Shui on Credit Co Ltd v. Commissioner of Inland Revenue
Read the full judgment text of FACV 1/2009 on BabelCite. This Court of Final Appeal judgment was delivered on 30 November 2009 before Chief Justice Li, Mr Justice Bokhary PJ, Mr Justice Chan PJ, Mr Justice Ribeiro PJ and Lord Walker of Gestingthorpe NPJ.
Revenue law – profits tax – Inland Revenue Ordinance (Cap. 112) – general anti-avoidance provision s.61A – May 1988 Shui On tax-defeasance scheme involving refinancing of HSBC syndicated loan on Wanchai East Shui On Centre through Mitsubishi loan of $600m and FPB Finance/Agnew/Centre Co. loan of $1,200m at fixed 9.375% with swap arrangements and assignment of eight-year interest stream to newly-formed taxpayer subsidiary for $600m – taxpayer amortising $606.2m purchase consideration as 'deferred expenditure' over eight years to generate accounting losses – whether deductibility of that expenditure under ss 16 and 17 of IRO could be argued by Commissioner on objection and appeal when assessments were made expressly under s.61A – whether s.61A assessments are separate self-contained charging provisions or merely an aid to ordinary charging provisions – Board of Review's general review function under s.64 and de novo appellate function under s.68 – characterisation of one-off payment for long-duration income-yielding asset as capital expenditure – application of Nchanga and Wharf Properties principles – s.17(1)(c) overriding Companies Ordinance accounting treatment – distinguish Secan, Myer and John Lewis Properties – indicia of capital payment in hands of assignee – hypothetical transaction and dominant-purpose observations under s.61A – whether s.61A requires tax benefit to exist in absence of section – appeal dismissed with costs order nisi to Commissioner.
Legal issues: Whether the Commissioner may rely on ss 14, 16 and 17 on their own after assessments made expressly under s.61A · Whether the $606.2m 'deferred expenditure' was deductible under ss 16 and 17 of the IRO · Whether the taxpayer was liable under s.61A of the IRO
Outcome: Appeal unanimously dismissed; the Court of First Instance and Court of Appeal decisions upholding the Board of Review's dismissal of the taxpayer's appeal were affirmed.
Cited by 12 cases · Cites 6 cases
|
FACV No. 1 of 2009 IN THE COURT OF FINAL APPEAL OF THE HONG KONG SPECIAL ADMINISTRATIVE REGION FINAL APPEAL NO. 1 OF 2009 (CIVIL) (ON APPEAL FROM CACV NO. 85 OF 2008) _____________________ Between :
_____________________
Dates of Hearing : 13 and 16 November 2009 Date of Judgment : 30 November 2009 _____________________ J U D G M E N T _____________________ Chief Justice Li: 1.I agree with the judgment of Lord Walker of Gestingthorpe NPJ. Mr Justice Bokhary PJ: 2.I agree with the judgment of Lord Walker of Gestingthorpe NPJ. Mr Justice Chan PJ : 3.I agree with the judgment of Lord Walker of Gestingthorpe NPJ. Mr Justice Ribeiro PJ : 4.I agree with the judgment of Lord Walker of Gestingthorpe NPJ. Lord Walker of Gestingthorpe NPJ : The Issues 5.In May 1988 a complicated pre-arranged scheme of transactions was effected between several companies in the Shui On group (three of them newly-formed subsidiaries) and several outside companies. The appellant, Shui On Credit Company Limited (“the taxpayer”) was one of the newly-formed subsidiaries. A scheme of that sort was likely to be of interest to the Commissioner of Inland Revenue (“the Commissioner”) as possibly falling within s.61A of the Inland Revenue Ordinance, Cap. 112 (“the IRO”), and so it proved. Profits tax assessments on the taxpayer for the years of assessment 1988-89 to 1996-97 (inclusive) were made in three tranches (six in 1995, two in 1997 and one in 1998). These assessments were all stated to have been made under s.61A of the IRO. 6.Section 61A was enacted in 1986 as a general anti-avoidance measure. It can be applied so as to assess a person to tax only if (among other conditions) a transaction has been effected (s.61A(1)) :
If the supposed tax benefit would not have been achieved even in the absence of s.61A (in colloquial terms, if for more mundane reasons the tax-avoidance scheme simply did not work) then logically s.61A cannot apply, as there is no tax benefit in the statutory sense. In a recent decision, Ngai Lik Electronics Co. Ltd v. CIR, 24 July 2009, FACV No. 29 of 2008, paras 93-97, this Court expressly left open the question whether that proposition is correct. The argument in the present case proceeded, in my view correctly, on the basis that it is correct. The question previously left open should now be taken to have been answered : a tax benefit in the statutory sense is required before s.61A is engaged, and so that section can apply only to a transaction which would otherwise avoid tax. 7.The tax benefit said to have been obtained by the taxpayer was the deduction from its assessable profits, in all nine years of assessment, of sums described as “deferred expenditure” (it remains to be considered whether that label was appropriate). Since the original assessments were made the matter has been more fully explored and the question has arisen whether the so-called deferred expenditure would (quite apart from s.61A) have been a proper deduction, having regard to the terms of ss 16 and 17 of the IRO. 8.In these circumstances there are three issues in the appeal, one turning on questions of administration and procedure and two substantive questions.
The Facts 9.In 1987 the principal asset of the Shui On group was a development in Wanchai East. The site was acquired in 1985, for a 75 year term, for $302m. The site was developed as a commercial centre, the Shui On Centre (“the Centre property”) with a syndicated loan facility of $360m arranged by Hong Kong and Shanghai Banking Corporation (“HSBC”) and secured on the Centre property. The development was fully completed during 1987 (at a total cost, including the acquisition of the site, of $596m) and units were let on terms of three to six years. 10.In 1987 the Centre property was held by South Castle Limited (“South Castle”), a wholly-owned subsidiary at what was then the lowest level of a perpendicular group structure: South Castle was owned by Shui On Properties Limited, (“Properties”), which was owned by Shui On Investment Company Limited (“Investment”), which was owned by Shui On Group Limited (“Group”), and since 1989 there has been a BVI company, Shui On Company Limited, as the ultimate holding company of the group. Apart from its secured indebtedness to HSBC South Castle also owed $209m to other group companies. The HSBC loan was due to be repaid on completion of the development but HSBC several times agreed with South Castle to roll it over, first until 11 December 1987, and then by further extensions until 11 May 1988 (when it was repaid as mentioned below). The Board of Review (“the Board”) accepted that the group had a genuine need for refinancing by means of a medium-term loan (that is, for seven to eight years) large enough to provide additional finance for other group activities. The Board did not accept that the scheme adopted and carried out was the most natural means of meeting that need. 11.The scheme and some later material events were described in detail by the Board in paras 15-41 of its written decision. The three new subsidiary companies formed to participate in the scheme were the taxpayer, incorporated in Hong Kong on 11 March 1988; Shui On Centre Company Limited (“Centre Co.”), incorporated in Hong Kong on 24 July 1987; and Glorion Holding Corporation (“Glorion”), incorporated in Nauru on 12 April 1988. For present purposes it is sufficient to repeat (as the Court of Appeal did) the more concise summary given by Reyes J in the Court of First Instance, in paras 8 to 13 of his judgment. In that summary “Mitsubishi” refers to Mitsubishi Bank; “Agnew” refers to a company named Agnew Park Limited, a wholly-owned subsidiary of BT Asia (HK) Limited (“BT Asia”); “Bankers Trust” refers to Bankers Trust Company, which had a 50% holding in BT Asia; and “FPB Finance” refers to FPB Finance Limited. Bankers Trust was the source from which the whole scheme emanated. 12.I gratefully adopt the concise summary by Reyes J:
The Commercial Reality 13.It is not easy to discern, from that summary, the real commercial character and effect of the scheme. That is in no way the fault of the judge, or of the Board (of whose more detailed description Reyes J made an accurate précis). The fact is that the scheme was very complicated and artificial, and the Board was inclined to the view (para.157 of the decision) that some of its complications were intended to make it less easy “to see through the scheme”. The Board got little assistance from the taxpayer’s witnesses who had a tendency, when questioned about aspects of the scheme, to suggest that someone else would be better equipped to answer that question (paras 66, 70 and 90 of the decision). The scheme seems to have been referred to both as a “tax defeasance scheme” and a “debt defeasance scheme” (for instance, paras 67 and 70 of the decision). 14.The Board produced a diagram (para.149) showing (in a simplified form) the flow of funds under the scheme. This is of some assistance in understanding the numerous transfers of money (almost all by book entries in bank accounts held at Bankers Trust) which were effected on 9, 10 and 11 May 1988. But it does not cover the flows of money which took place under the scheme between 1988 and 1996 (and in particular between 1988 and 1994, when there was a partial sale of the Centre property and the Mitsubishi loan was fully repaid). 15.The commercial reality was that the refinancing of the HSBC loan was undertaken by Mitsubishi, which was willing to make a secured loan of $600m for eight years on commercial floating-rate terms. Under the Mitsubishi loan interest was payable quarterly, and principal by half-yearly instalments. Initially the floating rate of interest was 7.25%, but as it happens it rose quite rapidly to reach 11.25% within a year. The natural course, as the Board observed (para.157), would have been to lend direct to South Castle, or Centre Co. (the new subsidiary which was to hold the Centre property). Instead, for tax reasons, the Mitsubishi loan was made to the taxpayer, a company with a paid up capital of $2 which did not own the Centre property. 16.The $1,200m loan made through the mechanism of FPB Finance and Agnew, by contrast, was in no sense an ordinary commercial loan. It was referred to in the taxpayer’s opening written submissions to the Board (para.31) as the primary loan (the Mitsubishi loan being described as the secondary loan). That was a tendentious description, since (on the Board’s unchallenged findings) the “primary loan” money went round in a circle in a couple of days. The main purpose of this loan (“the FPB/Agnew loan”) was to bring into existence the income stream (interest at a fixed rate of 9.375% p.a. on $1,200m, that is $112.5m p.a. for eight years) which Agnew assigned to the taxpayer by the assignment of 9 May 1988 (para.9(2) of the first instance judgment). In theory Agnew’s right to repayment of the principal in 1996 remained in existence, but it was a very shadowy existence, because (paras 9(5) and 10(5) of the first instance judgment) Glorion agreed that Bankers Trust should credit itself with $600m in consideration of Bankers Trust’s covenant to discharge (or procure the discharge of) the obligation to repay the principal of $1,200m on 11 May 1996, representing a 50% discount in respect of the obligation’s deferment for eight years. Glorion had become entitled to this sum of $600m as payment for new shares taken up by its holding company, South Castle, and that $600m was part of the sum of $1,200m paid to South Castle by Centre Co. on account of the purchase price ($1,310m) payable by South Castle to Centre Co. for the Centre property. 17.For practical purposes, therefore, the aim and effect of the complex manoeuvres involving the FPB/Agnew loan was to burden Centre Co. with a liability for interest at the rate of 9.375% p.a. on a shadowy principal sum of $1,200m, and as a result of Agnew’s assignment the taxpayer became entitled to this income stream. It was used by the taxpayer to pay to Mitsubishi both quarterly interest (at a floating rate) on the outstanding principal and half-yearly instalments of principal (the principal repaid amounting to about $52.5m in the first year and gradually increasing so that but for the early redemption it would have amounted to $102.3m in the last year of the eight-year term; the precise figures are set out in table 1A of the Commissioner’s written submissions forming Annexure C to the stated case). But because the Mitsubishi loan was at a floating rate there was no certainty that there would be a close match between the sums payable to and by the taxpayer. On the contrary, it was almost certain that they would not, because interest rates were volatile. The swap arrangements (para.9(3) and (4) of the first instance judgment) were therefore put in place to correct the likely mismatch. They could be likened to a bit of genetic engineering which transposed the characteristics of two organisms without completely destroying their identities; whether interest rates rose or fell, the effect of the swap was to keep Centre Co.’s liability under the FPB/Agnew loan at the same level as the taxpayer’s liability under the Mitsubishi loan. The Board (para.157) regarded the interposition of BT Asia into the swap transaction as a further bit of camouflage. 18.The taxpayer’s opening submissions asserted (para.37) that the agreements entered into by the group, its bankers and their respective subsidiaries were all supported by consideration. That may be correct if all the companies in the Shui On group, including the new subsidiaries, are regarded as a single economic entity. It is certainly not true if each company is looked at on its own. The most obvious example is Glorion, which had a paid-up capital of $600m (in contrast to the $2 issued share capital of the taxpayer and Centre Co.) but paid it all out for a purpose that benefited, not Glorion, but Centre Co., with the result that Glorion was left with accumulated losses of $600m (as appears from the balance sheet at Annex H to the taxpayer’s closing submissions, which are themselves Annexure B to the stated case). Yet the taxpayer’s case has always emphasized that profits tax applies separately to each company in a group, with no provisions for group relief. 19.It should also be noted, for the sake of completeness, that the only money which went out of the circle at once was the sum of $358m paid by Mitsubishi to HSBC in final discharge of the syndicated loan. This money was paid by Mitsubishi at the direction of the taxpayer, and so reduced to $242m the amount that the taxpayer had available to pay to Agnew for the assignment of the income stream. The balance of $358m was paid to Agnew by South Castle (para.10(3) of the first instance judgment). That was appropriate as South Castle was selling the Centre property free from HSBC’s security. After making that payment and injecting $600m into Glorion, South Castle was left with $242m which it paid to Investment in discharge of intra-group indebtedness. 20.The practical effect of these arrangements, once the scheme had been set up, is reflected in the summaries of the taxpayer’s profit and loss accounts set out in para.42 of the Board’s decision. For each full year of assessment the interest received from Centre Co. (the assigned income stream) was $112.5m; this was however reduced by variable amounts (the highest, almost $22m, in 1992-93) by the swap arrangements. The net income figure was then further reduced by (1) the interest element of the sum payable to Mitsubishi (a sum that decreased year by year as the outstanding principal was reduced); (2) “deferred expenditure w/o [written off]” (a sum that increased year by year, the eventual total being $606.2m, a figure explained in para.45 of the Board’s decision); and (3) other expenses (which were trifling). In each year of assessment these reductions produced a modest loss as the bottom line. The challenged assessments seek to turn these modest losses into significant profits by adding back the “deferred expenditure w/o”. The Procedural Issue 21.This issue depends on the true construction and effect of provisions in Parts X and XI of the IRO, which relate to assessments, objections to assessments and the appeal process. Under s.59(1) assessments are normally to be made by an assessor, but an assessment under s.61A may be made only by an Assistant Commissioner. Under s.60 the normal period under which an assessment or an additional assessment can be made is six years from the expiration of the relevant year of assessment. All the assessments now under appeal were made within the six-year period, although some were made quite close to its expiration. Under s.64 a taxpayer may object to an assessment. The assessment is then reviewed by the Commissioner, who may (s.64(3) and (4)) agree to the objection and reduce the amount of the assessment, or reject the objection in a reasoned determination. In the latter event, the taxpayer may appeal from the Commissioner’s determination to the Board, which may (s.68(8)) confirm, reduce, increase or annul the assessment, or remit it to the Commissioner. 22.Counsel for the taxpayer took the point that no additional assessment had been made in this case. It was not therefore open to the Commissioner, he submitted, to deal with the matter on grounds based simply on ss 16 and 17 of the IRO, without reference to any tax benefit within the meaning of s.61A, since the original assessments were all expressed to be made under s.61A. 23.What happened in this case, as a matter of fact, is that on 10 March 1995 a senior assessor wrote to the taxpayer’s accountants informing them of the Assistant Commissioner’s intention to make assessments under s.61A. The actual notices of assessment did not refer to s.61A, but the annexed profits tax computations bore the note “Assistant Commissioner’s note: The company is assessed under Section 61A of the Inland Revenue Ordinance”. 24.In developing his submissions the taxpayer’s counsel contrasted assessments under s.61A with what he termed “direct assessments” under s.59 or s.60. The distinction that he sought to make has little substance. Any assessment under Part X of the IRO is an assessment of an amount of profits tax charged at the appropriate rate on a sum of profits liable to that tax. “Assessment” is, as Deputy Judge A To said in CIR v. Common Empire Limited [2006] 1 HKLRD 942, para.39 (approved by the Court of Appeal in the same case [2007] 1 HKLRD 679, paras 6-7),
A notice of assessment is an official written notification of the amount of tax arrived at by that process. The amount of the assessment is its essential feature, and that is what a dissatisfied taxpayer’s objection is ultimately directed to (as appears from the language of s.64(3) and (4)). 25.An assessment made pursuant to the provisions of s.61A has the peculiarity that it may be made only by an Assistant Commissioner. That is no doubt because the assessment will often call for the analysis of a complex transaction involving large sums of money. But in other respects it is the same as any other assessment. It has no special time limits and no special procedure for objections or appeals. Moreover (as its position in Part X of the IRO indicates) it is not a separate charging provision. 26.The Board expressed the view (para.106) that “section 61A is not a charging provision at all”. It is certainly not a separate, self-contained charging provision. What it does is (like s.61) to extend the scope of the ordinary charging provisions in the IRO. The power conferred by s.61A(2)(b) is particularly far-reaching, as Lord Hoffmann NPJ explained in CIR v. Tai Hing Cotton Mill (Development) Limited (2007) 10 HKCFAR 704, para.17. The position was correctly summarised by Reyes J (para.52, in a passage approved by the Court of Appeal) when he said that s.61A is only an aid to the charging provisions in the IRO:
27.The submissions on behalf of the taxpayer were therefore based on the incorrect assumption that an assessment made pursuant to s.61A is essentially different from what was termed a “direct assessment”. They were also in error as to the Commissioner’s and the Board’s respective functions under Part XI of the IRO. 28.When the Commissioner then in office considered the taxpayer’s objections he took the view that the “deferred expenditure” was not a proper deduction under s.16(1) of the IRO. In the reasons set out at the end of his determination he stated:
This passage has been criticised, and there is a good deal of force in the criticism. It is almost self-contradictory unless it is read as stating two alternative conclusions. But the conclusion in the last sentence, at least, is unambiguous. That conclusion was challenged, equally unambiguously, in para.1 of the taxpayer’s notice of appeal, which asserted that the deferred expenditure was incurred to produce interest income chargeable to profits tax, and was therefore deductible under s.16(1) of the IRO. 29.As the Board correctly observed, by reference to the decisions in Mok Tsze Fung v. CIR [1962] HKLR 258 and (after the amendment of s.64 of the IRO) CIR v. The Hong Kong Bottlers Ltd [1970] HKLR 581, the Commissioner’s function, once objections had been made by the taxpayer, was to make a general review of the correctness of the assessment. In Mok Mills-Owens J said at pp 274-275 :
30.Similarly the Board’s function, on hearing an appeal under s.68, is to consider the matter de novo: CIR v. Board of Review ex parte Herald International Limited [1964] HKLR 224, 237. The taxpayer’s appeal is from a determination (s.64(4)) but it is against an assessment (s.68(3) and (4)). The taxpayer’s counsel drew attention to the fact that when Part XI was amended in 1965, the wording of s.68(4) was altered to refer to the onus of proving that the assessment was “excessive or incorrect” (rather than simply “excessive”). This, it was argued, showed that the amount of an assessment was no longer always the essential issue. Counsel for the Commissioner could not suggest any particular reason for the alteration, other than a general tidying-up of the language. Whatever the explanation, I am satisfied that the alteration was not intended, by what is sometimes called a side-wind, to make a major change in the scheme and effect of Part XI of the IRO. 31.The Commissioner and the Board were of course under a duty to proceed fairly, so that the taxpayer was not taken by surprise. But deductibility under ss 16 and 17 of the IRO was inevitably a live issue at every stage, because of the way in which “tax benefit” is defined. It is regrettable that the Commissioner’s reasons for his determination were not more clearly expressed, but that has not resulted in any unfairness. 32.In my judgment, the Court of First Instance and the Court of Appeal (the Hon Rogers VP, Le Pichon JA and Stone J) were right to uphold the Board’s decision, and the taxpayer fails on the procedural issue. Sections 16 and 17 of the IRO 33.It was therefore open to the Commissioner to rely on the effect of ss 16 and 17 of the IRO, apart from s.61A. Sections 16(1) and 17(1)(c) of the IRO are in the following terms :
34.In seeking to apply these provisions the Board referred to the decision of the Privy Council (on appeal from Hong Kong) in Wharf Properties Ltd v. CIR [1997] AC 505, following and applying the principles stated by the Privy Council (on appeal from the Federation of Rhodesia and Nyasaland) in Commissioner of Taxes v. Nchanga Consolidated Copper Mines [1964] AC 948. Seeking to apply those principles, the Board concluded that the $606.2m paid by the taxpayer (as consideration and legal fees) for the assignment of the income stream was expenditure of a capital nature, and so not deductible in consequence of s.17(1)(c) of the IRO. 35.In my judgment that conclusion was plainly correct. This is a clearer case than Wharf Properties, in which annual interest was paid (but on a loan for the acquisition of property which was not yet income-producing). The only possible criticism to be made about the Board’s treatment of this part of the case is that they were too ready to accept the taxpayer’s inclusion in its profit and loss accounts of the vague formula “deferred payments w/o” (which is not satisfactorily explained by the equally vague note in the accounts quoted in para.43 of the Board’s decision). The fact is that there was a single capital payment of $600m on 9 May 1998, together with approximately $6.2m legal fees paid at about the same time, and no actual payment was made in any later year of assessment. The deduction of ‘deferred expenditure [written off]’ was in effect an attempt to charge depreciation in respect of a wasting capital asset. That is something that tax law has not countenanced since the decision of the House of Lords in Coltness Iron Company v. Black (1881) 6 App Cas 315 (cited by Lord Wilberforce in IRC v. Church Commissioners for England [1977] AC 329, 340-341), except in the form of capital allowances under a special statutory regime. There are such special provisions in Part VI of the IRO, but there is no suggestion that they can apply in this case. 36.I respectfully consider that Reyes J was wholly correct in his reasoning in paras 63-65 :
The Court of Appeal rightly accepted the reasoning of Reyes J on this point without the need to add to it. 37.In his spirited assault on the lower courts’ reasoning and conclusion on this issue, counsel for the taxpayer took three main points. First, citing the words of Lord Millett NPJ in CIR v. Secan Limited (2000) 3 HKCFAR 411, he relied on the form in which the deferred expenditure had been shown in the taxpayer’s financial statements for all the relevant years of assessment. Second, he relied on the decision of the High Court of Australia in FCT v. The Myer Emporium Limited (1987) 163 CLR 199, in which it was held that the consideration for the assignment of a seven-year entitlement to interest on a loan was an income receipt in the hands of the assignor. Third, he relied on the taxpayer’s status (recorded in the agreed facts put before the Board) as a finance company whose principal activity was “to arrange financial activity”. I shall consider these arguments in turn. 38.The relevant passage in Lord Millett NPJ’s judgment in Secan is as follows (at p.419) :
In that case the taxpayer company had, in accordance with normal practice, capitalised interest charges incurred in the course of property development. When the development was completed, it sought, unsuccessfully, to change its basis of accounting and set earlier interest charges against its sales of completed flats. But for present purposes it is important to bear in mind, in considering Lord Millett NPJ’s observations, that they were subject to the overriding requirement of conformity with the IRO. Whether expenditure is on capital or income account for tax purposes is a question of law: Beauchamp v. FW Woolworth Plc [1990] 1 AC 478, cited by Lord Hoffmann in Wharf Properties at p.510. 39.In the present case the effect of s.17(1)(c) of the IRO is that the taxpayer’s accounts, although no doubt properly prepared for the purposes of the Companies Ordinance, Cap. 32, are not a reliable guide to the proper tax treatment. The taxpayer’s payment of $606.2m was a one-off capital payment, and the notional deferred expenditure shown in the accounts in later years was recording a loss of capital. 40.In Myer a large company in the retail trade loaned A$80 to a newly-acquired treasury subsidiary. The loan was for a fixed period of seven years, and the lender then sold the seven-year income stream to a bank for a lump-sum payment of A$45.37. There was therefore some superficial similarity to what happened in the present case. But the crucial distinction (as counsel for the Commissioner submitted, and I accept) is that Myer was concerned with the taxation of the assignor who sold the income stream and received the lump sum, not with the position of the buyer. There is no reason why the two treatments should be the same: Wharf Properties at p.511. In the hands of the taxpayer the income stream which it acquired from Agnew was a distinct item of property, and its treatment would depend on (among other things) how the taxpayer intended to make use of it. 41.That leads on to the third main point made on behalf of the taxpayer. The agreed facts in the case stated include (para.4) the following :
Building on this foundation, counsel for the taxpayer submitted that money is analogous to the stock in trade of a finance company, and that this marked the income stream, in the taxpayer’s hands, with the character of income. 42.The agreed description of the taxpayer’s principal activity is however an unstable foundation on which to build any substantial argument. “Financial activities” is a wide and loose expression. The fact is that the taxpayer was a single-purpose vehicle, brought into existence in order to perform its predetermined function in the tax-avoidance scheme. It did not acquire the income stream with the intention of selling it or otherwise turning it to account. Its function was to hold the assigned income interest and to serve as a conduit between Centre Co. and Mitsubishi. The argument that the assigned interest was trading stock, or something analogous to trading stock, must be rejected. 43.Counsel for the taxpayer referred the Court to what he described as the most recent English authority on the topic, Inland Revenue Commissioners v. John Lewis Properties Plc [2003] Ch 513. It concerned a transaction comparable to that in Myer, except that the assignment was of five years’ rental income (from retail premises owned by the group’s property holding company and let to its trading company). The judgment of Dyson LJ contains (at paras 80-87) a concise summary, derived from earlier case-law, of the indicia of a capital payment in the hands of the recipient (that is, in these cases, the assignor). The whole passage deserves attention but the five indicia, in brief summary, are (1) the duration of what is assigned; (2) the value of what is assigned; (3) the diminution in value (temporary or permanent) of what is retained by the assignor; (4) the lump-sum character of the consideration received by the assignor; and (5) the transfer of risk as a result of the assignment. All five indicia point to the same conclusion in this case. Section 61A of the IRO 44.The second substantive issue does not therefore arise for decision, and it would be inappropriate for this Court, as a court of last resort, to embark on any elaborate discussion of how s.61A of the IRO might have applied in a hypothetical situation. But I would make a few general observations. 45.Both sides recognised that since the hearing before the Board, which took place in 2004, the structure and effect of s.61A have been explored and elucidated in three important decisions of this Court, that is CIR v. Tai Hing Cotton Mill (Development) Limited (2007) 10 HKCFAR 704 and CIR v. HIT Finance Limited (2007) 10 HKCFAR 717 (in both of which judgments were handed down on 4 December 2007) and Ngai Lik Electronics Co. Limited v. CIR, FACV No. 29 of 2008 (in which judgment was handed down on 24 July 2009). Had the Commissioner and the Board had the benefit of those judgments some rather unsatisfactory features of these proceedings would no doubt have been avoided (though there has not been any procedural unfairness to the taxpayer, nor has the taxpayer’s counsel complained of any procedural unfairness). 46.In particular, both the Revenue and those advising taxpayers will take account of the important observations of Mr Justice Ribeiro PJ in Ngai Lik at paras 137-139. In para.137 Mr Justice Ribeiro PJ referred to :
These are, as he put it earlier in his judgment (para.36) “the three interlocking conditions – transaction, tax benefit and dominant purpose – [which] must be properly aligned and approached with the necessary degree of precision if the application of s.61A is not to miscarry”. 47.At the beginning of this judgment I referred briefly to the question whether s.61A can have any application to a tax-avoidance scheme which (in colloquial terms) simply does not work and indicated that it cannot apply, so answering the question that had been left open in Ngai Lik. The logic of the argument is that liability under s.61A presupposes non-liability in the absence of s.61A. Otherwise there would be no benefit in the statutory sense. 48.In HIT Finance, para.15, Lord Hoffmann NPJ made some observations which might be thought to imply (though they certainly did not express) a contrary view. Any observations of Lord Hoffmann NPJ are entitled to great respect, but it is to be noted that in that case the company referred to as Finance was not held liable under s.61A, and the company referred to as HITL (which was liable under s.61A) was not attacked under ss 16 and 17 until the case reached the Court of Appeal. I do not think Lord Hoffmann NPJ’s observations can be regarded as a considered contrary conclusion. 49.It is therefore incumbent on the Commissioner to make clear (as she did in HIT Finance) any alternative grounds on which she may seek to support an assessment. Alternative grounds may consist of familiar provisions of ss 16 and 17 of the IRO, or may be alternative ways of formulating the Commissioner’s case under s.61A (see Ngai Lik at paras 76 to 82). 50.The last point on which I wish to make some tentative comments is the way in which the Commissioner should, in the case of a complicated tax-avoidance scheme, approach the task of making an assessment so as to counteract the tax benefit which would otherwise be obtained by the relevant person. This is not entirely straightforward and it is best approached by stages. 51.The simplest situation is when a taxpayer has an existing source of income subject to profits tax, and participates in some free-standing transaction designed to produce a loss in order to set it against the income which would otherwise be taxable. If the three interlocking conditions are satisfied the appropriate action for the Commissioner is to make an assessment in the manner indicated in s.61A(2)(a) – that is by wholly disregarding the loss-making transaction. (Possibly the Commissioner might take the same course under s.61, relating to artificial or fictitious transactions, but in practice s.61 seems to have been little used since s.61A was enacted.) 52.If however the tax-avoidance scheme is more complicated, and brings into existence new sources of income as well as new deductions or losses, the task of counteracting the tax benefit requires the Commissioner to act under s.61A(2)(b), which is in wide terms : the assessment is to be made
This wide language must of course be read subject to the familiar constraints imposed by public law. The Commissioner must act reasonably and avoid any arbitrary or exorbitant exercise of the statutory power. That has been spelled out by Lord Hoffmann NPJ in Tai Hing (para.21) and by Mr Justice Ribeiro PJ in Ngai Lik (para.113). Mr Justice Ribeiro PJ put it as follows:
53.Even when s.61A(2)(b) is in play there are differing degrees of complication. Tai Hing was a relatively simple case in which a manufacturing company sold surplus land to a special-purpose subsidiary which was to develop the land in a joint venture with a developer. The special feature of the relevant transaction (the sale of the land by the manufacturer to the subsidiary) was that it was not a simple sale at market value, but a sale for a composite consideration fixed by a formula which was intended to secure, and did in the event secure, part of the development profit for the manufacturer, in whose hands it was not liable to profits tax. The result (but for s.61A) would have been that about $200m would have escaped profits tax in the hands of the subsidiary (which was trading as a developer) as compared with its profit if it had bought the land at its market value of $800m. The appropriate alternative hypothesis was therefore to postulate a straightforward sale at market value. Commissioner of Taxation v. Spotless Services Limited (1996) 186 CLR 404, the Australian case about the short-term deposit in the Cook Islands, which Lord Hoffmann NPJ referred to in his judgment in Tai Hing (para.20), is another example of a situation in which it was fairly easy to identify what form the “reasonably postulated hypothetical transaction” should take. 54.In Tai Hing (paras 21 and 29) Lord Hoffmann NPJ made some general observations about the appropriate hypothesis which counsel for the taxpayer seized on as indicating that the appropriate hypothesis was a matter for evidence, and even a matter that should have been put to the taxpayer’s witnesses. I do not think that it is the right approach, and I do not think that Lord Hoffmann NPJ intended his observations to be taken in that way. The scope of the Commissioner’s powers under s.61A(2) is a question of statutory construction (Tai Hing at para.15). The exercise of those powers is for the Commissioner’s judgment, subject to public law constraints. Of course the Commissioner must have regard to the facts as agreed or found by the Board. But any inquiry into the subjective attitudes of the taxpayer and its associates would be inconsistent with the objective approach that is one of the essential features of s.61A. In some cases the taxpayer or associates of the taxpayer may have been cross-examined before the Board as to why they preferred the impugned transaction to simpler and more natural alternatives. But that evidence would be directed to the issue of “sole or dominant purpose” and would not be of much relevance (and certainly not conclusive) as to the appropriate course to be taken under s.61A(2)(b). The Australian case of Commissioner of Taxation v. Hart (2004) 217 CLR 216 (in the passage at p.224 on which the taxpayer’s counsel relied) seems to be an example of that sort of thing. 55.To my mind one of the most difficult features of this appeal is that the Shui On scheme (like that in HIT Finance, but unlike those in Tai Hing and Ngai Lik) involved newly-acquired subsidiary companies, either in Hong Kong or in tax havens, which were there solely for tax-avoidance purposes. They would probably never have come into existence at all but for those purposes (in Tai Hing and Ngai Lik, by contrast, there were good commercial reasons for having the new subsidiaries, apart from artificial tax avoidance). There are obvious difficulties in determining the most appropriate means of counteracting a tax benefit obtained by a new subsidiary which, but for that benefit, would never have seen the light of day at all. That may be one reason why counsel for the Commissioner, in supplementary submissions, put forward a relatively elaborate alternative hypothesis which still envisaged a loan of $1.2 billion from the taxpayer to Centre Co. (a hypothesis which the taxpayer’s counsel attacked as unrealistic and lacking in any evidential basis). The other reason for the elaboration of the Commissioner’s alternative hypothesis may have been in order to forestall the argument that in assessing the taxpayer the Commissioner had chosen the wrong target. 56.This Court does not have to decide these questions and, for my part, I prefer to express no view about them. If they have to be decided on some future occasion it will, I hope, be in proceedings in which the essential issues are clarified and narrowed at an early stage, so as to enable them to be fully explored in argument. 57.Accordingly, I would dismiss the appeal. As to costs, there will be an order nisi awarding the Commissioner her costs of this appeal. Such order will become absolute 21 days from today unless a party notifies the Registrar before then that some other order as to costs is sought. In that event, costs will be dealt with by the Court on written submissions as to which the parties will seek procedural directions from the Registrar. Chief Justice Li : 58.The Court unanimously dismisses the appeal and makes the costs order nisi referred to above.
Mr David Bloom QC, Mr Barrie Barlow SC and Mr Stewart Wong (instructed by Messrs Mallesons Stephen Jaques) for the appellant Mr Michael Furness QC, Mr Ambrose Ho SC and Mr Paul H.M. Leung (instructed by the Department of Justice) for the respondent |
Cases cited in this judgment
Other judgments that cite this case