Federal Commissioner of Taxation v Ryan

The next question which arose for Mr Bourke was how to structure the refinancing. He took advice from Mr Cherry, then a partner of Arthur Young, Chartered Accountants and Mr Verzi of the same firm as to the income tax implications of the refinancing. Arthur Young replied to Mr Bourke by a letter dated 10 August 1988 written by a Mr Johnston under the supervision of Mr Cherry. The recommendation was that the funds necessary should be borrowed by CPF. That recommendation was clearly commercial and non tax driven. Obviously, if income tax were no consideration, there would then be a borrowing from the Australian borrower CPH or ACP (the loan to ACP had obvious commercial overtones given that company's cash flow) directly to the companies which were to use the funds borrowed. The Arthur Young advice was, however, to structure the transaction differently. It was suggested that the funds borrowed by CPF be lent initially at interest to ACP. As I have said, given that this company had a cash flow which would enable interest on the borrowings to be serviced, that part of the advice was obviously commercial. However, the advice then continued that ACP use the funds to subscribe for shares (either ordinary or redeemable preference) in MLG and that that company in turn invest the funds raised by the issue of shares to ACP in a new issue of redeemable preference shares by CPIL(UK). This company in turn would advance the funds through to Valassis via a structure involving the Netherlands and the Netherlands Antilles. The insertion of the Dutch/Netherlands Antilles companies is not suggested to have anything to do with Australian tax - it is related, one would assume, to the provisions of Double Tax Treaties and interest withholding tax arising from the United States. The letter then referred to an announcement that had been made by the Australian Government to the effect that as and from 1 July 1989 dividends received by Australian resident corporations from non-Australian sources would be treated as exempt income for Australian tax purposes. The effect of that proposal, if enacted, would have been, as the advice indicated, that dividends from non Australian sources received by a company with carry forward losses would be treated as reducing those losses, so that they would not be available to be used as deductions against assessable income - a consequence of s 80 of the Act. Arthur Young saw that result as "untenable". It would be obviated by the suggested mix of borrowing and share subscription. The letter spoke of it being : " … obviously sensible that any new overseas investment is structured in the correct manner." That would involve the subscription passing through an Australian entity which is in "a tax neutral position". No doubt, by "correct" the author of the advice meant that taxation consequences should be minimised. Mr Cherry said in evidence that in supervising the writing of the letter he gave no thought, nor was there any discussion about, the provisions of s 79D of the Act, the terms of which are set out later in these reasons. This was because he had, whether on behalf of the Consolidated Press Group or some other client was not made clear, taken counsel's advice earlier and had been advised that that section had no application. It would seem that Mr Cherry concurred with that advice. Whether the advice was correct is a matter which arises in the present case. It is perhaps interesting to note, however, that the 1989 income tax return of MLG prepared by Mr Cherry's firm and on which his name appears in the postal address proceeds on the basis that that section did have application. The discrepancy was not put to Mr Cherry. Before the letter had been written there were, however, two other developments. On 15 March 1988 the Chancellor of the Exchequer in the United Kingdom announced a change to the legislation of that country involving non resident United Kingdom companies, among other matters. Although the changes were to take effect from the date of the announcement, reprieve was given to companies incorporated prior to the date of the announcement. The proposal was to apply to such companies only after five years from the date of the announcement unless in the interim the company changed its central management and control to the United Kingdom, ie became resident there. After that date if the UK companies remained non resident, they were to be treated as resident in that country for tax purposes, with the consequences that the existing UK tax regime, including capital gains tax, would apply to such companies. A consequence was that after the expiration of the five year period a company incorporated in the United Kingdom, whether or not resident there, would be taxed on worldwide income. The second development was an Economic Statement and the issue of a consultative document on 25 May 1988 by the Treasurer of Australia proposing a change in the Australian taxation system dealing with foreign source income. The proposal was to be delayed until the start of the 1989-90 income tax year. Briefly, it involved taxing on an accrual basis certain income of non-resident entities in which Australian residents had an interest where the income was derived in a low-tax or no-tax country or had benefits from certain kinds of taxation concessions. It was to supplement legislation then on foot dealing with foreign tax credits for taxation payable in other countries. It may here be noted that the proposal ultimately found legislative form in the provisions of Division 10 of the Act, although in a slightly different form from the original proposals. The provisions dealing with foreign tax credits, the detail of which is not important for present purposes, were also taken into account by Mr Cherry in suggesting the so-called "tax-neutral" companies in the mixed debt-equity structure, although they were not mentioned in the Arthur Young advice. The proposals suggested by Arthur Young were not implemented immediately. This came about because in around November 1988 News Corporation, a competitor in some areas at least of the Consolidated Press Group, acquired the insert business competitor which had initiated the 1987 price war with the result that price stability in the industry was achieved (the price war apparently came to an end) and the Valassis Group returned to profitability. The next commercial event occurred in March/April 1989 when Mr Packer had talks with Sir James Goldsmith about a proposal to made a takeover bid for a United Kingdom company BAT Industries Plc ("BAT") . The participation by the Consolidated Press Group would require what Mr Packer nominated to Mr Bourke as "a fair amount of capital". Mr Bourke was charged with the responsibility of looking after the financial side for the group. The "fair amount of capital" was in the order of 250 million pounds sterling - although not all of this would necessarily be required to be committed to the bid. In April 1989 it was expected that the participation in the BAT takeover would produce profits, presumably in the form of dividends flowing from CPIL(UK) in a very substantial amount. Certainly by the time the bid went forward it was expected that assets held by BAT would be sold as quickly as possible ("unbundled" to use the terminology of Sir James Goldsmith) with dividends being realised within a 12 month period, although a more conservative estimate may have been longer, with an ongoing sustainable dividend emerging from what is said to have been a highly profitable tobacco business of BAT. The overall expected profit of the proposed participation by the Consolidated Press Group of Companies in the bid was estimated by Mr Bourke to be in the order of 1 billion pounds sterling. Mr Bourke recalled the Arthur Young advice of August 1988. He consulted Mr Cherry and Mr Verzi who repeated their previous advice as to the most advantageous structure for channelling the borrowing through to the English companies. No record of that advice was kept. In the result the advice was followed. On 28 April 1989 ACP applied for and was allotted 600,000 redeemable preference shares of $1.00 each at a premium of $500 per share in the capital of MLG - a total subscription price of A$300.6 million. On 2 May 1989 CPF lent ACP $A300.6 million to enable the share subscription to proceed. The shares were apparently allotted before the loan on 28 April 1989, although nothing turns, it would seem, on this. On 5 May 1989, CPIL(UK) allotted 2,400,000 fully paid ordinary shares of US$100 each to MLG. Funding for this came initially from an advance of US$240,000,000 from Westpac Banking Corporation to CPH which appears to have been made on 5 May 1989. A subsequent series of loans and share subscriptions resulted in one million fully paid redeemable preference shares of US$100 each being allotted in CPIL(UK) on 28 November 1989. In the meantime another company in the group, CP Investment (Singapore) Pte Limited ("CPI(Sing)")had been incorporated in Singapore. There was no attempt in evidence to explore why Singapore was chosen as against Australia, although it may be inferred that there was a taxation reason. As part of the BAT bid it subscribed for 195 ordinary shares of Ł1 each at a premium paying Ł8,835,125 in Hoylake Investments Limited ("Hoylake") which was to be the vehicle for the BAT bid. Hoylake was incorporated in Bermuda. Again no attempt was made to explore why Hoylake was incorporated in Bermuda, but again tax reasons may be inferred, particularly as that country is a recognised tax haven. CPI(Sing) owned 32.5 per cent of the Hoylake shares, the balance by interests associated with Sir James Goldsmith and Mr Jacob Rothschild, the partners with the Consolidated Press Group in the bid. Hoylake then commenced purchasing BAT shares on the market. The making of a formal bid was announced in a press release dated 11 July 1989. The announcement referred to the investors in Hoylake as including CPH, "a company owned by Mr Kerry Packer". By the time the bid was announced Hoylake had accumulated 1.25% of the issued share capital of BAT. On 10 July 1989 MLG lent interest free the sum of US$100,000,000 to CPIL(UK). The funds originated from United States dollar denominated accounts of CPF in the United States with Westpac and were treated as an advance by CPF to ACP to enable that company to pay subscription monies on redeemable preference shares in MLG. On 28 November 1989 these funds were used to pay for one million redeemable preference shares of US$100 each in CPIL(UK) which were allotted to it on that day. The US$100,000,000 was, when received, immediately lent by CPIL(UK) to CPI(Sing). Initially this loan was described as interest free, but in fact interest was charged on it at the rate of 15% from the date the loan was made until 6 October 1989 and thereafter at a rate of 16.25% per annum. Part of the Hoylake bid was conditional upon a sale of Farmers Group Inc, a company conducting an insurance business. That transaction required approval of the California Department of Insurance. Approval was ultimately denied in or about April 1990 and, in the result, the BAT bid was withdrawn on 23 April 1990. On or around 5 June 1990 Hoylake went into voluntary liquidation.

The United Kingdom Reorganisation It is obvious enough that the proposed United Kingdom and Australian tax changes received close attention although clearly no final decision could be made taking into account the proposed Australian changes until the legislation was introduced. This was noted by the board of CPIL(UK) at its meeting on 15 May 1989.

Mr Mackenzie corresponded with Mr Cherry on the subject seeking a report from him for submission to the next board meeting of CPIL(UK) to be held in March 1990. He wrote that he was not particularly interested in avoiding the attribution principles that were proposed for Australia in the so-called "accruals" proposal but was concerned that the group could be taxed both in Australia and overseas (presumably in at least the United Kingdom) and thus lose the benefits of franking credits which would otherwise be available to the ultimate parent company and presumably ensure that ultimate Australian distributions would not result in tax being payable at least in any substantial way to Australian corporations or individuals. By January 1990 a draft of the proposed Australian accruals legislation had been released. It was the subject of a number of discussions between Mr Cherry and Mr Mackenzie. The clear thrust of Mr Cherry's advice was that the holding structure in the United Kingdom should be relocated to a tax haven. Initially Bermuda and the Bahamas were considered. The pros and cons of each were considered. The matter was discussed at a meeting of directors of CPIL(UK) and CPIHL(UK) on 22 March 1990. Prior to that meeting Mr Bourke, Mr Cherry and Mr Mackenzie travelled to both the Bahamas and Bermuda to seek advice from advisers there as to the implications of establishing new holding companies for the international group. Mr Wolfensohn had a preference for Bermuda. The meeting resolved to recommend to members that each company be placed in voluntary liquidation and that an extraordinary general meeting of members be called on 11 April 1990 for that purpose. The directors of CPIL(UK) also resolved to declare dividends payable to members on 8 May 1990 at the rate of US$18.287 per share on the 2,810,000 redeemable preference shares (a total of US$51,387,000) and on the 2,658,295 fully paid ordinary shares (a total of US$48,613,000). The directors of CPIHL(UK) likewise resolved to declare dividends at the rate of US$0.3466 per share on the 66,000,000 fully paid redeemable preference shares (a total of US$22,878,000) and the 5,000,002 fully paid ordinary shares (a total of US$1,733,000) and at the rate of US$0.3154 per share on the 90,000,000 redeemable preference shares paid to US$0.91 (a total of US$28,389,000).

The board papers of the directors' meeting contain the following statement: "The overriding aim of the final structure is to ensure that the passive income referred to earlier is attributed to Australia as thereby franked dividends may be paid to the shareholders, ie. dividends which will be tax free in the shareholders' hands. Removal of the UK incorporated entities and their replacement with either Bahamas or Bermudian entities will ensure that: (a) this passive income flows tax free to Australia; and (b) there is no possibility of double taxation. While maintenance of the UK companies may not lead to additional corporate tax being paid overall, ie credit will be given in Australia for UK tax paid. It would not be possible to pay franked dividends out of that income." The Bahamas was chosen as the appropriate location for the new holding structure. On 5 April 1990 Consolidated Press International Holdings Limited ("CPIHL(B)")and Consolidated Press International Limited ("CPIL(B)") were each incorporated there. On 12 April 1990 MLG and CPH agreed with CPIL(B) to sell their holdings. The consideration was determined subsequently as a certain number of ordinary "A" class shares of US$1 each in the capital of CPIL(B). All of MLG's and CPH's holdings in CPIL(UK) and CPIHL(UK) were to be transferred as a consequence of the agreements entered into on that day. Although a calculation of the number of shares to be issued by the companies newly incorporated in the Bahamas was at the time of the meeting made in accordance with a draft valuation, the formal valuation for this purpose was made by the Australian firm of Ernst & Young (the successor firm to Arthur Young) and issued on 22 April 1990. No transfers were entered at this stage in the register of members of either company nor were any transfers then approved. On 27 April 1990 a meeting of a committee of directors of CPIL(B) was held in Hong Kong to determine the number of shares to be issued as consideration for the proposed transfers in accordance with the Ernst & Young valuation. In the result, 452,346,000 shares were to be issued to CPH and 118,287,000 to MLG. On 1 May 1990 Ernst & Young were asked to carry out a further valuation of CPIL(UK) as at 7 May 1990 in relation to the proposed acquisition by CPIL(B) of a further 2,400,000 fully paid ordinary shares of US$100 each in CPIL(UK) from MLG. The date of 7 May 1990 was chosen because it was relevant to indexation as by then the shares agreed to be sold would have been held more than one year since acquisition, thus raising the cost base for Australian capital gains tax by the indexation factor relevant. This, presumably, explains why these shares were excluded from the agreement entered into on 12 April. Entries in the meantime were made in the books of account of CPIL(UK) and CPIHL(UK) debiting the dividends which had been declared by the directors and were to be payable as at 8 May. No transfers to give effect to the agreements were ever registered in the register of members of either company, nor did the directors approve any transfer or resolve to direct registration. On 16 May 1990 each of CPIL(UK) and CPIHL(UK) resolved to go into voluntary liquidation and liquidators were appointed to each company on that day, being the day the liquidation took effect under the United Kingdom companies law. At the same meeting the members of each company resolved to authorise the liquidators to distribute the whole or any part of the assets of each company to a member in specie. Subsequent events record the steps necessary to complete the liquidation of each company. The detail is of little importance. Although complicated on their face they represented, in essence, the distribution in specie of assets. On the same day, at the request of the liquidators appointed to each company, each of CPH and MLG directed and authorised them to make "any payments consequent upon the crediting of dividends declared by the Company on 8th May 1990" and "any distributions to members … in the course of the winding up… direct to CPIL(B) in place of any payment or distribution to [the members]." On the next day the liquidators of CPIHL(UK) purported to satisfy the debt due by CPIL(UK) to CPIL(B) arising from the dividend payable on 8 May 1990 by assigning to CPIL(B) the first US$53,000,000 of a debt of US$84,220,334.05 due to CPIHL(UK) by Conpress (Singapore) Pte Limited. A deed of assignment and satisfaction was executed to effect this transaction. Similar action was taken by the liquidators of CPIL(UK) to satisfy the debt purportedly due as a result of the dividend declaration as well as certain other debts by assigning to CPIL(B) the first US$106,751,299.80 of a debt due to CPIL(UK) from CPI(Sing). The general ledger of CPIL(UK) showed that the amount of US$106,751,299.80 was in part a payment of the US$100,000,000 dividend paid by CPIL(B) on behalf of CPIL(UK). On 8 June 1990 an arrangement was made between the liquidators of CPIHL(UK) and CPIL(B) for the distribution in specie of the balance of the assets of CPIHL(UK) and a Deed of Assignment between CPIL(B) and CPIHL(UK) was executed whereby CPIHL(UK) assigned the debts referred to in the above mentioned agreement. The next day a Deed of Assignment and Agreement for Novation was entered into assigning certain debts in consideration of CPIL(B) agreeing to assume certain obligations and liabilities of CPIL(UK) to creditors. Formal novation agreements were entered into between 9 June 1990 and 11 June 1990.

The Part IVA / Section 79D Issue The assessment in issue proceeded, so far as is relevant to this first issue, on the basis of Part IVA of the Act, the general anti-avoidance provision. In general terms it may be said that the Commissioner submits that there were two schemes to which the provisions of that Part applied and that the Applicants obtained as a result of those schemes a tax benefit. Each scheme is said to commence with the application for shares in MLG by ACP and conclude with the allotment of shares by CPIL(UK) in the case of one scheme and by CPIHL(UK) to MLG in the case of the other scheme. The relevant tax benefit is identified, in each case, as being the obtaining by CPH and MLG of deductions in a year of income for interest payable available to be applied against assessable income of these companies in circumstances where, but for the schemes, the interest would not have been deductible against that assessable income, because precluded from deduction by virtue of the provisions of s 79D of the Act. Although the identification of a tax benefit is an ingredient, and indeed an essential ingredient, in the application of Part IVA to a particular scheme, so that it is not usual to separate out the other ingredients of Part IVA from the identification of the tax benefit, it is convenient so to do in the present case because there is a question of interpretation of s 79D which divides the parties and which is vital to the question whether there ever could have been a tax benefit. Before proceeding to that section it is necessary to state that the concept of tax benefit obtained in connection with a scheme, essential to the application of Part IVA, is defined in s 177C(1) of the Act which, relevantly, provides as follows: "(b) a deduction being allowable to the taxpayer in relation to a year of income where the whole or a part of that deduction would not have been allowable, or might reasonably be expected not to have been allowable, to the taxpayer in relation to that year of income if the scheme had not been entered into or carried out." A consequence of s 177D is that the Part can have no application unless a taxpayer has obtained or would, but for s 177F have obtained, a tax benefit in connection with the scheme.

Section 79D - Interpretation The critical question of interpretation dividing the parties is whether, as the Applicants content, s 79D as in force in the 1989-90 income tax year has no application where in the year of income no foreign source income is derived, or whether, as the Commissioner contends, it operates to disallow deductions otherwise available to a resident taxpayer even where no foreign source income is, in the year of income, derived. In the years of income ended 30 June 1989 and 1990 s 79D provided as follows: "(1) Where the amount of a class of income derived by a taxpayer in a year of income from a foreign source is exceeded by the sum of: (a) any deductions allowed or allowable from the assessable income of the taxpayer of the year of income that relate exclusively to income of that class derived from that source; and (b) so much of any other deductions allowed or allowable from that assessable income (other than apportionable deductions) as, in the opinion of the Commissioner, may appropriately be related to income of that class derived from that source; the deductions to which paragraphs (a) and (b) apply shall be reduced respectively by amounts proportionate to those deductions and equal in total to the amount of the excess. (2) In subsection (1), 'class of income' and 'foreign source' have the same meanings as in section 160AFD." Sections 160AFD(6) and (7) define respectively "class of income" and "foreign source" in the following terms: "(6) For the purposes of this section: (a) interest income constitutes a single class of income; (b) offshore banking income constitutes a single class of income; and (c) all other income constitutes a single class of income." "(7) In this section - 'foreign source' in relation to a taxpayer, means - (a) a business carried on by the taxpayer at or through one or more permanent establishments in a foreign country; or (b) any other business, commercial or investment activity carried on by the taxpayer in a foreign country." Section 79D was inserted into the Act by Act No 78 of 1988, s 16 first applicable to assessments for the year of income commencing 1 July 1988. The insertion of s 79D was enacted, as appears from the Explanatory Memorandum which accompanied the Taxation Laws Amendment Bill (No 2) 1988 (which bill subsequently became Act No 78 of 1988) to overcome a perceived deficiency in the then law to be found in ss 51(6) and (7) as those sections stood at the time of the amendment (see the notes to Clause 14 of the Explanatory Memorandum). Section 51(1) is the general provision for deduction of what may be called business or working expenses. Section 51(6) had provided: "Where the amount of a class of foreign income derived by a taxpayer in a year of income from a foreign source is exceeded by the sum of - (a) any deductions allowed or allowable from the assessable income of the taxpayer of the year of income that relate exclusively to income of that class derived from that source; and (b) so much of any other deductions allowed or allowable from that assessable income (other than apportionable deductions) as, in the opinion of the Commissioner, may appropriately be related to income of that class derived from that source, a deduction is not allowable under subsection (1) in respect of the amount of the excess." Section 51(7) provided that the expressions "class of income" and "foreign source" were to have the same meanings as in s 160AFD. These provisions had a correlation in s 160AFD of the then Act. That section formed part of the provisions concerned with credit being given in Australia for taxes payable elsewhere, generally referred to as the "foreign tax credit" which were enacted in 1986, first applicable to the year of income commencing 1 July 1987 and replacing earlier provisions dealing with credits to be given in respect of tax payable in Papua New Guinea. The basic scheme of the foreign tax credit provisions was that if a resident taxpayer had assessable income of a year which included foreign income and, being liable so to do, had paid foreign tax on that foreign income the taxpayer was entitled to a credit of whichever was the lesser of the foreign tax (subject to an immaterial reduction in amount) and the Australian tax. Foreign income was segregated into classes - interest income and non interest income (see s 160AFD), but that does not affect the present discussion. Because of the dual requirement of foreign income of a year of income and tax paid in respect of that income in the year, it was obvious that the provisions of s 160AFD(1) operated only in a year where there was a derivation of foreign income, and not in a year where no foreign income was derived. It may not be thought, therefore, surprising that s 51(6) was expressed to operate only where there was a class of foreign income derived in the year of income and was not expressed as operating where none was derived. However, s 160AFD dealt with the situation where a resident taxpayer had foreign losses in the preceding seven years. Such foreign losses were required to be recouped against the relevant class of foreign income before the taxpayer was taken to have foreign source income in the year of the particular class. The purpose of s 51(6) was to quarantine deductions allowable under s 51(1) but relating exclusively, or in the Commissioner's opinion appropriately related, to the foreign source income so that those deductions were to offset foreign source income, rather than other non foreign source income. Certain deductions (referred to as "apportionable deductions", a defined expression in s 6(1)) were excluded from the quarantining regime. Two defects were, it would seem, perceived in s 51(6). The first was that it quarantined (subject to "apportionable deductions" not here relevant) only deductions under s 51(1). Deductions under other sections of the Act were not quarantined. Second, it did not deal at all with cases where there were foreign losses - not surprising as carry forward losses were not, as such, deductible under s 51(1). These related defects were cured by ensuring that s 79D operated to extend the quarantining to "any deductions allowed or allowable" provided that they related to the relevant class of foreign income. The Explanatory Memorandum which accompanied the Taxation Laws Amendment Bill (No 2) 1988, after referring to the proposed deletion of s 51(6), commented in relation to clause 16, proposing the insertion of s 79D, as follows: "Clause 16 will insert in the Principal Act proposed new section 79D which will in effect ensure that a current year foreign loss can only be carried forward in terms of section 160AFD of the Principal Act for offset against future foreign income of the same class from the same foreign source. Subsection 79D(1) specifies that where the amount of a 'class of income' derived by a taxpayer in a year of income from a 'foreign source' is exceeded by the sum of certain deductions, the respective deductions are to be reduced by amounts proportionate to those deductions and equal in total to the amount of the excess. In practical terms this means that each deduction will be reduced by a proportionate amount of the excess. … To illustrate the operation of new section 79D, consider a situation where the amount of a class of income derived by a taxpayer in a year of income from a foreign source is $1000. The deductions allowable under the Principal Act against that income are $200, $400 and $600 under subsections 51(1), 53(1) and 54(1) respectively. The aggregate deductions allowable exceed the amount of income by $200 so that the respective deductions are to be reduced by a proportionate amount of the excess. This means that the deductions allowable in the year of income under subsections 51(1) 53(1) and 54(1) will be reduced respectively by 200/1200 of 200 ie. $33, 400/1200 x 200 ie. $67 and 600/1200 x 200 ie. $100. The reductions equal in total $200, being the amount of the excess." At the same time s 160AFD was likewise amended to ensure that it too related to all relevant deductions and income and required, therefore, a carry forward of foreign losses of the preceding seven years, the overall loss being calculated in effect as the excess of allowable deductions over the amount of the taxpayer's income of a relevant class of income derived from a foreign source. To complete the legislative history it may be noted that a new s 79D was substituted by Act No 5 of 1991, applicable to assessments for the 1990-91 year of income and subsequent years, to deal specifically with the issue which has arisen in the present case. As substituted, the new s 79D spelt out what was to happen where the taxpayer did not derive any assessable foreign income of a class in the year of income. Section 160AFD operated only in relation to pre-1990 losses and s 79E was substituted, inter alia, to allow post 1989 losses against assessable foreign income where an election was made under subsection (6). Interestingly, that subsection used language similar to that originally employed in s 79D, when it provided: "A taxpayer who has derived assessable foreign income in a year of income…" The language is in contrast to the substituted s 79D(1) which provided, inter alia: "The taxpayer did not derive any assessable foreign income of that class in the year or income." The substitution of the new s 79D is, however, expressly directed by Parliament to play no part in the process of interpretation of the old s 79D. Section 51(1) of Act No 5 of 1991 provides that the substitution is to be disregarded for the purpose of interpreting the former s 79D in relation to an assessment to which the old section would be applicable. Thus, while the substitution will not aid an argument advanced by a taxpayer that the previous section was defective and the defect corrected, it will likewise not aid an argument by the Commissioner that the original Parliamentary intention in introducing s 79D could be reinforced by looking to the substitution in 1991. For all present purposes the substitution must be ignored.

The language of s 79D appears clear and unambiguous. It operates in the 1990 year when there is "an amount" of a "class of income" having a foreign source, which is derived. There may be occasions when zero can be regarded as "an amount" cf Federal Commissioner of Taxation v Ryan (1998) 98 ATC 4323 where regard is had to the context, although the result in that case can hardly be thought to have been anticipated by the earlier cases which it swept aside. But, be that as it may, it is, to say the least, difficult to imagine that the words "amount derived" were intended to encompass the case where nothing at all was derived, especially when the context is in relation to a particular class of income having a particular source. It is hard to imagine how a zero amount can be treated as having a particular source, foreign or otherwise. In general terms it may be said that the task of statutory construction is to expose the meaning of the words which Parliament has enacted. This is reflected in the so-called "golden rule" of interpretation of Lord Wensleydale: Grey v Pearson (1857) 6 HLC 61 at 106; 10 ER 1216 at 1234. So it will generally be the case that the Court will give to legislation the ordinary grammatical meaning which it bears. As Gibbs CJ reminded us in Cooper Brookes (Wollongong) Pty Ltd v Federal Commissioner of Taxation (1980-1) 147 CLR 297 at 304: " … it is not unduly pedantic to begin with the assumption that words mean what they say." However, it must likewise be said that even without provisions such as ss 15AA and 15AB of the Acts Interpretation Act 1901, it is fundamental that the statutory construction proceed to give effect to the manifest or expressed intention of Parliament. As McHugh J pointed out in Corporate Affairs Commission (NSW) v Yuill (1991) 65 ALJR 500 at 511 the literal or grammatical meaning may not be the meaning which Parliament intended to enact. Hence there will be a need to depart from the literal meaning where there is good reason so to do: Cooper Brookes at 321. As Mason and Wilson JJ there said: " … It [ie the propriety of departing from a literal interpretation] extends to any situation in which for good reason the operation of the statute on a literal reading does not conform to the legislative intent as ascertained from the provisions of the statute, including the policy which may be discerned from those provisions." In CIC Insurance Ltd v Bankstown Football Club Ltd (1997) 187 CLR 384 at 408, Brennan CJ, Dawson, Toohey and Gummow JJ said: " … the modern approach to statutory interpretation (a) insists that the context be considered in the first instance, not merely at some later stage when ambiguity might be thought to arise, and (b) uses 'context' in its widest sense to include such things as the existing state of the law and the mischief which, by legitimate means such as those just mentioned, one may discern the statute was intended to remedy. Instances of general words in a statute being so constrained by their context are numerous. In particular, as McHugh JA pointed out in Isherwood v Butler Pollnow Pty Ltd (1986) 6 NSWLR 363 at 388, if the apparently plain words of a provision are read in the light of the mischief which the statute was designed to overcome and of the objects of the legislation, they may wear a very different appearance. Further, inconvenience or improbability of result may assist the court in preferring to the literal meaning an alternative construction which, by the steps identified above, is reasonably open and more closely conforms to the legislative intent." Not surprisingly, the Commissioner relies upon the decision of the High Court in Cooper Brookes in support of a construction which the literal language of s 79D does not bear. There is no doubt but that Cooper Brookes is of great significance is elucidating the role of the Court in statutory construction. However, at least some of the comments in that case must be seen by reference to the issue which arose for decision. The question was as to the operation of s 80C(3) of the Act as it then stood. On a literal reading of that subsection the taxpayer fell outside it and thus certain losses were available as a deduction to it. However, s 80C(3) was the result of amending legislation directed at taxation avoidance and the interpretation sought to be given to it by the taxpayer brought about the consequence that the amendment did not effect the purpose it was obviously intended to effect. There was a legislative "mistake" as Gibbs CJ saw it, so that "the expression of its [ie Parliament's] intention miscarried" (see at 306). In the result, the Court preferred the construction which effected the legislative purpose, rather than the construction which otherwise produced an outcome which was irrational and absurd. Even if Cooper Brookes is not to be confined to cases involving the construction of amending legislation directed at tax avoidance, it is obvious that the decision to depart from the literal words which Parliament had enacted was made easier by the context in which the case arose. There are, too, cautions to be found in the various judgments that emphasise that courts can not depart from the literal meaning of words merely because the result may be inconvenient or even seem unjust: see per Gibbs CJ at 305, Stephen J at 310, Mason and Wilson JJ at 321 and see too per McHugh J in Saraswati v the Queen (1990-91) 172 CLR 1 at 22. This must be particularly so in the field of income tax where failure to include in a return an amount of income, or the claiming of a deduction, not otherwise allowable in the context of self-assessment, may bring about, subject now to safe haven provisions, substantial penalties. The legislative history provides no clear answer to the present dilemma, nor do either resort to extrinsic materials nor the legislative context in which s 79D is to be found. There is nothing intrinsically absurd about quarantining a deduction against foreign income when foreign income is derived, but not quarantining a deduction where no foreign income is derived. As is clear, deductions may be incurred at a time prior to income being derived, just as deductions may be incurred and allowable after the income to which those deductions are directed may have ceased. The latter problem was reasonably clear after the decision of the High Court in AGC (Advances) Ltd v Federal Commissioner of Taxation (1974-5) 132 CLR 175, but see Placer Pacific Management Pty Ltd v Federal Commissioner of Taxation (1995) 95 ATC 4459, the former considerably earlier. On the Commissioner's interpretation a deduction would be lost in the latter case and may never be allowed in the former if it should turn out that no foreign income is ever derived. In my view, as relevant to the 1990 year of income, s 79D should be construed in accordance with its ordinary and natural meaning. On this basis there could be no tax benefit in that year. While clearly the version of s 79D as applicable to assessments in the 1991 year would operate to produce a tax benefit in that year, it is difficult to see how it would be concluded that a scheme put into place prior to the 1991 year would be concluded to have a dominant purpose of producing a tax benefit in accordance with legislation not enacted at the time the scheme was put into place. I shall return to that problem later. In the view which I take it is not strictly relevant to deal with the other matters raised by the Applicants both as to s 79D or Part IVA on this branch of the case. However, as it is likely that whatever the outcome the present case will be appealed, it is desirable that I indicate my views on the balance of the argument. Before turning to Part IVA, a further submission related to s 79D should be noted. The above comments have assumed that the interest incurred was related to income from a foreign source (as defined). Not so, say the Applicants. If the scheme had not been entered into or carried out it can be assumed that ACP would have subscribed for the shares in the UK companies and in the result, at some stage, have derived dividend income. However, it is submitted that that dividend income would have had an Australian source for ordinary Australian income tax purposes and, to the extent that it differs, would not be from a foreign source in the defined sense. At the heart of this submission is the view that ACP was a holding company, the activities of which would be holding shares, rather than subscribing for them. Its central management and control and decision making process was in Australia and, as a holding company, the source of its dividend income for ordinary purposes should be seen to be where that central management and control, that is where its decision making process was, namely Australia. Reference is made to the decision of the High Court in Esquire Nominees Limited v The Commissioner of Taxation (1973-4) 129 CLR 177 and Thiel v Federal Commissioner of Taxation (1990) 171 CLR 338. The latter case is relied upon for a slightly different reason by the Commissioner. In the former of these cases, the issue for decision was whether the dividend income of a resident of Norfolk Island, who was the trustee of a trust holding shares in another company resident in Norfolk Island, was to be assessed as if derived from a source in Norfolk Island with the consequence that s 95 of the Act had no operation to include it in the computation of net income of that trust estate. The case accepts both that source of income is a matter of fact as well as the requirement that what must be looked at is what a "practical man would regard as a real source of income": Nathan v Federal Commissioner of Taxation (1918) 25 CLR 183 at 189. Barwick CJ, in holding the source of the dividend in question to be Norfolk Island, placed emphasis on where the fund of profits were out of which the dividends were to be paid (see at 212). So too did Menzies J at 221. Stephen J regarded a number of factors as relevant including the location of the fund of profits out of which the dividend was paid (see at 226). Each of these views would tend against the Applicants' submission, for the fund of profits out of which the dividend from the UK companies would be paid would be overseas, not in Australia. However, the Applicants seize upon the following passage from the judgment of Barwick CJ at 212: "Further a company may make profits without trading in goods or commodities or for that matter in securities. It may make profits simply by investment and may do so though its investment portfolio consists only of shares in one other company or even of all the shares in one other company. In such a case its net income from its investment will be its profits. Further, in my opinion, the place where the company makes its investment income will be the place where it has its central management and control. It will, of course, be different in the case of a company conducting manufacturing or trading activities. In the case of such companies the place where these activities are carried on can be seen in fact to be the geographical source of the profits these activities yield." With respect, this passage ignores both the facts in that case, where the Commissioner was seeking to argue source as dependent upon situs of shares or ultimate funds emanating from Australia in a subsidiary of the holding company, as well as the comments of the Chief Justice which precede this passage and of the other members of the majority of the Court to which reference has earlier been made. The distributed profit in that case, that is to say the dividend from the Norfolk Island holding company, was in Norfolk Island as was also the central management and control. Here the fund out of which dividends would in time be paid was in the United Kingdom, even if the central management and control of ACP was in Australia. In any event, the case had nothing to do with the statutory question which the combined effect of s 79D and the definition of "foreign source" in s 160AFD poses. It is necessary, therefore, to turn to Thiel. Thiel was concerned with the interpretation of the words "profits of an enterprise of one of the Contracting States" in Art. 7 of the Double Tax Agreement between Australia and Switzerland. It was not concerned with the statutory question here raised. For the Commissioner it is said that ACP had a "business, commercial or investment activity carried on … in a foreign country". The Applicants deny for this purpose that ACP carried on an investment activity or for that matter did so (or should be taken to do so) as part of carrying on an activity overseas. Mason CJ, Brennan and Gaudron JJ in Thiel gave to the words "carried on" in the Treaty a meaning not involving repetition, in contradistinction to other places in the Act where the words are used in juxtaposition to "carrying on" and "carrying out". In these other places repetition is indicated. It was, as their Honours observed, necessary to look at the context of the Treaty to reach a conclusion. The context in s 79D is whether an Australian resident derives income from an investment activity which it carried on in a foreign country. It would seem highly unlikely that the legislature intended to quarantine deductions against foreign income where there was no repetition of the activity but not do so where the activity was done once and once only. In my view s 79D should be construed, consistent with the general law of source in any case, to bring about the result that a dividend from an overseas company in which the taxpayer has invested is to be treated as having a foreign source for the purposes of the section. It is not difficult on an ordinary reading of the words to conclude that the dividend which, for present purposes, ACP is to be taken to have derived if it had not taken up redeemable shares in MLG would have been income which it would have derived from an investment activity which it carried on in the United Kingdom, namely the subscription of shares there. The dividend would in the relevant sense be income "from" the activity of subscription and in the result would be taken to have a foreign source. I turn now to consider Part IVA. I do so on the basis that contrary to my view s 79D would, as the Commissioner submits, have applied to ACP even where ACP derived no income from a foreign source in the year of income.

For Part IVA to apply to a scheme entered into or carried out after 27 May 1981, each of the elements of s 177D must be satisfied. Those elements may be said to be:

1. A "scheme" as that expression is defined in s 177A(1) of the Act.
2. The obtaining by a taxpayer of a "tax benefit" as that expression is defined in s 177C of the Act.
3. A conclusion, having regard to the various matters in s 177D(b) as to the dominant purpose of one of the people who entered into or carried out the scheme or part thereof. The word "scheme" is widely defined in s 177A. In a particular case it may extend to an arrangement or course of action or course of conduct or it may consist merely of a single action. Nor would it seem to matter that the scheme identified by the Commissioner is itself but a part of a larger scheme so long as it brings about a tax benefit and is thus susceptible to cancellation under s 177F, although there may be circumstances where the adoption of a smaller scheme as part of a larger scheme might bring about the result that the circumstances treated as the "scheme" are such that they are incapable of standing on their own feet and are thereby robbed of all practical meaning: Federal Commissioner of Taxation v Peabody (1994) 181 CLR 359 at 384. Here there is, on any view of the matter, a scheme. As already indicated, (for present purpose it is unnecessary to distinguish between each of the two schemes upon which the Commissioner relies and the discussion proceeds upon the basis of taking one only of them) the scheme is the acquisition by ACP of redeemable preference shares in MLG and the acquisition by MLG of redeemable preference shares in CPIL(UK).

Tax Benefit Obtained in Connection with the Scheme On the assumption that I am wrong on the proper construction of s 79D there would be a tax benefit which ACP would obtain in connection with the scheme within s 177C(1)(b), namely: "a deduction being allowable [ie. under s 79D] to the taxpayer [ie. ACP] in relation to a year of income where the whole or a part of that deduction would not have been allowable, or might reasonably be expected not to have been allowable, to the taxpayer in relation to that year of income if the scheme had not been entered into or carried out." The test of reasonable expectation required to be satisfied will be satisfied where it can be predicted that if the relevant scheme had not been entered into or carried out ACP would have done something which would give rise to a deduction being allowable to it and the prediction is sufficiently reliable as to be regarded as reasonable: cf Peabody at 385. It is reasonable to expect that had the scheme as defined not been entered into or carried out ACP would either have subscribed for shares in CPIL(UK) or made loans to that company. Neither alternative matters to the present analysis, although I should think it more likely than not that the investment would have been by way of shares, since that was the way the actual investment by MLG into CPIL(UK) was structured. Why would it be reasonable to expect anything else if ACP had invested directly? I have, so far, preceded on the basis that the change of language in s 79D applicable to assessments for the year of income ended 30 June 1991 has no relevance. The Commissioner says it does and that his case is stronger in the 1991 income tax year. In one sense that is true since it is easy enough to say in respect of events which gave rise to the 1991 assessment that if ACP had invested directly in shares in CPIL(UK) it would have lost a deduction under s 79D in its amended form applicable to assessments for the 1991 year of income. It is not true, however, once one turns to the third factor, namely the necessary conclusion.

The Section 177D Conclusion As the High Court pointed out in Federal Commissioner of Taxation v Spotless Services Ltd (1996) 186 CLR 404 the conclusion which is to be drawn under s 177D depends entirely upon objective facts. Subjective motivation will be irrelevant to the conclusion. Hence, whether or not Mr Cherry or others had a purpose of ensuring tax deductibility to ACP of the interest deductions to which it was otherwise entitled would not be relevant in arriving at that conclusion. Likewise the conclusion to be drawn must be a conclusion that a reasonable person would draw: Spotless at 421-2. The conclusion to be reached must relate to the dominant purpose of a person who either entered into or carried out the scheme or a part of it. A purpose will be dominant if it is the ruling, prevailing or most influential purpose: Spotless at 416. Thus where there are two or more purposes objectively evident then the one which is most influential will be the dominant purpose. The fact that the transaction was commercial does not require the conclusion that the dominant purpose would fall outside the part, for there is no true dichotomy between schemes which are commercial and those which are tax driven: Spotless at 415-6. I do not think that the important comment of McHugh J at 425, with which, with respect, I agree, is necessarily at odds with the majority view. When one comes to apply the tests in s 177D(1)(b) it is clear that there are two purposes which can be seen to have driven the scheme as identified by the Commissioner. The use of various tax haven companies would, if one looked at the overall scheme, suggest tax avoidance in the generally accepted meaning of that term. But it is not the wider scheme upon which the Commissioner chooses to fasten. The two purposes which caused the structure to be adopted on the assumption, contrary to my view, that s 79D operated to disallow a deduction in a year of income where no foreign assessable income was derived, were the obtaining of the deduction under s 79D and the adoption of a structure which would not detract from the tax credit relief. Having regard to the definition of "tax benefit" in s 177C the latter was not, at the relevant time, a "tax benefit". The question is therefore which of these two benefits should be seen to be dominant in the relevant sense. The s 79D advantage would be of no particular consequence unless no immediate foreign source income was anticipated to flow by way of dividend from the United Kingdom companies, for once an adequate stream of foreign income flowed, the interest deduction would be available to offset it. Until that time, on the Commissioner's interpretation of s 79D, the interest deduction would not be available to offset anything on the assumption that it could reasonably have been expected that ACP would have invested by way of share subscription in the United Kingdom. Because the structure adopted involved no foreign income being derived, the interest income would be able to be deducted without regard needing to be had to s 79D, and as against Australian source income. Likewise, the tax credit advantage would arise only where foreign income was included directly or indirectly in assessable income of a resident taxpayer. That too could reasonably have been expected to arise in the future. It was, however, not an immediate problem. While it is clear on the facts that a substantial income stream was expected and that there was substantial foreign tax payable on foreign income in subsidiaries of the United Kingdom companies, it was more likely than not that any question of tax credit availability would arise in a year of income subsequent to the year in which the investment was made. With some doubt I am of the view that a conclusion would be drawn that the dominant purpose of some person who participated in the scheme, and in particular those (perhaps not Mr Cherry, but there were others) who advised the group at Arthur Young and later Ernst & Young, was to bring about the result that a deduction would be allowed to the Applicants which, but for the scheme, would have been disallowed to them because of the application of s 79D. I reach this conclusion because it seems to me that the interest deduction was more immediate than the adoption of a neutral structure for non interference with tax credits. It might perhaps be said that one of the problems in the present case lies in artificially dissecting part of a scheme from the totality of the scheme adopted. The arrangement as a whole was directed to a commercial end much more significant than tax. Part of the structure was devised because of tax, but the separating out of the tax and non tax benefits leaves outside the structure both the borrowing of ACP and the subscription of monies for shares by CPIL(UK). That, however, is a consequence of the decision of the High Court in Spotless. It is interesting here to note the problem of adopting the amended version of s 79D and applying to it the Part IVA analysis. All the steps said to constitute the plan were carried out in the 1990 tax year, that is to say prior to the amendment made to s 79D applicable in the year of income ended 30 June 1991. How would it be reasonable to conclude that these steps were designed to attract the amended s 79D when that section was not even law at the time the steps were taken? In my view the application of the Part can only be tested with respect to the obtaining of a tax benefit in accordance with the law as applicable to the time the scheme is entered into or carried out. It can not be tested retroactively after the scheme has been entered into and carried out merely because in the meantime the law has changed in respect of assessments for that year. The time for testing the dominant purpose must be the time at which the scheme was entered into or carried out and by reference to the law as it then stood. It is unnecessary to consider what the case may be if the entry into the scheme and the carrying out of the scheme were in different income tax years, and in one of those years the law differed from that in another. It suffices to consider that case when or if it arises. In summary, and having regard to the construction which I prefer of s 79D, I am of the view that the Applicants should succeed on what has been referred to as the Part IVA/s 79D issue.

The Dividend Stripping Issue Section 177E forms part of Part IVA of the Act. It provides: "(1) Where - (a) as a result of a scheme that is, in relation to a company - (i) a scheme by way of or in the nature of dividend stripping; or (ii) a scheme having substantially the effect of a scheme by way of or in the nature of a dividend stripping, any property of the company is disposed of; (b) in the opinion of the Commissioner, the disposal of that property represents, in whole or in part, a distribution (whether to a shareholder or another person) of profits of the company (whether of the accounting period in which the disposal occurred or of any earlier or later accounting period); (c) if, immediately before the scheme was entered into, the company had paid a dividend out of profits of an amount equal to the amount determined by the Commissioner to be the amount of profits the distribution of which is, in his opinion, represented by the disposal of the property referred to in paragraph (a), an amount (in this sub-section referred to as the 'notional amount') would have been included, or might reasonably be expected to have been included, by reason of the payment of that dividend, in the assessable income of a taxpayer of a year of income; and (d) the scheme has been or is entered into after 27 May 1981, whether in Australia or outside Australia, the following provisions have effect: (e) the scheme shall be taken to be a scheme to which this Part applies; (f) for the purposes of section 177F, the taxpayer shall be taken to have obtained a tax benefit in connection with the scheme that is referable to the notional amount not being included in the assessable income of the taxpayer of the year of income; and (g) the amount of that tax benefit shall be taken to be the notional amount. (2) Without limiting the generality of sub-section (1), a reference in that sub-section to the disposal of property of a company shall be read as including a reference to - (a) the payment of a dividend by the company; (b) the making of a loan by the company (whether or not it is intended or likely that the loan will be repaid); (c) a bailment of property by the company; and (d) any transaction having the effect, directly or indirectly, of diminishing the value of any property of the company." The legislative purpose seems clear enough. The legislature had, it would seem, taken the view that there would be some difficulty in fitting dividend stripping schemes within Part IVA. Both tax benefit and conclusion as to purpose could present problems. To obviate such problems s 177E deemed the Part to apply to dividend stripping schemes and spelt out the consequences, particularly that there was a tax benefit in a particular amount arising from the Part and that there was no need to test the scheme against the conclusion as to dominant purpose. The Commissioner proceeded upon the basis that Part IVA applied to the transfers of shares by CPH and MLG in each of CPIL(UK) and CPIHL(UK) to CPIL(B) and the subsequent liquidation of the two United Kingdom companies, loans, payment of dividends or liquidation distributions. He made determinations under s 177F(1) of the Act to include in the assessable income of CPH for the year ended 30 June 1990 the sum of $69,681,830 described as a deemed dividend from CPIHL(UK) and the sums of $16,757,189 and $49,726,875 described as deemed dividends from CPIL(UK). He likewise made determinations under s 177F(1) including in the assessable income of MLG the sums of $81,748,275 and $27,547,974 described as deemed dividends from CPIL(UK). He did this notwithstanding, as will be seen, that profits available in CPIL(UK) were less than those which the Commissioner considered had been distributed under the scheme. I will return later to the difficulty with the Commissioner's determination. However, as a fall back position if the provisions of s 177E were found to have no application the Commissioner submitted that the ordinary provisions of Part IVA were applicable to enable an amount to be included in the assessable income of MLG and CPH. As can be seen from the provisions of s 177E for that section to have application a number of elements must be present. These are: 1. There must be a scheme as defined in s 177A(1). There is no doubt that the steps identified by the Commissioner constitute a scheme as so defined. 2. The scheme must be either: (a) by way of dividend stripping; (b) be in the nature of dividend stripping; or (c) have substantially the effect of a scheme by way of or in the nature of dividend stripping. 3. As a result of the scheme property must be disposed of. 4. The Commissioner must be able to form the opinion in s 177E(1)(b) that the relevant disposal represented in whole or part a distribution of profits of the company of the accounting period in which the disposal took place, an earlier, or a later accounting period and have actually done so. 5. The hypothesis in s 177E(1)(c) must be satisfied, namely that if before the scheme was entered into a dividend had been paid out of profits to the extent the Commissioner determines were treated as having been distributed (see (4) above) it could reasonably be expected that amounts would have been included in the assessable income of a taxpayer in a year of income. It is necessary to consider whether the last three of these criteria are here satisfied.

Dividend Stripping It is not clear to me what the distinction is between a scheme of dividend stripping, one that is in the nature of dividend stripping and one that has the effect of either. Nor is the concept of dividend stripping one of great clarity. The words have no recognised technical meaning nor are they the subject of definition, presumably because they were thought to be widely understood. The Explanatory Memorandum which accompanied the Income Tax Laws Amendment Bill (No. 2) 1981 which became Act No 110 of 1981 and introduced Part IVA into the Act has this to say about s 177E(1)(a): "Schemes within the category of being, or being in the nature of, dividend stripping schemes would be ones where a company (the 'stripper' purchases the shares in a target company that has accumulated profits that are represented by cash or other readily-realisable assets, pays the former shareholders a capital sum that reflects those profits and then draws off the profits by having paid to it a dividend (or a liquidation distribution) from the target company. In the category of schemes having substantially the same effect would fall schemes in which the profits of the target company are not stripped from it by a formal dividend payment but by way of such transactions as the making of irrecoverable loans to entities that are associates of the stripper, or the use of the profits to purchase near- worthless assets from such associates." 1981 was not the first time the expression had been used in the Act. It had appeared in s 46A(1) as inserted into the Act in 1971 in the context of the words: " …transaction, operation, undertaking, scheme or arrangement that the Commissioner is satisfied was by way of dividend stripping." Reference may be made as well to ss 46B and 52A. In 1970 in Investment and Merchant Finance Corporation Limited v Federal Commissioner of Taxation (1970) 120 CLR 177 at 179 Windeyer J said: "The case arises out of what was called in the evidence a 'dividend-stripping operation'. That term has become well known in English revenue law. I quote from Halsbury's Laws of England, 3rded, vol 20, p 201: 'Dividend stripping is a term applied to a device by which a financial concern obtained control of a company having accumulated profits by purchase of the company's shares, arranged for these profits to be distributed to the concern by way of dividend, showed a loss on the subsequent sale of shares of the company, and obtained repayment of the tax deemed to have been deducted in arriving at the figure of profits distributed as dividend.' So well known has the term become that the second edition of Fowler's Modern English Usage (1965) has a brief explanation, under the heading 'Bond washing and dividend stripping', introduced by - 'Most of us are familiar with these terms, but few know much more about them than that they are devices for the legal avoidance of taxation. In the course of the duel provoked by them between the tax avoider and the legislature they have developed a protean variety of detail, but their essence remains the same.' " At 179 Windeyer J referred to a passage in Simon's Income Tax Law, vol 2, p 192 et seq. and vol 2A, p 679 with apparent approval. There the learned authors say of dividend stripping: "This device was operated by financial concerns which, by a series of transactions, gained control of trading companies having large undistributed reserves of accumulated profits, and obtained for themselves the benefit of a large dividend without incurring any liability to surtax, and later were enabled to create what amounted to an artificial trading loss on which tax relief at the standard rate could be claimed from the Revenue. The method was for the financial concern to buy up the shares of such a company, and so to arrange matters that it received for itself a very large dividend out of the accumulated profits of the company. The next step was for the financial concern to sell the shares for an amount very substantially less than the price paid by it for the shares, and this selling price would normally be the difference between the original purchase price and the dividend, so that the purchase price would have equalled the dividend plus the selling price. The difference between the purchase price and the selling price of the shares, which in effect would be the amount of the dividend, could then be shown in the financial concern's trading accounts as a loss, albeit an artificial one, in respect of which tax relief could be claimed from the Revenue." Although reference to the above cases, generally in the context of the then anti-avoidance section, s 260, which made no reference to dividend stripping, is helpful in identifying in a general way the characteristics of dividend stripping, none turned on the expression itself. There have been numerous cases in Australia that might easily fall within the description of dividend stripping. The words were used by the High Court in reference both to Investment and Merchant Finance Corporation Ltd v Federal Commissioner of Taxation (1971) 125 CLR 249 and Rowdell Pty Limited v Federal Commissioner of Taxation (1963) 111 CLR 106. Schemes such as those in Federal Commissioner of Taxation v Patcorp Investments Ltd (1976) 140 CLR 247 and Brookton Co-operative Society Ltd v Federal Commissioner of Taxation (1981) 147 CLR 441 at 449 have been similarly categorised. In Patcorp Investments, Gibbs J at 300 described arrangements such as those in Bell v Federal Commissioner of Taxation (1953) 87 CLR 548; Newton v Federal Commissioner of Taxation (1958) 98 CLR 1; Hancock v Federal Commissioner of Taxation (1961) 108 CLR 258 as well as Federal Commissioner of Taxation v Ellers Motor Sales Pty Ltd (1972) 128 CLR 602 as cases involving dividend stripping: see R.I. Rosenblum, "Anti-Avoidance: Part IVA of the Income Tax Assessment Act" (1994) 2(5) Taxation in Australia, Red Edition at pp 268-274. Likewise, there is a plethora of case law in the United Kingdom where the expression "dividend stripping" has been used, although not in a statutory context. Many of these cases are dealt with in an article by Mr C J Vincent, "Dividend Stripping: stricto sensu or strictly senseless?", (1989) 24(2) Taxation in Australia, at 82. From them and from a study of the Australian cases, including those to which reference has earlier been made, the learned author concludes that a scheme will only be a dividend stripping scheme if it exhibits the following features: "1. The acquisition of shares in a ('target') company by a party: whether by a sale from the existing members, or via an allotment thereto from the target company. 2. The payment of a dividend, or a liquidator's distribution, to the purchaser out of target company profits: whether pre or post acquisition; and whether distributed immediately, or over a period of time. 3. The purchaser claiming not to be taxable upon the distribution received thereby. This claim may have ensued from - (a) the adoption of a tax accounting method which excludes dividends qua income; (b) the application of an express statutory exempting provision; (c) the allowance of an inter-corporate dividend rebate; or (d) the allowance of a countervailing deduction for a loss on the disposition, or diminution in value, of the shares in question. 4. The vendor shareholders, or the target company, obtaining a capital sum not substantially less in monetary terms than the quantum of profits distributed to the purchaser." To be fair the author does not suggest that these are an exhaustive list of the features of a dividend stripping arrangement, merely the minimum factors which have been present in a transaction to which courts have applied the term "dividend stripping". In the course of argument senior counsel for the Commissioner submitted that there were a number of characteristics of a dividend strip which were present here. Those characteristics listed were: 1. The size of the dividend (or profits) relative to other assets. 2. The fact that the shares in the United Kingdom companies were expected, if not dealt with, to be productive of profits in the future. 3. The fact that there were profits available to pay a dividend. 4. That if a dividend were paid that dividend would end up in the assessable income of Australian resident shareholders. 5. The sale of shares by the Australian resident shareholders in a way which realised the value of the profits in a capital form and left those profits to go to a company in the Bahamas. It was submitted that although many of the dividend stripping schemes in the cases arising both in the United Kingdom and Australia had produced advantages to the stripper, eg loss deductions, rebates etc, that was not an essential ingredient in a dividend stripping scheme. I agree: cf Collco Dealings Ltd v Inland Revenue Commissioners [1962] AC 1. Nor would it matter that the so called "stripper" was related to the shareholders rather than independent of them: cf Ellers Motor Sales at 623 per Walsh J, although what is there said is rather in the context of s 260 of the Act than in the context of the statutory expression "dividend stripping". In my view a scheme will be a dividend scheme or in the nature of a dividend scheme if a reasonable observer looking at the transaction would say of it that its essential character is dividend stripping. What is that essential character? Clearly it involves a company which is pregnant with accumulated profits out of which a dividend would reasonably be likely to be declared or has already been declared or where the company is about to receive profits in the future out of which a dividend would reasonably be likely to be declared. In each case the result would be that shareholders would become liable to pay tax on those dividends. It involves the readying up of that company for sale, either by converting its assets into cash or purchasing back operating assets so that the substantial assets of the target company are cash or loans back. (The fact that steps had been taken before contacting a stripper would hardly, however, remove the transaction from the category of dividend stripping). It involves the sale, or allotment of shares in the target company to the stripper. It involves the subsequent payment of a dividend to the stripper by the target company or a deemed dividend of the kind referred to in s47 of the Act, so as to recoup the stripper for the outlay for the shares. As presently advised I see no reason why the fact that the consideration for the sale of shares in the target company is not cash but an allotment of shares necessarily excludes, as the Applicants suggest, the scheme from being in the nature of dividend stripping. Obviously not all sales of shares, even if cum dividend, are in the nature of dividend stripping. Nor is the sale of 100% of shares in a company necessarily dividend stripping, even if the company has accumulated profits. What is missing in the first case and may be missing in the second is the conclusion that an objective observer would reach as to why the scheme has taken place. For a scheme will only be a dividend stripping scheme if it would be predicated of it that it would only have taken place to avoid the shareholders in the target company becoming liable to pay tax on dividends out of the accumulated profits of the target company. It is that matter which distinguishes a dividend stripping scheme from a mere reorganisation. In my view an objective examination of what took place here would not lead to the conclusion that there was a dividend stripping scheme, or for that matter a scheme in the nature of dividend stripping, if that is a significantly different thing. At least one of the United Kingdom companies did have substantial accumulated profits - that much is clear. Both also had substantial investments in overseas companies from which dividends could be derived. The United Kingdom companies had no need to distribute accumulated profits. Any accumulated profits could have sat there forever. The sale of shares and subsequent liquidations were brought about not to enable the shareholders to receive capital instead of dividend distributions, although that was one consequence of what happened, but as part of a reorganisation of the United Kingdom companies for reasons which had to do with United Kingdom and Australian tax other than in respect of dividends which might be derived from the accumulated profits by way of dividend.

Does the Scheme have the Effect of a Dividend Stripping?

However, there is an alternative to be considered, namely, whether, although there was no scheme properly to be characterised as a dividend stripping scheme, nevertheless there was a scheme or there were schemes which had substantially "the effect of" a scheme by way of or in the nature of a dividend stripping. This alternative requires the focus to move from the nature of the scheme and its essential character to a focus on effect. Some idea of what Parliament contemplated would be encompassed within the category of schemes having the effect of dividend stripping schemes is to be gleaned from the Explanatory Memorandum which accompanied the Income Tax Laws Amendment Bill (No 2) 1981. It is there stated: "In the category of schemes having substantially the same effect would fall schemes in which the profits of the target company are not stripped from it by a formal dividend payment but by way of such transactions as the making of irrecoverable loans to entities that are associates of the stripper, or the use of the profits to purchase near worthless assets from such associates." If it be necessary for a scheme to be in the ordinary sense, dividend stripping that there be an actual dividend, and it is unnecessary to decide this (although the Explanatory Memorandum proceeds on the basis that a scheme where profits are taken out on liquidation involving a deemed dividend under s 47 of the Act would be encompassed within dividend stripping as that term is ordinarily used), then no doubt a scheme in which the accumulated profits are stripped by a liquidation distribution, rather than a dividend in the usual sense, would likewise be one, the effect of which would be the same as dividend stripping, although the essential character of the scheme might be said to be different. It may be noted that the relevant "effect" has only to be "substantially" the same as a scheme by way of or in the nature of dividend stripping. Just what the word "substantially" adds is likewise not clear. Perhaps all that need be said of it is that while some of the effect of the scheme may be different the overall effect will be either virtually the same or, at least, to a large extent, the same. I do not, however, accept the submission of the Applicants that there is no difference between the purpose of a scheme and its effect, so that there is likewise no difference between a dividend stripping scheme and one having substantially the effect of a dividend stripping scheme. No doubt for the purposes of applying s 260 of the Act is was true to say that in that context purpose and effect had no real difference in meaning: Federal Commissioner of Taxation v Newton (1956-7) 96 CLR 577 at 630 per Williams J, approved by the Privy Council in the same case at (1958) 98 CLR 1 at 8. But there the issue was the purpose or effect of a contract, agreement or arrangement. Here purpose plays no part of the statutory language, but is present only so far as it aids the characterisation of the scheme. In my view there is a clear difference between a dividend stripping scheme on the one hand and one that has the effect of such a scheme on the other. In my view the relevant "effect" is to be judged by reference to the vendor of the shares in the target company and the target company itself, although there may be some relevance in the effect of the scheme upon the purchaser.

What then was the effect of the scheme? Before the sale of the shares each of the Applicants had shares in CPIL(UK) and CPIHL(UK). As at 30 June 1989 CPIHL(UK) had, according to its audited accounts to that date, accumulated profits of US$86,825,000. Given that the consolidated balance sheet of the company showed a lesser figure for accumulated profits it may be assumed that unless the company continued to trade, it was unlikely that it would generate more profits. As at the same date CPIL(UK) had accumulated losses of US$69,449,000. Even more losses were to be found in the consolidated accounts as at the same date. In the period from 1 July 1989 to 31 December 1989 CPIL(UK) derived an operating profit after tax in the sum of US$65,147,000 which, while available to be used as the source of a company law dividend could, if not paid out as a dividend, be used to replace lost share capital. Since, ultimately, the directors of CPIL(UK) purported to pay a dividend of $US100,000,000 it may be assumed that to that extent, at least, a fund of profits was available, albeit for the current year, out of which a dividend could have been paid. In the period from 1 July 1989 to 30 September 1989, CPIHL(UK) had an operating profit after taxation of US$17,557,000 bringing retained profit carried forward to US$94,048,000. As at 10 May 1990 an unaudited balance sheet of CPIHL(UK) showed accumulated profits of US$33,456,205.29, although the discrepancy is not easily explained, it might take into account the dividend that had been declared and was payable on 8 May 1990, that is to say, two days before the balance sheet was drawn up. The purchase price of the shares (that is to say, ultimately, the number of shares to be allotted by way of consideration) which the Applicants transferred in CPIL(UK) and CPIHL(UK) to CPIL(B) and CPIHL(B) was calculated by reference to the net value of the assets of the companies transferred. On the assumption that, as the Applicants submit, no dividend was in law ever properly declared by the directors of CPIL(UK) and CPIHL(UK) so as to create a debt in favour of any shareholder (this is a matter discussed later) no actual dividend ever reached the hands of the purchasers, CPIHL(B) and CPIL(B). However, the purchasers subsequently received a distribution of assets from CPIL(UK) and CPIHL(UK) by way of a transfer of assets in the liquidation. In the case of CPIL(UK) it is difficult having regard to the fact that it had a negative balance in its accumulated profits account, although a current year profit, to see that the effect of any scheme was one of dividend stripping. By the time the scheme was undertaken no further profits were to be earned, and the current year's profit could be offset by prior years' losses. The matter is different in the case of CPIHL(UK), where as well as current year profits there were considerable accumulated profits, because the result was that the shareholders received capital for their shares in an amount including the amount standing to the credit of the accumulated profits account, and as a result of the liquidation there was a distribution in specie to the purchaser. The accounts of the purchaser are not in evidence, so that it is unknown whether that company ever had any profits as a result of the transaction, although it can be assumed that it did not, for it could bring the assets acquired in specie into its account by reference to the cost of shares acquired, ensuring that it had no accumulated profits in a company law sense.. The taxation position may well have been different, cf Harrowell v Federal Commissioner of Taxation (1967) 116 CLR 607. I am, however, of the view that the scheme, but only in respect of CPIHL(UK), was capable of being seen as one having the effect of a dividend stripping scheme such as to make s 177E applicable, but subject to the other matters with which that section is concerned. It can be said that there was no particular need for CPIHL(UK) to declare dividends. Had it been kept in existence it could have chosen to remain with accumulated profits for the foreseeable future. However, I can not see how that bears on the effect of the scheme, even if relevant to the purpose of the scheme.

Disposal Although it is argued to the contrary on behalf of the Applicants, it is hard to see that there was no disposal of property of the United Kingdom companies, either within the defined sense in s 177E(2) or in the ordinary sense of the words. It may be correct to say, as the Applicants do, that there was no payment of dividends in the company law sense, a matter which I will deal with briefly later. But there were, if no dividends paid, distributions of assets in the liquidation in specie which clearly enough are disposals. Further, those disposals resulted from the scheme. A formal submission was made that s 177E is concerned only with a disposition of an asset of a company of which it is the beneficial owner and that accordingly the section had in accordance with the United Kingdom case of Ayerst (Inspector of Taxes) v C & K (Construction) Ltd [1976] AC 167 no application because on liquidation the company had ceased to be the beneficial owner of its assets. First I do not think s 177E should be so construed, for to do so would make nonsense of the clear legislative purpose to include schemes involving liquidation distributions in the category of dividend stripping schemes. Second, there is, perhaps a conflict between the House of Lords decision in Ayerst and the decision of Menzies J in Franklins Selfserve Pty Ltd v Federal Commissioner of Taxation (1970) 125 CLR 52 at pp 70-1 which the High Court refused to resolve in Federal Commissioner of Taxation v St Hubert's IslandPty Ltd (1978) 138 CLR 210. As presently advised I would prefer the view of Menzies J in the present context. A formal submission that St Hubert's Island itself was wrongly decided may be noted as well. It is binding upon me. Was a Valid Dividend Paid? Senior Counsel for the Applicants relied upon an argument that the purported declarations of dividends by the United Kingdom companies resolved upon by the directors were void. This was in support of an argument that in the present case no dividends were ever paid and there was thus no disposal. The argument depends upon the Articles of Association of each company which at relevant times were in the form of Table A to the Companies Act 1948 or 1981 as the case may be. Under each of these Acts, Table A provides that the Directors have power only to pay interim dividends. The declaration of a dividend and the amount of that dividend, subject to the dividend not exceeding an amount recommended by the directors, is a matter left to an ordinary resolution of members. The argument receives support from the decision of the High Court in Brookton Co-operative at 454-6 where it was held that a dividend purportedly declared by the directors of a company having comparable articles was invalid and thus did not create a debt owing by the company to the shareholders. Although it is clear enough that had the shareholders realised the invalidity they would have passed a resolution declaring a dividend, they did not do so. They never turned their mind to the question. That, in my view, distinguishes the present case from one where acts of the shareholders might be taken to ratify an ultra vires act. I do not think it ultimately matters to the outcome of the present appeals to reach a conclusion on the point. If the resolution was valid, then the actions which occurred subsequently involved a payment of a dividend and the effect of the scheme is more clearly dividend stripping. If, on the other hand, the resolution was, as the Applicants submit invalid, then the distribution of assets were distributions by the company in liquidation. In either case the result of the scheme in respect of CPIHL(UK) was dividend stripping. In the case of that company there was also a disposal resulting from the scheme. The Commissioner's Opinion - s 77E(1)(b) It is clear from the language of s 177E(1)(b) that for s 177E to operate so as to deem the scheme to be one to which Part IVA applies the Commissioner must form an opinion that the disposal represents a distribution of profits of the target company, although the relevant profits may be of the accounting period in which the disposal occurred, or of any earlier or later accounting period. The opinion is relevant, as well, to the operation of para (c), which makes it clear that the Commissioner must determine a figure to be the amount of profits represented by the disposal. It is submitted on behalf of the Applicants that the formation of an opinion by the Commissioner was vitiated in law and, for that reason, whether otherwise there was a dividend stripping, the provisions of Part IVA had no operation. It is necessary to turn to what the Commissioner did. In each determination which the Commissioner made, he first identified the facts and concluded that there was a scheme, that it involved dividend stripping, and that there was a disposal. He then turned to the question whether the disposal represented a distribution of profits. The determination continues (virtually identical language is used in each case): "Paragraph (b) of sec. 177E(1) contains a further requirement that, in the opinion of the Commissioner, the disposal of the relevant property represents, in whole or in part, a distribution to a shareholder or another person of profits of the company. This will occur in the accounting period in which the disposal occurred or any earlier or later accounting period. The Commissioner contends that there was [sic] sufficient profits out of which the dividend could have been paid. In support of the above, it should be noted that the net assets of the two companies [after adding back the dividends of $153m to CPIL (Bahamas) from their Balance Sheets at 10 May 1990 were: CPIL(UK) 86,456,205US$ CPIHL(UK) 103,282,414US$ 189,738,619US$ @ 0.7606 = $249,459,136 AUS. Further, when the dividend is deemed to have been paid out of profits, the Commissioner contends that at the time, it is not necessary for there to be sufficient profits available for the payment of a dividend, as s177E(b) also refers to future profits. It is the Commissioner's contention that the above net assets of both companies represents profits available for distribution." The determination continues by referring, inter alia, to MacFarlane v Federal Commissioner of Taxation (1986) 86 ATC 4477 at 4482-3 and Federal Commissioner of Taxation v Slater Holdings Ltd (1984) 156 CLR 447 as to the meaning of profits. While as Gibbs CJ, with whose reasons Mason, Brennan and Deane JJ agreed, observed in the latter case at 460 the question of what is profit may need to be answered in accordance with the circumstances of the particular case, a starting point is the definition of Fletcher Moulton LJ in In re The Spanish Prospecting Co Ltd [1911] 1 Ch 92 at 98, quoted by his Honour: " 'Profits' implies a comparison between the state of a business at two specific dates usually separated by an interval of a year. The fundamental meaning is the amount of gain made by the business during the year. This can only be ascertained by a comparison of the assets of the business at the two dates." Neither case on any view of the matter permits a conclusion to be reached that the word "profits" equates to net assets. One thing that is not capable of being a profit is paid up capital. Apart from it being conceded that the Commissioner had the names of the United Kingdom companies reversed, so that the figures shown, for example, as the net assets of CPIL(UK) are, it would seem, the net assets of CPIHL(UK) and vice versa, it is now also conceded that CPIL(UK) did not on any view of the matter have profits sufficient to support treating any disposal of assets as representing a distribution of profits. The Commissioner was only able to get to the result he did by pooling the assets of the two companies, when clearly he had to consider each separately for the purposes of s 177E(1)(b). The fact that the paid up share capital of CPIL(UK) was, as at 10 May 1990 US$546,819,650 and that of CPIHL(UK) was US$103,252,414 hardly seems to have disturbed the Commissioner from the arbitrary course which he pursued, even if it is true that each company had more net assets than are shown on the determination. What the determination makes clear is that the Commissioner never turned his mind to the question of what amount of profits existed in each company, let alone how much thereof was represented by the relevant disposal which he identified. The determination was made at a time when it was known that each target company had been liquidated. Once liquidated it could never have more profits than it had immediately before. It is clear that the discretion of the Commissioner under s 177E(1)(b) miscarried. As Part IVA can not apply to a case under s 177F unless the Commissioner has formed the relevant opinion, and as he did not, the assessments, so far as they are dependent upon s 177E must be set aside.

The Statutory Hypothesis - s 177E(1)(c) It is not necessary to pursue the statutory hypothesis in s 177E(1)(c) having regard to the failure of the Commissioner properly to form the opinion under s 177E(1)(b) as the two matters run together. It suffices to say that it could be expected that the full amount of profits available for dividend (although certainly not the paid up capital of the companies) would have been included in the assessable income of the Applicants if a dividend had been declared by each United Kingdom company immediately before the scheme was entered into.

Can the Commissioner Now Utilise s 177D? It is the Commissioner's submission that if he fails to succeed under s 177E he can in this Court support the assessment under s 177D. With respect, there are considerable difficulties in the Commissioner's path. First, it is hard to see, without the aid of s 177E how it can properly be said that there was a tax benefit to the Applicants as that expression is defined in s 177C. In relation to the year of income in question it can hardly be said that if the sale etc of the shares had not been entered into or carried out the Applicants would have received dividends. There was not the slightest need for the United Kingdom companies to pay a dividend in that year and certainly nothing in the evidence to suggest that this was remotely likely to happen. Nor do I think the conclusion under s 177D could reasonably be drawn as to dominant purpose. I do not think that it is necessary to consider each of the separate matters in that section having regard to the absence of tax benefit. There is another difficulty - the Commissioner's exercise of discretion under s 177F was dependent upon treating the transaction as one falling within s 177E and thus having the deeming features which that section postulated. The Commissioner clearly did not apply his mind to the matters in s 177F by reference to Part IVA without the application of s 177E. It may be the case that this is not fatal since Part IVA (and in particular s 177D) itself applies without the exercise of discretion. Further, although ss 177F(1) and (3) proceed on the basis of an exercise of discretion, the existence of the discretion is not made to depend upon the Commissioner's opinion about tax benefit or that such a benefit results in connection with the scheme, for these are matters of objective fact: cf Peabody at 382. While it may be thought to be extraordinary as a matter of administrative law that the exercise of a discretion by reference to wrong matters may nevertheless be a valid exercise of discretion, that proposition may flow from Peabody itself. It is unnecessary in the present case to decide whether the exercise of discretion under s 177F would be valid where it is considered in respect of Part IVA being made applicable by s 177E, when Part IVA indeed applies, but because of s 177D alone. It is preferable to await a case where the outcome depends upon it. Conclusion I am of the view that the Commissioner's assessments based on the provisions of Part IVA/s 177E have been shown to be excessive. Likewise those based on the Part IVA/s 79D issue. The objection decisions relating to the matters dealt with in these reasons will thus be set aside and allowed. Because, however, other matters, initially at issue between the parties and the subject of the objection decisions have been settled, I propose to stand the matter over until a date to be fixed with counsel and direct the Applicants to bring in short minutes of order to reflect the settlement as well as the matters subject to my reasons. Prima facie the Respondent should pay the Applicants' costs of the present proceedings. However, I will hear submissions, if required, on this question. I certify that this and the preceding forty four (44) pages are a true copy of the Reasons for Judgment herein of the Honourable Justice Hill

Associate: Dated: 13 October 1998 Counsel for the Applicants: Mr D H Bloom QC Mr R F Edmonds SC Mr A J Payne

Counsel for the Respondent: Mr B J Shaw QC Mr G T Pagone QC Mr G J Davies Mr M Moshinsky

Solicitor for the Respondent: Australian Government Solicitor

Date of Hearing: 21 to 28 September 1998

List of Companies, Trusts and Their Initials ACP Australian Consolidated Press Limited BAT BAT Industries Plc CPF Consolidated Press (Finance) Limited CPH Consolidated Press Holdings Limited CPI(Sing) CP Investment (Singapore) Pte Ltd CPIF Consolidated Press International Finance Plc CPIHL(B) Consolidated Press International Holdings Limited (incorporated in the Bahamas) CPIHL(UK) Consolidated Press International Holdings Limited (incorporated in the UK) CPIL(B) Consolidated Press International Limited (incorporated in the Bahamas) CPIL(UK) Consolidated Press International Limited (incorporated in the UK) CPUS Consolidated Press US Inc MLG Murray Leisure Group Pty Limited