Inmarsat Global Limited v The Commissioners for HMRC

What is at issue in these proceedings is whether the appellant, Inmarsat Global Limited (“Inmarsat”), is entitled to capital allowances on costs incurred by the International Maritime Satellite Organisation (“IMSO”), to whose trade Inmarsat has succeeded, on the launch of certain satellites. The case raises questions as to the interpretation and application of sections 61(4) and 78 of the Capital Allowances Act 1990 (“the 1990 Act”) and how those provisions interact.

IMSO was established by the Convention on the International Maritime Satellite Organization (INMARSAT) (“the Convention”) which was entered into on 3 September 1976 and came into force on 16 July 1979. IMSO was headquartered in the United Kingdom and was a body corporate for United Kingdom tax purposes, but by the Inmarsat (Immunities and Privileges) Order 1980 it was exempt from taxes on income and gains.

The present proceedings concern three “second generation” (or “I-2”) and three “third generation” (or “I-3”) satellites which IMSO commissioned and launched in the course of its trade of operating a telecommunications satellite system. As the Upper Tribunal (“the UT”) explained in paragraph 6 of its decision, the same basic contractual structure applied in relation to all six satellites:

The launch contracts in respect of the I-2 satellites were entered into before the contract for their construction was novated in favour of NSM. In contrast, the launch contracts relating to the I-3 satellites were concluded only after Abbey had acquired the benefit of the construction contract and agreed to lease the satellites to IMSO.

The UT said in paragraph 8 of its decision:

The I-2 satellites were launched on dates between October 1990 and December 1991. The I-3 satellites were launched on various dates in 1996.

NSM and Abbey claimed capital allowances on the expenditure they incurred in acquiring the satellites for their (actual or deemed) leasing trades. No such allowances were, however, claimed in respect of the launch costs, those costs having been incurred and paid by IMSO and forming no part of the leasing arrangements.

In 1999, Inmarsat acquired the business and assets of IMSO’s trade in exchange for shares in Inmarsat. As part of these arrangements, Inmarsat gained the benefit and burden of the satellite leases pursuant to novation agreements. It is common ground that Inmarsat thereby succeeded to IMSO’s trade for the purposes of section 78 of the 1990 Act.

Following the succession, Inmarsat sought to claim writing-down allowances in relation to the launch costs which IMSO had incurred. The parties having been unable to agree on the claims, the question whether Inmarsat is entitled to capital allowances in respect of the satellites was eventually the subject of a reference to the First-tier Tribunal (“the FTT”). The FTT (Judge Kevin Poole) answered the question in favour of HM Revenue and Customs (“HMRC”), and the UT (Adam Johnson J and Upper Tribunal Judge Jonathan Richards) dismissed an appeal by Inmarsat. Inmarsat now, however, challenges the decisions of the FTT and UT in this Court.

The statutory regime governing capital allowances is nowadays embodied in the Capital Allowances Act 2001. For the purposes of this judgment, however, I shall take the law to be as it was when IMSO’s business was transferred to Inmarsat in 1999. At that stage, the law relating to capital allowances was primarily to be found in the 1990 Act.

Under the 1990 Act, a person who has incurred capital expenditure on machinery or plant for the purposes of their trade can potentially qualify for capital allowances. The allowances can be of two kinds: first-year allowances and writing-down allowances. Provision for the latter is made in section 24 of the 1990 Act, which is in chapter I of part II of the Act. Section 24 establishes a regime under which a person can become entitled to an allowance equal to 25% of the difference between “qualifying expenditure” and any relevant “disposal value”. So far as relevant, section 24 provides:

“Disposal value” is explained in section 26 of the 1990 Act as follows:

Chapter VI of part II of the 1990 Act contains specific provisions in respect of fixtures. As section 51(1) explains, the chapter applies “to determine entitlement to an allowance under this Part in respect of expenditure on the provision of machinery or plant which is so installed or otherwise fixed in or to a building or any other description of land as to become, in law, part of that building or other land”.

Chapter VII of part II of the 1990 Act, headed “Miscellaneous expenditure”, includes section 61, dealing with “Machinery and plant on lease”. Section 61(4) states:

The other provisions of the 1990 Act which are of particular relevance to the present appeal are to be found in chapter VIII of part II, which comprises sections 73-83 and is headed “Supplementary provisions”. Sections 77 and 78 are concerned with succession to a trade. Section 77 is in point where “a person (referred to … as ‘the successor’) has succeeded to a trade which was until that time carried on by another person (referred to … as ‘the predecessor’)” and “the two persons are connected with each other”. In such a case, the predecessor and successor may elect for machinery or plant to “be treated as sold by the predecessor to the successor at a price which does not give rise to a balancing allowance or balancing charge”. In that respect, section 77 provides:

Section 78 of the 1990 Act applies where there has been no election under section 77. So far as material, it provides:

Sections 113 and 337(1) of the Income and Corporation Taxes Act 1988, to which there is reference in section 78 of the 1990 Act, address respectively a change in the persons carrying on a trade (section 113) and companies beginning or ceasing to carry on a trade (section 337(1)). Section 337(1), which it is common ground applied in the present case, reads:

Reverting to the 1990 Act, section 81 concerns, among other things, the position where a trader uses machinery or plant which was the subject of a gift. So far as material, section 81 provides:

There was also reference in submissions to provisions relating to contributions which are contained in part VIII of the 1990 Act, headed “Supplementary provisions”. Section 154 states:

Section 155 of the 1990 Act, which has effect where section 154 applies, lays down further rules relating to capital contributions. Section 155(3) provides:

Inmarsat’s case is that it is entitled to writing-down allowances in relation to IMSO’s expenditure on launch costs by virtue of sections 61(4) and 78 of the 1990 Act. It contends that, in paying those costs, IMSO incurred capital expenditure “on the provision for the purposes of a trade carried on by [it]” of the satellites “which [IMSO was] required to provide under the terms of [their leases]” and, hence, that the satellites were to be “treated … as belonging to [IMSO]” pursuant to section 61(4). When, Inmarsat maintains, it acquired IMSO’s business and assets in 1999, it succeeded to IMSO’s trade for the purposes of section 78, with the result that property used for the purposes of IMSO’s trade, including its deemed interest in the satellites, was to be “treated as if … it had been sold to” Inmarsat. That being so, Inmarsat was to be treated as having expended the open market value of the interest on acquiring it, and the interest was to be treated as belonging to Inmarsat in consequence of that expenditure. It follows, so the argument runs, that the requirements of section 24 were met and Inmarsat qualified for writing-down allowances.

In paragraph 28 of its decision, the UT identified the following issues as arising between the parties:

The FTT decided Issue 1 in favour of HMRC. Judge Poole said in paragraph 59 of his decision:

On that basis, as Judge Poole noted in paragraph 61, it did not matter whether Inmarsat’s contentions on section 61(4) of the 1990 Act were accepted, but he nevertheless proceeded to consider them. With regard to Issue 2, Judge Poole concluded in paragraph 67:

Turning to Issue 3(a), Judge Poole considered that a distinction fell to be drawn between the I-2 and I-3 satellites. So far as the I-2 satellites were concerned, clause 6 of the agreement to acquire and lease dated 28 September 1988 between NSM and IMSO, with which the master lease agreement required IMSO to comply, included an undertaking by IMSO that it would “use all reasonable endeavours to achieve Successful Injection and Satisfactory Operation in a timely fashion”. Judge Poole said in paragraph 104 that he agreed with counsel for Inmarsat that it was “necessary, in order to comply with that obligation, to incur the costs involved in procuring the launch services for those Satellites”. Judge Poole continued, however:

The documentation in respect of the I-3 satellites contained nothing comparable to clause 6 of the agreement to acquire and lease dated 28 September 1988 between NSM and IMSO, but Mr Kevin Prosser QC, who was appearing with Ms Barbara Belgrano for Inmarsat, as he also did before us, argued that a “requirement to provide” arose from a covenant in the master lease agreement dated 20 December 1991 to “satisfy all Pertinent Laws”. Judge Poole rejected that submission, observing in paragraph 101 of his decision:

Judge Poole also decided Issue 3(b) in favour of HMRC. He said in paragraph 111 of his decision:

Turning finally to Issue 3(c) and the significance of the “tailpiece” of section 61(4) of the 1990 Act, Judge Poole concluded in paragraph 118:

The UT differed from Judge Poole on Issues 2, 3(a) and 3(b), but agreed with him on Issue 1 and so upheld his decision.

The UT’s conclusion on Issue 1 can be seen from paragraph 56 of its decision, where it said:

With regard to Issue 2, the UT said in paragraph 74:

Turning to Issue 3(a), the UT stated in paragraph 93 that it “accept[ed] Inmarsat’s submission that, if the I-3 Satellites had not been launched, IMSO would have been in breach of its obligations under the applicable Lease”, saying among other things:

The UT also determined Issue 3(b) in favour of Inmarsat. It explained as follows:

The UT did not find it necessary to determine Issue 3(c).

Issue 1 raises a question as to the implications of the deeming for which section 78(1) of the 1990 Act provides. Under section 78(1), property which was in use for a trade before a succession and is then used for the purposes of the new trade is to “be treated as if … it had been sold to the successor”. Does that mean, as Mr Prosser contends, that the property is to be deemed to belong to the successor regardless of whether it in fact does so? Or is the position rather, as was maintained by Mr Michael Gibbon QC, who appeared for HMRC with Mr Richard Vallat QC and Mr Ronan Magee, that section 78(1) is concerned merely with valuation and that it has no application unless the successor comes to own the property?

We were referred to various cases in which the Courts have considered the implications of deeming provisions. It is sufficient, I think, to cite from the most recent of them, the decision of the Supreme Court in Fowler v Revenue and Customs Commissioners [2020] UKSC 22, [2020] 1 WLR 2227. Lord Briggs there said at paragraph 27:

In the present case, Mr Prosser argued that it is an inevitable corollary of treating property “as if … it had been sold” that the recipient must be considered to have acquired ownership. A “sale” involves by definition a transfer of title. In circumstances, therefore, where the previous trader owned the relevant property or, independently of section 78, is deemed to have done so, he is necessarily considered to have passed that ownership to the successor. Mr Prosser recognised that that would not be the case if section 78 had said otherwise, but he pointed out that the provision nowhere states that it applies only where the property at issue has come to belong to the successor.

Mr Prosser drew a distinction between the successor’s position and that of the predecessor. He and Mr Gibbon agreed that, for section 78(1) to be applicable, the relevant property must have belonged, or be deemed to have belonged, to the predecessor immediately before the succession. Mr Prosser gave two reasons for this. The first was that a predecessor cannot sensibly be deemed to sell what it does not own. The second was that section 78(1) is designed to achieve tax neutrality, by enabling the successor to step into the predecessor’s shoes for capital allowance purposes. The policy, Mr Prosser argued, is that the successor should qualify for capital allowances where a deemed discontinuance of the predecessor’s trade has given rise to a disposal event requiring the predecessor to bring a disposal value into account, but a deemed discontinuance will be a disposal event only if the property belongs, or is deemed to belong, to the predecessor.

Mr Prosser is of course correct that, for the predecessor to have to bring into account a disposal value, it must be the case not just that the trade has been permanently discontinued, but that the property in question “belong[ed] to [the predecessor] at some time in the chargeable period”: see section 24(6)(b) of the 1990 Act. Nor could it make sense for the predecessor to have to bring into account a disposal value for property which had not belonged to it merely because it had used the property in its trade and the property was now being used by the successor. What, however, is striking is Mr Prosser’s acceptance that section 78(1) applies only where the predecessor owned, or is deemed to have owned independently of section 78(1), the material property when, first, the section does not say so and, secondly, the section deems the property to have been “sold”. It is Mr Prosser’s case, as I have said, that a sale necessarily involves a transfer of title. He contends on that footing that the successor has to be seen as acquiring ownership, but it could equally well be argued that the predecessor must be assumed to have had ownership to pass. Yet, for very good reason, Mr Prosser does not suggest that the deemed sale has the consequence that the predecessor must be taken to have had title and so to have to bring a disposal value into account for property which was never his. A conventional sale would involve a transfer of property from A to B. On the basis of Mr Prosser’s submissions, however, the hypothetical sale posited by section 78(1) would have the consequence that B would be deemed to have gained title but A would not be assumed to have had it.

The true position, I think, is that the successor is no more to be taken to have obtained ownership than the predecessor is to be assumed to have had it in the first place. The role of section 78(1) is to explain how property should be valued where it has passed from the predecessor to the successor without a sale. It is not to deem property to have passed or, more specifically, to deem the predecessor to have had ownership or the successor to have achieved it. The provision does not state that property is to be deemed to belong to the successor, and no such inference can be drawn from property’s being treated “as if … it had been sold”. Section 78(1) has a valuation function. It says nothing about whether section 24(1)(b) is satisfied.

I am reinforced in that view by section 78’s failure to deal with matters which it might have been expected to address if the effect of that provision had been to deem ownership to transfer to the successor. The UT said this on the subject:

With regard to the contrast with sections 61(4) and 155(3) to which the UT referred in paragraph 54 of its decision:

Section 61(4) provides for machinery or plant to be treated as belonging to a lessee “for so long as it continues to be used for the purposes of the trade” and explains that, “as from the determination of the lease, section 24(6) shall have effect as if the capital expenditure on providing the machinery or plant had been incurred by the lessor and not by the lessee”; and

Section 155(3) is explicit that, where the trade is transferred, writing-down allowances are in future to be “made to the transferee, and shall not be made to the transferor”.

Mr Prosser argued that this does not matter. The position, he suggested, is simply that the draftsman saw no need to add to the disposal events for which section 24(6)(c) provides. Under that provision, there is a disposal event if, among other things, a trader loses possession of machinery or plant (section 24(6)(c)(ii)), machinery or plant begins to be used for purposes other than those of the trade (section 24(6)(c)(iv)) or the trade is permanently discontinued (section 24(6)(c)(v)). In the circumstances, Mr Prosser said, the fact that section 78 does not itself spell out when a disposal event is to occur does not reveal a gap in the legislation. It was not necessary for section 78 to specify a point at which machinery or plant is to be treated as ceasing to belong to a successor.

Doubtless, the legislation could be read in the way Mr Prosser suggested if section 78(1) were understood to deem machinery and plant to belong to a successor. However, the fact that section 78(1) says nothing about when any deemed “belonging” is to terminate is still, I think, telling: the draftsman would, as it seems to me, be likely to have said something on the topic if section 78(1) had been intended to deem a successor to have title to property which he did not in fact own, a point to which comparison with sections 61(4) and 155(3) gives extra weight. Further, the draftsman would probably have addressed the valuation issues which could arise. Suppose, for example, that B has continued a trade previously carried on by A using plant to which A retained title and that the plant now ceases to be used for the purposes of the trade on its sale by A to a third party. Would B be required to bring into account the full “price which the … plant would have fetched if sold in the open market” pursuant to section 26(1)(f) regardless of the extent to which he has had the benefit of writing-down allowances and despite the fact that the proceeds of sale will all have gone to A? Or what?

Mr Prosser, however, advanced a number of arguments for reading section 78(1) as deeming machinery and plant to belong to the successor. In the first place, he contrasted section 78(1) with section 77(4), which is stated to apply to property which after a succession “belonged to the successor and was in use for [the purposes of the trade]” (emphasis added). That, Mr Prosser submitted, shows that the draftsman said so in terms where the intention was that property had to belong to the successor. In a similar vein, Mr Prosser pointed out that in section 78(1) the draftsman had spelt out the need for property to be “either in use or provided and available for use for the purposes of the new trade”. If, Mr Prosser argued, it had been intended that the successor should have to own the property as well as use it, the draftsman would have said as much.

However, section 77(4) stipulates not only that the relevant property must have “belonged to the successor” after the succession but that it has to have “belonged to the predecessor” previously and section 78(1) omits any reference to ownership by either the successor or the predecessor. Mr Prosser nonetheless accepts that section 78(1) will not apply unless the property in question belonged to the predecessor or is deemed to have done so independently of section 78. By the same token, it cannot be inferred that the draftsman would have said so if it had been intended that the successor should have to acquire ownership. Further, the fact that section 77(4) differs from section 78(1) in expressing “belonging” requirements is unsurprising in circumstances where the wording of section 78(1) can be traced back to section 60 of the Income Tax Act 1945 whereas that of section 77(4) derives from the Finance Act 1988. Different hands were at work.

Another contention advanced by Mr Prosser was that HMRC’s interpretation of section 78(1) renders it of little use. Where, Mr Prosser said, the trade for the purposes of which capital expenditure has been incurred on the provision of machinery or plant is deemed to have been permanently discontinued without the machinery or plant being sold, there will be a disposal event under section 24(6) with, by virtue of section 26(1)(f), a disposal value equal to “the price which the machinery or plant would have fetched in the open market”. Since section 78(1) deems there to have been a sale netting “the price which that property would have fetched if sold in the open market”, it would add nothing in this respect. Turning to the position of the successor, Mr Prosser pointed out that section 81 addresses a situation where a person “brings into use for the purposes of a trade carried on by him machinery or plant which belongs to him in consequence of a disposition by way of gift” and that, by its own terms, section 78(1) has no application where property is “sold” (including where property is the subject of an “exchange”: see section 150(4)). It follows (so Mr Prosser said) that, were section 78(1) inapplicable unless property were actually transferred to the successor, the provision would apply only in “very narrow and highly unusual circumstances”: where the successor either (a) gives actual consideration otherwise than in the form of money or other property or (b) gives no consideration at all but the transaction is nevertheless not a gift.

For his part, Mr Gibbon, while accepting that section 78(1) would operate only in relatively unusual circumstances, suggested that bankruptcy provided an example. When a bankruptcy order is made, the bankrupt’s estate, including goodwill, vests in the trustee under section 306 of the Insolvency Act 1986 (“the 1986 Act”), and the trustee has power to carry on any business of the bankrupt so far as may be necessary for winding it up beneficially pursuant to section 314(1) of the 1986 Act and paragraph 1 of schedule 5 to that Act. The bankrupt will retain “such tools, books, vehicles and other items of equipment as are necessary to the bankrupt for use personally by him in his employment, business or vocation” (see section 283(2) of the 1986 Act) and “cannot be restrained from carrying on in his own name a similar business, nor from soliciting former customers, nor compelled to covenant not to compete with his former business”: see Muir Hunter on Personal Insolvency at paragraph 3-619. Even so, it may well be that, if a trustee chooses to carry on the bankrupt’s business, section 78(1) will apply in relation to property used in the business. Mr Prosser suggested that the trustee would need the permission of the creditors’ committee or the Court to carry on the bankrupt’s business, but the requirement to that effect formerly to be found in section 314(1) of the 1986 Act was removed by the Small Business, Enterprise and Employment Act 2015 and, even before that, the creditors’ committee could potentially ratify continued trading after the event: see section 314(2) as it was in force up to May 2015.

In any case, Mr Prosser fairly accepted that he could not go so far as to say that HMRC’s interpretation would render section 78(1) entirely redundant. While, therefore, the provision might be applicable no more than rarely, it would not be superfluous.

Turning to the policy underlying section 78(1), Mr Prosser maintained that the purpose of the provision is to ensure that succession is tax neutral and, accordingly, that a successor to a trade can step into the predecessor’s shoes so far as capital allowances are concerned. I agree with Mr Gibbon, however, that it is not evident that Parliament intended a successor to a trade to qualify for capital allowances in respect of property which he has never owned and on which he has incurred no capital expenditure.

In all the circumstances, I agree with Judge Poole and the UT that section 78(1) has no application unless the successor becomes the actual owner of the relevant property (or potentially, as Judge Poole said, is deemed by some other provision to become its owner). Since, in the present case, Inmarsat never acquired ownership of the satellites, Issue 1 and, with it, the appeal as a whole fall to be determined in favour of HMRC. Section 78(1) cannot entitle Inmarsat to writing-down allowances.

The conclusions I have arrived at thus far suffice to dispose of the appeal. The issues relating to section 61(4) of the 1990 Act having, however, been the subject of full argument, I think I should address them.

The UT noted that “neither party was able to provide any satisfactory explanation of the underlying overall policy behind [section 61(4) of the 1990 Act]” and that, as a result, it had “been unable … to derive as much assistance from submissions on the policy or purpose of s61(4) as [it] might have hoped”: see paragraphs 61 and 62 of the decision. In response, Mr Gibbon took us through the origins of section 61(4). As he explained, a provision in comparable terms was introduced by the Finance Act 1971 (“the 1971 Act”). Section 46(2) of the 1971 Act provided:

At that stage, therefore, the legislation did not include, as section 61(4) of the 1990 Act came to, the words “and … the machinery or plant is not so installed or otherwise fixed in or to a building or any other description of land as to become, in law, part of that building or other land, then, if the machinery or plant would not otherwise belong to him”. Those words were inserted into section 46(2) of the 1971 Act by section 59(5) of the Finance Act 1985 (“the 1985 Act”), while section 59(1) of the 1985 Act provided for schedule 17 to the Act to apply to determine entitlement to capital allowances “in respect of expenditure on the provision of machinery or plant which is so installed or otherwise fixed in or to a building or any other description of land as to become, in law, part of that building or other land”. Section 46(2) of the 1971 thus ceased to apply to expenditure on fixtures, which must hitherto have been its prime focus.

Lord Browne-Wilkinson touched on the reasons for the change in Melluish v B.M.I. (No. 3) Ltd [1996] 1 AC 454. He noted at 477 that in Stokes v Costain Property Investments Ltd [1984] 1 WLR 763 the Court of Appeal had “expressed the view that the law as they had found it was not satisfactory” and that, “[a]s a result, Parliament enacted further provisions regulating the right to capital allowances in relation to fixtures in the Act of 1985”. He added, however, at 479 that, “although Stokes v. Costain Property Investments Ltd. was the initiating event, on any view Schedule 17 [to the 1985 Act] goes much wider”.

When the 1990 Act was passed, section 46(2) of the 1971 Act, as amended in 1985, became section 61(4) of the 1990 Act and the fixtures code found in schedule 17 to the 1985 Act was carried forward into chapter VI of part II of the 1990 Act.

Mr Gibbon suggested that it can be inferred that section 61(4) was included in the 1990 Act “just in case it still covered things”. That seems likely to me. At any rate, the explanatory notes in respect of what was enacted as section 70 of the Capital Allowances Act 2001 as part of the Tax Law Rewrite Project explained that the clause was “based mainly on section 61(4) and (8) of the [1990 Act]” and provided “for the rare circumstances in which a lessee has to provide plant or machinery under the terms of a lease but does not own it” (emphasis added).

It can be seen from section 61(4)(a) of the 1990 Act that, for the subsection to apply, a lessee must incur “capital expenditure on the provision for the purposes of a trade carried on by him of machinery or plant”. In this respect, section 61(4) reflects section 24(1)(a), which refers to “a person carrying on a trade [who] has incurred capital expenditure on the provision of machinery or plant wholly and exclusively for the purposes of the trade”.

The question raised by Issue 2 is whether IMSO’s expenditure on launch costs was incurred on “the provision … of … plant” within the meaning of section 61(4)(a). It is common ground both that the launch costs represented capital expenditure and that the satellites were “plant” used for the purposes of IMSO’s trade.

Judge Poole and the UT differed on the point. Judge Poole distinguished the cases to which he had been referred as all “concerned with ancillary costs associated with an acquisition” and observed that any “provision of plant” “must have at its heart the plant itself” and that “simply moving someone else’s plant from A to B … cannot … amount to the ‘provision’ of that plant”: see paragraph 66 of his decision. He therefore did “not consider IMSO to have incurred capital expenditure (in the form of its costs of launching the Satellites) on the provision for the purposes of its trade of those Satellites”: paragraph 67. In contrast, the UT took the view that there could not have been much doubt that the launch costs would have been “expenditure on the provision of” the satellites if IMSO had been the owner and that it made no difference that the satellites were not in the ownership of IMSO: see e.g. paragraphs 74 and 76 of its decision.

We were taken in this context to two decisions of the House of Lords: Inland Revenue Commissioners v Barclay, Curle & Co Ltd [1969] 1 WLR 675 (“Barclay, Curle”) and Ben-Odeco Ltd v Powlson [1978] 1 WLR 1093 (“Ben-Odeco”). In Barclay, Curle, the taxpayer had incurred capital expenditure on the construction of a dry dock and excavation to accommodate it. The House of Lords held by a majority that the dry dock was “plant” (as the taxpayer had contended) rather than an “industrial … structure” (as the Inland Revenue had suggested). It was further held that the excavation costs were expenditure on “the provision of … plant”. As to that, Lord Reid said at 680:

In Ben-Odeco, the taxpayer, whose trade consisted in hiring out an oil rig, had incurred commitment fees and interest in respect of loans which it had taken out to finance the construction of the oil rig and the fees and interest were rightly charged to capital. The taxpayer claimed that the expenditure qualified for capital allowances, but the House of Lords, by a majority, decided otherwise. Having cited a Canadian case concerned with a statute referring to the “capital cost to the taxpayer”, Lord Wilberforce said at 1097:

In the light of these authorities, it is not surprising that Mr Gibbon accepted before us, as he had below, that the costs of launching the satellites would have constituted capital expenditure on “the provision … of … plant” within the meaning of section 61(4) if they had been incurred by NSM and Abbey as the satellites’ lessors. However, Mr Gibbon submitted that, where a person spends money only on the transport or installation of something (which he does not even own), the expenditure would not naturally be described as “the provision … of… plant” and would anyway be “too remote” having regard to the legislation’s purpose. In the present case, Mr Gibbon argued, the payments which IMSO made in respect of launch costs lacked the necessary nexus with “belonging”/ownership. To be relevant, Mr Gibbon said, expenditure has to be on something that is “leading towards ownership in due course”: “freestanding” expenditure will not do, there must be a sufficient connection with acquisition of ownership.

For his part, Mr Prosser asked rhetorically how, say, transport costs could be incurred “on the provision … of … plant” if incurred by an owner but not if incurred by someone else. The nature of the expenditure, Mr Prosser said, is the same in either case. Expenditure on the “provision … of … plant” is not limited to purchase costs, Mr Prosser submitted, regardless of whether the money is spent by an owner, though the legislation imposes a separate “belonging” condition. When considering whether something involved “provision … of … plant”, the focus with both section 61(4) and section 24(1)(a) is on the effect or function of the expenditure, on what it does in connection with the provision of the plant, not on the identity of the person paying.

Plainly, expenditure relating to plant will not give rise to capital allowances unless it results in the plant belonging, or being deemed to belong, to the person incurring it. Section 24(1)(b) imposes a requirement that, “in consequence of [the taxpayer’s] incurring that expenditure, the machinery or plant belongs or has belonged to him”, and section 61(4), where it applies, deems machinery or plant to belong to the lessee. Expenditure on plant must, moreover, be incurred for the purposes of a trade carried on by the taxpayer if allowances are to be available. There is therefore no question of wholly freestanding expenditure qualifying for allowances. Someone who incurs costs without thereby gaining (or being deemed to gain) ownership will not obtain allowances.

On the other hand, I do not see the need for expenditure to be on the “provision … of machinery or plant” as itself imposing any requirement that expenditure should “lead towards ownership”. As Lord Wilberforce explained in Ben-Odeco, the UK legislation focuses “on the plant and the expenditure on the plant”, “confin[ing] the limiting curve to the plant itself”. What matters for the purposes of sections 24(1)(a) and 61(4)(a) of the 1990 Act is thus what role the expenditure played in relation to the relevant plant. To give rise to capital allowances, expenditure must facilitate the “provision” of machinery or plant, but there is no reason to read “provision” as limited to acquiring ownership. Neither section 24(1)(a) nor section 61(4)(a) says anything to that effect, and Barclay, Curle affords authority that “provision” “must cover something more than the actual supply” and, in particular, can cover costs of installation without which plant would be “of no use for the purposes of the trade”. The terms of sections 24(1)(a) and 61(4)(a) indicate that attention should be directed at whether “the provision of machinery or plant” was for the purposes of the taxpayer’s trade, not at whether it was such as to “lead to ownership”. In the circumstances, I agree with Mr Prosser that it is hard to see how costs could be incurred “on the provision … of … plant” if incurred by an owner but not if incurred by someone else.

Mr Gibbon, however, submitted that the approach espoused by Inmarsat would involve “read[ing] into the statute provisions for apportioning the market value between the lessor and lessee which are simply not there” (to borrow words from paragraph 59 of Judge Poole’s decision). In the same vein, as the UT explained in paragraph 83 of its decision, HMRC argued below that section 61(4) “did not operate to create ownership of a part or share in the plant and machinery” but “either deemed the plant and machinery to belong (exclusively) to the lessee, or it did not” and that “since s61(4) contains no mechanism to ascertain the comparative extent of each person’s ownership of the relevant plant, that was a strong indication that it did not envisage any deemed division of ownership of the kind for which Inmarsat argues”.

The UT explained in paragraph 84 of its decision why it saw difficulties with HMRC’s submission:

The UT said in paragraph 85 of its decision that, in the circumstances, “HMRC’s arguments as to the presence of anomalies have not persuaded us that the term ‘capital expenditure … on the provision … of machinery or plant’ in s61(4) of CAA1990 should be construed any differently from similar phrasing in statutory predecessors to s24”. Likewise, I do not think the fact that Mr Prosser’s approach to section 61(4) could generate doubt as to how a lessee’s “disposal value” should be assessed can justify the adoption of HMRC’s interpretation of section 61(4)(a). In the first place, it seems to me, as I have mentioned, that other considerations favour Mr Prosser’s construction of section 61(4)(a). Secondly, section 61(4)(a)’s reference to capital expenditure on the “provision for the purposes of a trade carried on by him of machinery or plant” reflects the similar wording in section 24(1)(a), but the latter provision has much older origins and I cannot see why any issue which might arise in relation to section 61(4) should dictate how a concept which was already well-established by the time a predecessor of section 61(4) was first introduced should be understood. Thirdly, as the UT pointed out, the “contributions code” to be found in sections 154 and 155 of the 1990 Act could give rise to comparable debates. Fourthly, again to echo the UT, with fixtures excluded from section 61(4) “it is difficult to think of a real-world situation where s61(4) would wish to treat the lessee’s ownership as being exclusive”.

In all the circumstances, I agree with the UT’s conclusions on Issue 2.

For relevant purposes, Issue 3(a) can be taken to be whether IMSO was “required” under the terms of the leases relating to them to procure the launch of the I-3 satellites. Before the FTT, HMRC also disputed whether IMSO was under such an obligation in respect of the I-2 satellites, but Judge Poole found in favour of Inmarsat on the point and HMRC did not appeal against the finding. By the time the matter was before the UT, therefore, HMRC did “not challenge the FTT’s conclusion that there was such a ‘requirement’ in relation to the I-2 Satellites” (as the UT noted in paragraph 86 of its decision).

To explain the rival arguments, I need to say more about the contractual documentation relating to the I-3 satellites.

IMSO entered into a contract with General Electric Technical Services Company, Inc (“GETSCO”) on 1 February 1991 for the satellites’ construction and purchase. On 20 December 1991, a novation agreement was concluded as between GETSCO, IMSO and Abbey under which Abbey agreed to buy the satellites in IMSO’s place. On the same day, Abbey and IMSO entered into further contractual documents: a lease facility agreement, a master lease agreement and, in respect of each of the satellites, a lease schedule.

A recital to the lease facility agreement explained that it set out “the terms and conditions and principles and assumptions on and subject to which the Owner [i.e. Abbey] will make a commitment to the Lessee [i.e. IMSO] to purchase Assets and lease them to the Lessee”. The lease facility agreement entitled IMSO to require Abbey to enter into, among other things, the novation agreement, the master lease agreement and lease schedules on the basis that the leasing of each satellite would “commence forthwith upon the Owner acquiring title to and property in, and risk of loss of, that Asset” under the novation agreement. The lease facility agreement defined “Delivery” and “Delivered” to refer to “intentional ignition of the first stage engine of the launch vehicle” for a satellite and, by clause 9.05, if a satellite was not “Delivered” on or before 31 December 1998:

The original contract for the purchase of the I-3 satellites from GETSCO had, by article 28, given IMSO a right to terminate the contract in whole or in part, for its convenience, at any time prior to completion. Under the novation agreement, that right remained with IMSO rather than passing to Abbey. The novation agreement further contained an acknowledgment by the parties that Abbey would not acquire title to any satellite until “Intentional Ignition”.

The master lease agreement provided for the I-3 satellites to be leased by Abbey to IMSO on the terms and conditions contained in the master lease agreement and lease schedules. By clause 4.01, “[t]he leasing of each [satellite] … under this Agreement shall commence on its Delivery”, and “Delivery” was once again defined by reference to “intentional ignition of the first stage engine of the launch vehicle”. Clause 7, headed “OPERATION; MAINTENANCE; COMPLIANCE WITH LAW”, included at clause 7.03 a provision with the heading “Compliance with Law” which stipulated:

The lease schedules in respect of the I-3 satellites reflected the master lease agreement. Each of them stated that Abbey agreed to lease and IMSO agreed to take on lease the satellite in question “on the Delivery Date for that Spacecraft and on and subject to the terms and conditions of the Master Lease Agreement, this Lease Schedule and other Relevant Documents”.

In his oral submissions on Issue 3(a), Mr Prosser pinned his colours to the mast of clause 7.03 of the master lease agreement. As he pointed out, clause 7.03 required IMSO to satisfy the requirements of the Convention, which included the following provisions:

As article 3:

The purpose of the Organization [i.e. IMSO] is to make provision for the space segment necessary for improving maritime communications and, as practicable, aeronautical communications, thereby assisting in improving communications for distress and safety of life, communications for air traffic services, the efficiency and management of ships and aircraft, maritime and aeronautical public correspondence services and radiodetermination capabilities.

The Organization shall seek to serve all areas where there is need for maritime and aeronautical communications.

The Organization shall act exclusively for peaceful purposes”; and

As article 5:

The Organization shall be financed by the contributions of Signatories. Each Signatory shall have a financial interest in the Organization in proportion to its investment share which shall be determined in accordance with the Operating Agreement.

Each Signatory shall contribute to the capital requirements of the Organization and shall receive capital repayment and compensation for use of capital in accordance with the Operating Agreement.

The Organization shall operate on a sound economic and financial basis having regard to accepted financial principles.”

The UT agreed, taking the view, as I have mentioned, that “a point must have come at which it would have made no commercial sense for IMSO to decide not to launch” and that IMSO would at that point have been in breach of clause 7.03 of the master lease agreement if it had not launched the satellites. Judge Poole had disagreed, on the footing that “a generalised obligation to comply with all relevant laws” could not give rise to “a specific obligation to incur expenditure on launching six specific satellites”.

It is not clear to me that clause 7.03 of the master lease agreement applied in advance of a lease taking effect. The master lease agreement provided for the terms and conditions contained in it, including clause 7.03, to be those on which an I-3 satellite “is leased”, and each lease schedule stated that a satellite was to be taken on lease “on and subject to the terms and conditions of the Master Lease Agreement” “on the Delivery Date for that Spacecraft”. It strikes me as well arguable, therefore, that clause 7.03 was to be one of the terms on which a satellite would be leased once it was “Delivered” and that it had no application until then. Were that right, clause 7.03 could not assist Inmarsat since “Delivery” was to take place only on “intentional ignition of the first stage engine of the launch vehicle” or, in other words, when the satellite was launched. On that basis, clause 7.03 would not have operated to oblige IMSO to satisfy the requirements of the Convention until launch had already happened and could not have required IMSO to procure the launch.

Even assuming, however, that clause 7.03 of the master lease agreement was applicable before a satellite was “Delivered”, I do not see it as having imposed on IMSO a requirement to launch the satellites. To my mind, it is implausible to suppose that Abbey would, or could, have invoked clause 7.03 and the Convention if IMSO had not proceeded with the launch of the 1-3 satellites. The Convention brought IMSO into being and served as its constitution, dealing with matters such as its organs, the composition, functions and procedure of each of them, audit requirements and privileges and immunities. In theory at least, there could doubtless have been circumstances in which Abbey might have wished to complain, relying on clause 7.03 of the master lease agreement, that IMSO, or an organ of it, was exceeding its powers. I find it hard to conceive, however, that Abbey could have hoped to establish a breach of clause 7.03 on the basis that IMSO was operating otherwise than “on a sound economic and financial basis”. As I have said, the UT concluded from the fact that the I-3 satellites were ultimately launched that “a point must have come at which it would have made no commercial sense for IMSO to decide not to launch”, but I do not think that is right. At most, it can be inferred that IMSO saw launch as consistent with its purpose and operation “on a sound economic and financial basis”. It cannot be deduced that launch was the only course open to IMSO which could be squared with the Convention, and it strikes me as unreal to imagine that it would have occurred to Abbey to try to show that IMSO’s sole option was to proceed with launch. HMRC submitted that the “generalised obligations described in the … Convention can be satisfied in a myriad of different ways”, adding that it might have been “more in accordance with a requirement to ‘operate on a sound economic and financial basis’ for the lessors to incur the launch costs and reduce the lease rental payments yet further by reference to the capital allowances available to the lessors on these”. I agree that it cannot be assumed, and has not been demonstrated, that operating “on a sound economic and financial basis” in pursuit of its purpose necessarily obliged IMSO to incur the launch costs.

The reality, in my view, is that Abbey would have looked elsewhere for a remedy if launch had not taken place. More specifically, it would have relied on clause 9.05 of the lease facility agreement, under which, if launch of a satellite did not happen on or before 31 December 1998, IMSO was obliged to terminate Abbey’s obligations (in default of which, Abbey could), with the consequence, pursuant to clause 9.02(2), that the purchase contract would be re-novated to IMSO and a “Re-Novation Sum” would be payable.

In short, I have not been persuaded that the terms of the leases of the I-3 satellites “required” IMSO to procure their launch.

The question raised by Issue 3(b) is whether, for section 61(4) of the 1990 Act to apply, expenditure must have been incurred only after the term of the relevant lease had begun. Judge Poole’s answer was in the affirmative, on the basis, as he explained in paragraph 111 of his decision, that “there is a natural chronological flow about [section 61(4)], which necessarily implies that the lease must be in existence before the capital expenditure is incurred”. In this respect, however, the UT differed from Judge Poole, concluding that Issue 3(b) should be determined in Inmarsat’s favour.

Arguing that Judge Poole had been correct, Mr Gibbon pointed out that section 61(4) speaks of a “lessee” incurring capital expenditure. Mr Gibbon relied, too, on section 61(8), which states that “lease” “includes an agreement for a lease where the term to be covered by the lease has begun” (emphasis added) and that “lessee” is to be “construed accordingly”. The word “where”, Mr Gibbon submitted, imposes a temporal requirement and confirms the need for the term to have started before expenditure is incurred.

For his part, Mr Prosser supported the UT’s decision. As used in section 61(8), Mr Prosser said, “where” means “in a case where”, not “when”. While, therefore, section 61(4) will not be in point unless the term of a lease has commenced, the subsection is, Mr Prosser maintained, capable of applying to expenditure incurred before then.

I prefer Mr Prosser’s submissions. The better view, I think, is that section 61(4) operates only if the term of a lease has begun and so the “lessee” has become such, but that expenditure incurred earlier can potentially be relied on. Were the position otherwise, a “lessee” could not claim any allowance in respect of, say, capital expenditure which he had been required by an agreement for a lease to make in advance of the start of the term, which would appear to make little sense. In fact, a “lessee” might not be able to invoke section 61(4) even in relation to sums spent on capital expenditure during the term. By section 159(3), “an amount of capital expenditure is to be taken to be incurred on the date on which the obligation to pay that amount becomes unconditional (whether or not there is a later date on or before which the whole or any part of that amount is required to be paid)”. If, therefore, an agreement for a lease imposed an unconditional obligation to make capital expenditure, it would be deemed to be incurred at that stage, and so before the term, even if the relevant plant or machinery were not in fact provided until later.

The “tailpiece” of section 61(4) of the 1990 Act states:

Mr Gibbon pointed out that, when Inmarsat succeeded to IMSO’s trade, the leases of the satellites were novated and that novation “typically extinguishes the original contract (between A and B) and replaces it by another (between A and C)”: see Chitty on Contracts, 34^th ed., at paragraph 22-092. That being so, Mr Gibbon said, each satellite lease will have been determined for the purposes of section 61(4).

Since, however, section 78 looks to the position “immediately before the succession takes place”, Mr Gibbon did not suggest that section 78 had been rendered inapplicable in the present case by operation of the tailpiece. He rather relied on the tailpiece as a factor bearing on how section 78 should be interpreted (i.e. in relation to Issue 1).

In the circumstances, the UT did not express a view on Issue 3(c). It said in paragraph 106 of its decision, “given the way in which HMRC seek to rely on the tailpiece to s61(4), we do not consider that we need to decide the question of whether the novation of the Leases was a ‘determination’ for the purposes of that tailpiece and we will not do so”.

I, too, see no necessity to comment on Issue 3(c) and prefer not to do so.

I would dismiss the appeal. Had it mattered, I would respectfully have agreed with the UT on Issues 2 and 3(b), but with Judge Poole on Issue 3(a). However, I agree with both Judge Poole and the UT that Issue 1 should be decided in favour of HMRC and that is determinative of the outcome of the appeal.

I am grateful to my Lord, Lord Justice Newey for his clear exposition of the issues in this case. I agree with his conclusions and add only this in relation to Issue 1. The purpose of section 78(1) is to value property which has passed from a predecessor to a successor without a sale (see paragraph 40 above). The mechanism to achieve that valuation is an assumed (but fictitious) sale of that property at open market value by the predecessor to the successor. The central error in Inmarsat’s case lies in seeking to press that statutory fiction further, to deem ownership of the property to have passed to the successor. That is not what section 78(1) provides and is not what the statutory fiction created by section 78(1) was intended to do; nor is it an inevitable corollary of that statutory fiction. Inmarsat’s case offends the principles summarised in Fowler, set out at paragraph 36 above.

Because the satellites do not belong (and never did belong) to Inmarsat, the expenditure in question cannot qualify for capital allowances (section 24(1)(b) of the 1990 Act) and this appeal must be dismissed.

I agree with both judgments.