on the required activity or is it not? Another way of approaching that question, in my judgment, may be found in Lord Walker’s formulation in Tower McCashback at  2 AC 457, at  ( emphasis supplied):
“The composite transactions in this case, like that in the Ensign Tankers case ( and unlike that in the Barclays Mercantile case) did not, on a realistic appraisal of the facts, meet the test laid down by the Capital Allowances Act 2001, which requires real expenditure for the real purpose of acquiring plant for use in a trade.”
175. So, in determining the purpose or object of the payment of the IBC in December 2008, I conclude that I must determine on a realistic appraisal of all the facts the “real purpose” of the payment. A purpose or object which is less remote is to be given priority to one which is more remote.
176. I have set out ( at  to ) my conclusions as to the purpose of the special allowance code, the six conditions which must be satisfied for the allowance to be available, and the principles underpinning the detailed workings of the code. Taking all those factors into account, does a purposive construction of the statute support MOUK’s construction or that of HMRC?
177. In my judgment, the following points are the strongest arguments in favour of MOUK’s construction:
(1) If the purpose of the code is to afford a special allowance for qualifying decommissioning expenditure, the IBC payment does not offend against this purpose. The timing of the allowance is accelerated, but as and when actual decommissioning expenditure is incurred, that expenditure satisfies the statutory conditions. The allowance is available, once and only once, for the type of expenditure contemplated by the code.
(2) There is no explicit restriction in the code which links the allowance to the decommissioning services actually performed at the point when expenditure is incurred. The test looks entirely to the position when expenditure is incurred. The changes introduced by the Finance Act 2009 which introduced such a restriction were operative, and did not apply in December 2008.
(3) The expenditure must have been incurred on decommissioning services which met the statutory conditions, because the DSA restricts the services to be paid for from the IBC to services which satisfy those conditions.
(4) The changes to the code which were introduced in order to facilitate relief for “mid-life” decommissioning show that the relief was intended to be available throughout the (lengthy) decommissioning process.
178. The following points, in contrast, support HMRC’s construction:
(1) The use of “on” as the causal connector points to an intention on the part of the draftsman that there should be a particularly close connection between the expenditure and the relevant decommissioning activities. MOUK’s construction would have the effect that the special allowance was available for expenditure incurred “in connection with” decommissioning activities.
(2) “ Decommissioning” is defined as an activity, so it would not be appropriate to interpret the provisions as being satisfied by expenditure “ on the provision of” decommissioning.
(3) HMRC’s construction would result in MOUK being entitled to the special allowance at the same time as their co-venturers in Brae. That would be consistent with the principle enunciated in Ben-Odeco that taxpayers in objectively similar situations should, absent express contrary provision, be taxed in a similar fashion.
(4) The language to be construed, “incurred…on decommissioning”, is not the type of prescriptive, closely-articulated language dealt with in Trigg. A purposive construction in this case is no more likely to support a purely literal construction than in any other case.
179. In my judgment, it cannot be inferred from the Finance Act 2009 changes that prior to their introduction MOUK’s construction prevailed. As with many provisions which prospectively deal with deficiencies or loopholes in legislation, those who have relied on the deficiency in their tax planning seek to argue that the change proves their reading of the prior legislation, while HMRC seeks to argue that the change is merely clarificatory. In my opinion, all that may reliably be drawn from the change in this case is that the draftsman of the provisions prior to their amendment had not properly contemplated the type of “acceleration technique” devised for MOUK by Ernst & Young.
180. The changes to the code which facilitated relief for “mid-life” decommissioning expenditure addressed a different issue to that in this case. Indeed, on MOUK’s construction, those changes were never needed—and need not have been lobbied for—because well before the closing down of a field relief could always have been obtained for such expenditure by using the advance payment structure adopted by MOUK.
181. As regards the HMRC arguments I refer to at , I find them to have force, save that the Ben-Odeco presumption regarding similar taxpayers is arguably not in point, in that, if MOUK’s argument succeeds, they placed themselves in a different position to their co-venturers.
182. In construing the code purposively, I have concluded that three other points militate in favour of the construction argued for by HMRC.
183. First, MOUK’s construction would have the effect that the special allowance would be obtained on an advance payment to a subsidiary of the sort in this appeal regardless of the time delay between the payment and the actual decommissioning expenditure. Mr Peacock accepted that on MOUK’s construction there was no temporal limit on the effectiveness of the arrangement, so that even if if took 10, 20 or 30 years for the $300 million to be spent on decommissioning, the initial payment nevertheless attracted the special allowance when incurred.
184. Secondly, on MOUK’s construction the allowance would be available even if none of the $300 million was in fact spent on decommissioning, as long as the purpose of the payment when incurred was decommissioning. I questioned Mr Peacock as to the impact on MOUK’s analysis if subsequently some or even all of the $300 million had, unexpectedly, not been spent on decommissioning Brae. This might have happened if, for instance, the $300 million was an over-estimate, or the statutory code changed to remove the relief, or MOUK became bankrupt, or it sold Brae. Mr Peacock’s response was that in such a situation the full $300 million remained eligible for the special allowance.
185. Mr Peacock submitted that this second consequence followed in particular from Guthrie, where the taxpayer in fact acquired no plant at all. I am not persuaded by that argument. The facts in Guthrie are far removed from the tax engineering in this appeal, and in my judgement are not authority that it is irrelevant whether the expenditure incurred is in fact spent on decommissioning services.
186. I regard it as highly unlikely that the draftsman intended that the special allowance should be available for expenditure on services performed 20 or 30 years after the claimed payment, or regardless of whether qualifying decommissioning services in the amount of the payment were performed at all. Such outcomes are so unlikely to have been contemplated and intended by the draftsman that they border on the absurd. As Collins LJ stated in Harding v HMRC  EWCA Civ 1164, at :
“… if a literal construction would lead to injustice or absurdity, and the language admits of an interpretation which would avoid it, then such an interpretation may be adopted… But there may be cases in which the anomaly cannot be avoided by any legitimate process of interpretation…”
187. The third point which in my judgment supports HMRC’s construction is to test the consequence of that construction. The consequence would be that MOUK would, like any other ring-fence taxpayer, be able to claim the special allowance as and when actual (third-party) decommissioning expenditure was incurred. This contrasts with the outcome of the HMRC construction in Guthrie (no tax relief at all for the victim of what the Court described as “the grossest fraud”) and the similar outcome which weighed heavily with Lord Salmon in his dissenting judgment in Ben-Odeco.
188. I have set out earlier what I regard as the state of the authorities on what is meant by a purposive construction, and viewing the facts realistically. As Lord Hodge emphasises in Rangers, part of that approach entails a recognition that tax is generally imposed (and, I would add, reliefs given) by reference to economic activities which exist “in the real world”. Lord Hodge referred with approval to the statement of Carnwath LJ in the Court of Appeal in Barclays Mercantile that taxing statutes “draw their life-blood from real world transactions with real world effects”.
189. In this appeal, it is paramount in viewing the facts realistically always to have in mind the critical question. The question is not whether MOUK incurred expenditure or real expenditure, or whether MOUK’s motives on entering into the arrangements included the avoidance of UK taxation, or whether payment of the IBC was money well spent, or whether services provided following the implementation of the DSA comprised expenditure eligible for the special allowance. The question is what MOUK incurred expenditure on when it paid $300 million to its subsidiary in December 2008. That involves an enquiry as to the “real purpose” of MOUK in making that payment. The facts, whether viewed realistically or not, are relevant only in so far as they illuminate that enquiry.
190. In considering the real world and economic effects of the transactions in this appeal, it must be borne in mind that the architect of the arrangements, Ernst & Young, explained and presented them to the board of MOUK as being an “acceleration technique” as a result of which cash did not leave the Marathon group. The precise basis of the legal relationships arising under the arrangements between MODS and other person was in my view opaque. The structure was entirely artificial in that it made no non-tax sense to implement it. The $300 million remained within the group unless and until it was expended years later on real decommissioning services.
191. Those facts are in my judgment relevant to the determination of the real purpose or object of the payment of the $300 million in December 2008. They support my primary finding of fact, which is that the real purpose or object of MOUK in paying that amount on that date was to accelerate the special allowance, and not to meet decommissioning expenditure.
192. I have explained above my findings of fact as regards MOUK’s purpose or object in making the payment. MOUK may also have had a longer-term purpose or object of meeting real-time decommissioning obligations, but I find that any such purpose was clearly more remote, and not MOUK’s immediate, dominant purpose in paying that amount in December 2008. The tax purpose in no way tainted or denatured the long-term more remote purpose, but nor did the long-term more remote purpose operate to alter the real purpose. MOUK was fully aware when it paid the IBC to MODS that Ernst & Young had forecast that the main decommissioning programme would not be likely to take place until 2018 to 2024 ( see  ).
193. The economic effect of the arrangements was simply to put aside monies for future decommissioning costs in an earmarked fund within the group. In the real world, that was their economic effect.
194. I summarise at  and  the actual decommissioning costs met by MOUK between 2008 and 2016. The total such costs, regardless of eligibility for the special allowance, was slightly less than $50 million. If any person had been asked what decommissioning costs MOUK had incurred by 2016, in my estimation they would say slightly less than $50 million, and not $300 million.
195. In view of my finding as to the real purpose or object of MOUK in paying the IBC, it is not necessary for me to determine the separate question of whether or not all the statutory conditions for the relief were satisfied at the time of that payment. I will however do so as that may assist in the event of any appeal.
196. I set out at  the six conditions which must be satisfied in order for the allowance to be claimed. The issue is whether or not MOUK was in a position to know when the IBC was paid whether all of the relevant conditions would in fact be satisfied.
197. I describe at  my conclusion that the point in time at which the relevant conditions in sections 163 and 164 must be satisfied is when the expenditure is incurred.
198. It is not clear from the code how the provisions are to operate in a normal situation, where expenditure is not pre-paid, where a condition is satisfied at the point when expenditure is incurred, but subsequently ceases to be satisfied. In relation to certain of the conditions—for instance, the requirement that the taxpayer is carrying on a ring-fence trade—the draftsman may well have intended that subsequent failure of the condition should have no effect on the allowance. But for others—notably the requirement that the relevant plant or machinery is not replaced—the position is unclear.
199. The DSA attempts to close off failure of the conditions by defining the services which are to be paid for from the IBC in a way which largely mirrors the statutory conditions. There is, however, no requirement in the DSA that the relevant plant or machinery is not replaced. This is recognised as a material risk to the desired tax analysis in the Ernst & Young report on Project Fawkes.
200. I strongly suspect that the draftsman simply did not contemplate, or at least adequately address, the possibility of years or decades passing between the incurring of the expenditure and the activity in respect of the plant or machinery. I find the issue a particularly difficult one, but I conclude on balance that MOUK did satisfy the statutory conditions because it satisfied them, taking account of the terms of the DSA, at the time when it incurred the expenditure. The hiatus in the DSA relating to replacement of decommissioned plant or machinery is not fatal to MOUK’s position in this respect because even in a normal case when decommissioning occurs, the statute is silent as to how the consequences of replacement are dealt with.
201. This is, of course, of no avail if, as I have found, MOUK did not incur expenditure on decommissioning at all in December 2008.
202. HMRC raised a separate argument that MODS was not free to use the IBC because it was subject to an implied Quistclose trust.
203. One might have expected this argument to be relevant to whether or not MOUK had “incurred” expenditure. Indeed, in HMRC’s Statement of Case it was raised in this context. However, Mr Nawbatt confirmed that the argument went only to what the IBC was incurred “on”.
204. It is not necessary for HMRC to succeed on this point in order to succeed in the appeal. Since, however, it was raised and argued, and in the event of an appeal against this decision, I have considered and decided it.
205. In Barclays Bank Ltd v Quistclose Investments Ltd  AC 567 (HL), a loan was made by Quistclose to Rolls Razor specifically for the purpose of paying a dividend which had already been declared, and the money was placed in a separate account at Barclays. When Rolls Razor went into liquidation before paying the dividend, the House of Lords held that the money in the account was held on trust for Quistclose, and Barclays, having notice of the trust, could not set the balance of the account off against Rolls Razor’s other assets.
206. Quistclose trusts must now be understood in the light of the House of Lords decision in Twinsectra Ltd v Yardley  UKHL 12.
207. Twinsectra established that arrangements which had the effect of creating a trust could do so regardless of the subjective intentions of the settlor. In that case, money was lent by Twinsectra to a solicitor on the terms of an undertaking that it would be retained and used solely for the acquisition of property. The House of Lords held that the arrangements gave rise to a Quistclose trust regardless of whether the lender had subjectively intended to create a trust; the proper construction of the undertaking was determinative.
208. Mr Nawbatt argued that on a proper construction of the DSA and taking into account the wider arrangements, a Quistclose trust arose on payment of the IBC. That was because MODS was not free to use that sum as it wished, by virtue of the obligation in the DSA to invest the IBC in and only in Permitted Investments.
209. Mr Nawbatt also pointed out that while the DSA stated that MODS would execute a floating charge in favour of MOUK, such a charge was not in fact executed. This indicated, he argued, that ownership of the IBC had not in fact passed to MODS.
210. A helpful explanation of the judgments in Twinsectra, which I adopt in this decision, is found in the Court of Appeal decision in Bieber v Teathers Ltd  EWCA Civ 1366. For the purposes of disposing of the issue in this appeal, I need only refer to parts of the judgment of Patten LJ (with whom the other judges agreed). The relevant parts are as follows:
“ It is convenient at this stage to summarise the necessary conditions for the imposition of a Quistclose type trust of the kind alleged…
 These principles were reviewed by the House of Lords in Twinsectra Ltd v Yardley  2 All ER 377,  2 AC 164 and the judge directed himself in accordance with the following summary of the law ( 2 BCLC 585 at -):
‘16. First, the question in every case is whether the payer and the recipient intended that the money passing between them was to be at the free disposal of the recipient: Re Goldcorp Exchange  2 BCLC 578,  1 AC 74 and Twinsectra Ltd v Yardley  2 All ER 377 at ,  2 AC 164.
17. Second, the mere fact that the payer has paid the money to the recipient for the recipient to use in a particular way is not of itself enough. The recipient may have represented or warranted that he intends to use it in a particular way or have promised to use it in a particular way. Such an arrangement would give rise to personal obligations but would not of itself necessarily create fiduciary obligations or a trust: Twinsectra  2 All ER 377 at ,  2 AC 164.
18. So, thirdly, it must be clear from the express terms of the transaction (properly construed) or must be objectively ascertained from the circumstances of the transaction that the mutual intention of payer and recipient ( and the essence of their bargain) is that the funds transferred should not be part of the general assets of the recipient but should be used exclusively to effect particular identified payments, so that if the money cannot be so used then it is to be returned to the payer: Toovey v Milne (1819) 2 B&Ald 683 and Quistclose Investments  3 All ER 651 at 654,  AC 567 at 580.
19. Fourth, the mechanism by which this is achieved is a trust giving rise to fiduciary obligations on the part of the recipient which a court of equity will enforce: Twinsectra  2 All ER 377 at ,  2 AC 164. Equity intervenes because it is unconscionable for the recipient to obtain money on terms as to its application and then to disregard the terms on which he received it from a payer who had placed trust and confidence in the recipient to ensure the proper application of the money paid: Twinsectra at … ’ ”
211. Turning to the facts of this case, and considering them in the context of these conditions, it is in my judgment clear that the IBC was not impressed with a Quistclose trust.
212. The DSA does indeed oblige MODS on terms to invest the IBC in and only in Permitted Investments. However, even if the effect of that obligation was that the sums were not at the free disposal of MODS, as to which I am not convinced, I find no basis to determine objectively that “ the essence of the bargain” between MOUK and MODS was that if the IBC could not be so used it must be returned to MOUK. Nor is there any evidence to suggest that “the mechanism by which this is achieved is a trust giving rise to fiduciary obligations on the part of [MODS] which a court of equity will enforce”: Twinsectra at .
213. In my judgment nothing can be objectively inferred in this context from the failure to execute a floating charge as contemplated by the DSA. That failure could have arisen for any number of reasons.
214. Clause 3.4 of the DSA states plainly that the IBC was non-refundable. I asked Mr Nawbatt where I would find the obligation on MODS to return the IBC if it was not used for a specified purpose. His reply was that it should be inferred, but I find no evidence on which to base any such inference.
215. For these reasons, HMRC’s argument on this point fails.
216. For the reasons given above the appeal is dismissed.
217. This document contains full findings of fact and reasons for the decision. Any party dissatisfied with this decision has a right to apply for permission to appeal against it pursuant to Rule 39 of the Tribunal Procedure (First-tier Tribunal) (Tax Chamber) Rules 2009. The application must be received by this Tribunal not later than 56 days after this decision is sent to that party. The parties are referred to “Guidance to accompany a Decision from the First-tier Tribunal (Tax Chamber)” which accompanies and forms part of this decision notice.