FEDERAL COURT OF AUSTRALIA

Federal Commissioner of Taxation v Brewing Investments Ltd [2000] FCA 920

INCOME TAX – assessable income – liquidation distribution received by the taxpayer, an Australian resident company, where distribution made out of liquidation distribution received by a non-resident company from liquidator of another non-resident company – whether distribution to taxpayer company made by liquidator out of income – whether s 47(1) of the Income Tax Assessment Act 1936 (Cth), by deeming liquidation distribution to be a dividend paid out of profits, applied to deem liquidation distribution to the taxpayer to be made out of income.

Income Tax Assessment Act 1936 (Cth) ss 44(1), 47(1)

WORDS AND PHRASES:  “for the purposes of this Act”,  “dividends paid ... out of profits”

Harrowell v Commissioner of Taxation (1967) 116 CLR 607 considered

Gibb v Commissioner of Taxation (1966) 118 CLR 628 discussed

Commissioner of Taxation (NSW) v Stevenson (1937) 59 CLR 80 referred to

Hill v Permanent Trustee Co of New South Wales Ltd [1930] AC 720 referred to

Inland Revenue Commissioners v George & Burrell [1924] 2 KB 52 referred to

Federal Commissioner of Taxation v Uther (1965) 112 CLR 630 discussed

Federal Commissioner of Taxation v Blakely (1951) 82 CLR 388 discussed

Archer Bros Pty Ltd (in liq) v Federal Commissioner of Taxation (1953) 90 CLR 140 discussed

Federal Commissioner of Taxation v W E Fuller Pty Ltd (1959) 101 CLR 403 discussed

Inland Revenue Commissioners v Blott [1920] 2 KB 657 cited

Parke Davis & Co v  Commissioner of Taxation (1959) 101 CLR 521 discussed

Glenville Pastoral Co Pty Ltd (in liq) v Federal Commissioner of Taxation (1963) 109 CLR 199 considered

THE FEDERAL COMMISSIONER OF TAXATION v

BREWING INVESTMENTS LTD (ACN 004 233 005)

V 64 of 2000

HILL, HEEREY AND SUNDBERG JJ

10 JULY 2000

SYDNEY

ON APPEAL FROM A JUDGE OF THE FEDERAL COURT OF AUSTRALIA

THE COURT ORDERS THAT:

1.         The appeal be allowed.

2.         The judgment appealed from be set aside and in lieu thereof it be ordered that:

            (i)         the objection decision appealed from be affirmed;

            (ii)        the applicant pay the respondent Commissioner’s costs.

3.         The respondent pay the appellant’s costs of the appeal.

Note:    Settlement and entry of orders is dealt with in Order 36 of the Federal Court Rules.

ON APPEAL FROM A JUDGE OF THE FEDERAL COURT OF AUSTRALIA

REASONS FOR JUDGMENT

HILL J

1                     Just as it seems that the present system of business taxation involving the concept of assessable income less allowable deductions is to be replaced by what is said to be a more commercially acceptable approach to the computation of taxable income, the Court is presented with an appeal which concerns issues which have bedevilled the present conventional system of taxation for at least a century and yet have still not been completely resolved.

2                     The present appeal is an appeal from the judgment of a Judge of this Court.  The facts are agreed.   It raises the issue of the operation of s 47(1) of the Income Tax Assessment Act 1936 (Cth) (“the Act”) where the assets of a non-resident company with accumulated profits (largely resulting from the derivation of non Australian income) are distributed to its parent company on liquidation, the parent also being not a resident of Australia, and where in turn that company is liquidated and a distribution is made by the liquidator to the ultimate parent company which is a resident of Australia.  The learned primary Judge held that the ultimate parent company, the respondent, Brewing Investments Ltd (“BIL”) was not required to include in its assessable income for the year ending 30 September 1993, a substituted accounting period adopted in lieu of the year of income ending 30 June 1993, any amount, whether under s 47(1) directly, or in combination with s 44(1) of the Act.  It is from this judgment that the appellant, the Commissioner of Taxation (“the Commissioner”), appeals.

The agreed facts

3                     The following statement of facts is taken from the judgment under appeal.

“BIL (formerly Henry Jones Investments Limited) is an Australian resident company for the purposes of the Act.  It was the sole beneficial owner of all the issued share capital of Rowsom Investments Limited (“Rowsom”).  Rowsom was in turn the sole beneficial owner of all the issued share capital of Clarkson Holdings Limited (“Clarkson”); and Clarkson was the sole beneficial owner of all the issued share capital of Elders Investments Limited (“EIL”).  None of Rowsom, Clarkson and EIL was a resident company of Australia for the purposes of the Act.  They were residents of unlisted countries for the purposes of Part X of the Act.  Rowsom and Clarkson were incorporated in the Turks and Caicos Islands.  EIL was incorporated in Bermuda.  All three companies had their central management and control in Hong Kong.  Neither Rowsom nor Clarkson derived any relevant income from Australian sources.  Although EIL derived income from both Australian and non-Australian sources, the Commissioner expressly disavowed any reliance in this appeal on that fact.

Clarkson was placed in voluntary liquidation on 3 June 1991.  On 10 June 1991, Clarkson’s liquidator distributed net assets amounting to US$52,013,763 to Rowsom.  For accounting purposes, the distribution was treated as a capital receipt in Rowsom’s books of account.  In the books of Clarkson, the distribution was represented by the following shareholders’ funds accounts:

Clarkson’s Retained Earnings consisted of Income Reserves amounting to US$10,415,507 and Capital Profits Reserves (on the sale of shares and exchange gains) amounting to US$25,517,590.

In September 1993, EIL was placed in voluntary liquidation and, on 24 September 1993, EIL’s liquidator distributed net assets amounting to US$62,264,631 to Rowsom.  For accounting purposes, the distribution was treated as a capital receipt in Rowsom’s books of account.  In the books of account of EIL, the distribution was represented by the following shareholders’ funds accounts:

EIL’s Retained Earnings consisted wholly of Income Reserves.  These Reserves included profits which had been accumulated and not distributed before the date of liquidation.

In September 1993, Rowsom was placed in voluntary liquidation and, on 24 September 1993, Rowsom’s liquidator distributed net assets amounting to US$70,238,424 to BIL.  For accounting purposes, the distribution was treated as a capital receipt in BIL’s books of account.  In the books of account of Rowsom, the distribution was represented by the following shareholders’ funds accounts:

As at 24 September 1993, US$70,238,424 was equivalent to Aus$107,299,762.  Rowsom’s Capital Profits Reserve contained the distributions by the liquidators of EIL and Clarkson.  Those distributions comprised:

As already noted, in its 1993 return, BIL included in its assessable income an amount of Aus$48,599,855, which was equivalent to US$31,813,465, made up of:

The Capital Profits Reserve was made up of all of EIL’s reserves (US$17,038,103), represented as income in EIL’s profit and loss account, and that part of the distribution made by Clarkson’s liquidation which represented interest income (US$10,415,507).  It is not in contest that the balance of the distribution from Rowsom to BIL (US$38,524,959) was a return of capital and, on no view, was deemed dividends.

As we have seen, BIL subsequently departed from this approach.  It now claims that the only amount properly to be included in its assessable income is the amount that represents the retained earnings of Rowsom, namely, US$4,359,855.  The Commissioner declines to accept the validity of BIL’s altered approach.”

The statutory provisions

4                     Income tax as assessed in accordance with the Act is imposed upon taxable income, that being the result reached when there is deducted from assessable income all allowable deductions.  The starting point for the computation of assessable income is s 25 which, in the case of a resident taxpayer, includes the income of that taxpayer derived directly or indirectly from sources within or outside Australia.  In the case of a non-resident it will be only such income as is derived from sources in Australia that are included in assessable income.  Income derived by a non-resident from sources wholly outside Australia is exempt income under s 23(r).

5                     Section 44(1) of the Act is concerned with dividends.  It provides:

“The assessable income of a shareholder in a company (whether the company is a resident or a non-resident) shall, subject to this section and to section 128D –

(a)       if he is resident – include dividends paid to him by the company out of profits derived by it from any source; and

(b)       if he is non-resident – include dividends paid to him by the company to the extent to which they are paid out of profits derived by it from sources in Australia.”

6                     At the relevant time s 47(1) of the Act provided:

“Distributions to shareholders of a company by a liquidator in the course of winding up the company, to the extent to which they represent income derived by the company (whether before or during liquidation) other than income which has been properly applied to replace a loss of paid up capital, shall, for the purposes of this Act, be deemed to be dividends paid to the shareholders by the company out of profits derived by it.”

The judgment appealed from

7                     As I have already noted the learned primary Judge rejected the submission of the Commissioner that s 47(1) of the Act, which was without territorial limitation, operated recursively to make successive distributions on the winding up of a series of interrelated companies assessable pursuant to s 44(1).  It was the Commissioner’s argument before her Honour, as before us on appeal, that in consequence the distributions by the liquidators of Clarkson,  EIL and Rowsom all constituted income, deemed by force of s 47(1) of the Act to be dividends paid out of profits with the consequence that the ultimate liquidation distribution by Rowsom to BIL constituted assessable income pursuant to s 44(1)(a).

8                     In her Honour’s view the case of the Commissioner rested largely on whether the  case fell within the decision of the High Court in Harrowell v Commissioner of Taxation (1967) 116 CLR 607.  The case of BIL on the other hand depended upon what was said by the majority of that Court in Gibb v Commissioner of Taxation (1966) 118 CLR 628.  After a discussion of the case law and, in particular, the two cases just mentioned, her Honour found that Harrowell was inapplicable to the facts before her.  The decision in Harrowell, in her Honour’s view, depended upon the fact that the two companies in the chain, the liquidation of which resulted in the ultimate distribution to the taxpayer, were residents of Australia so that s 47(1) operated at each stage together with s 44(1) of the Act so as to treat each liquidation distribution as being included in the assessable income of the recipient.  Section 47(1) was not, in her Honour’s view, intended to operate on its own to treat a liquidation distribution as income in circumstances where s 44(1) had no application to bring the distribution into assessable income.

9                     Her Honour’s decision is best summarised in the following passage from the judgment, intended as a summary of the reasons:

“As I have already indicated, I reject the Commissioner’s submissions as to the operation of s 47(1) with respect to the distributions from Clarkson and EIL to Rowsom because at the time those distributions were made there could be no deeming ‘for the purposes of [the]Act.  At the time those distributions were made, neither s 44(1) nor any other relevant provision of the Act had any relevant application to the distributions from Clarkson and EIL to Rowsom.  As the Commissioner properly conceded, neither s 44(1) in conjunction with s 47(1) nor any other provision of the Act rendered the receipt of those distributions assessable income.  Critically, for the purposes of this appeal, the distributions from Clarkson and EIL did not give rise to a receipt by Rowsom in the nature of income in the ordinary sense: see Burrell and the authorities cited earlier in connection with Burrell.  Nor did those distributions give rise, for the purposes of the Act, to deemed dividends pursuant to s 47(1).  In this circumstance, as BIL submits, Harrowell can have no direct application.  If those distributions did not represent income in Rowsom’s hands, then the subsequent distribution by Rowsom to BIL did not represent income derived by Rowsom; and, in consequence, that distribution could not, for the purposes  of the Act, be deemed to be dividends paid to BIL out of profits derived by Rowsom, by force of s 47(1) or otherwise.”

10                  Senior counsel for BIL relies upon and urges the Court to adopt her Honour’s reasoning as expressed in this passage.

The Commissioner’s submissions

11                  The Commissioner at the close of argument on the appeal summarised his submissions in writing as follows:

1.         Section 47(1) is a provision dealing with a transaction according to its character.

2.         Section 47(1) has an application which is independent of considerations of locality.

3.         Section 47(1) deems what the shareholder receives from the liquidator to be, for all purposes of the Act, income derived by the shareholder.

4.         The operation of s 47(1) is not confined to the bringing of distributions within s 44(1).

5.         These propositions can be combined as follows:

Where – as in the present case – a question arises about whether s 47(1) applies to a liquidator’s distribution to a resident shareholder (so as to render the distribution assessable income of the shareholder under s 44(1)(a)) the following rule applies:

‘If the distribution in question has its source in an earlier liquidator’s distribution, then if the earlier distribution is of the character defined by s 47(1) – that is, it represents income derived by the company making the first distribution – then it necessarily follows that the later distribution is a distribution to which s 47(1) applies’.”

The submissions of BIL

12                  As I have already indicated, senior counsel for BIL relied upon the reasons which the learned primary Judge has given.  In reply to the written submission which I have set out, he joined issue with the Commissioner on the fourth submission, which, if incorrect, would result in the rejection of the conclusion as stated in the fifth submission.  It was the submission for BIL that s 47 had no operation, that is to say, did not apply, either alone or with s 44(1), to the distributions to Rowsom.  That company, as a non-resident, had no taxable income under the Act and no tax payable.  Thus when the liquidator of Rowsom distributed its assets to BIL there was a distribution out of capital, not income, with the consequence that s 47(1) had no application, alone or in combination with s 44(1), to include any amount in the assessable income of BIL.

13                  To understand and evaluate the competing submissions it is necessary to discuss in some detail  the historical background to s 47(1) and the many cases in which the High Court has analysed it or the principles underlying it.

Discussion of the cases bearing upon s 47(1) of the Act

14                  As is well known s 47(1) had its origin in the differences that exist in a system of taxation of income between the liability for tax of a person in receipt of dividends and the liability for tax of a person in receipt of a liquidator’s distribution.

15                  As the joint judgment of Rich, Dixon and McTiernan JJ in  Commissioner of Taxation (NSW) v Stevenson (1937) 59 CLR 80 at 97-100 points out, in a system directed at the taxation of income it could be possible to treat what is received by a shareholder, other than an original shareholder, as, at least in part, a return to the shareholder of a part of the amount paid for the shares.  However, such an approach is not adopted.  Rather the law distinguishes between income and capital by reference to “the source of the receipt”.    Distributions by a company out of profits, or at least those earned during the period in respect of which a distribution is made, “clearly fall”, as their Honours say, within the concept of income.  However, it has not proven possible to distinguish between shareholders in accordance with the relation to which a particular individual may stand to the particular profit distributed and  so no attention is paid to the fact that, as their Honours say, to one shareholder the distribution may be but a return of the amount the shareholder paid for the share. 

16                  In terms of company law and the law of income tax the distribution by a company of profits, be they profits accumulated or current year profits, is seen as the “detachment”, “release” or “liberation” of the profits.  The share itself remains in existence and is held by the shareholder.  So, as the Privy Council said in Hill v Permanent Trustee Co of New South Wales Ltd [1930] AC 720, a case concerned with the meaning of “income” for trust law purposes, it was said at 734 that:

“If payment to the shareholders is made out of profits it is income of the shares, and no statement of the company or its directors can change it from income into corpus.”

17                  It may be noted that the influence on Australian income tax law so far as it distinguishes between income and capital, rather than the analysis made by economists in distinguishing between the two concepts, has been the subject of criticism by some commentators and forms the basis for the drive towards the fresh approach to the calculation of taxable income proposed in what is presently referred to as “Option 2”: cf Krever “Simplicity and Complexity in Australian Income Tax” in Peterson & Gallagher (eds) Tax and Transfer Reform in Australia and Germany (Berliner Debatte Wissenschaftsverlag 2000) at 67-69.

18                  But, whether for trust law purposes or for the purposes of a tax based upon concepts of income and capital (absent a special provision such as s 47), where a company goes into liquidation and the mass of assets of the company, representing as it might accumulated profits or profits earned up to the date of liquidation is distributed to the shareholder on liquidation the rule adopted is different.  In such a case shareholders are returned the ultimate capital value of the intangible property constituted by their shares which no longer exist.  Profits are not detached, released or liberated leaving the share intact.  There is, as their Honours said in Stevenson at 99, “no dividend” and

“no distribution of profits because they are profits.  The shareholder simply receives his proper proportion of a total net fund without distinction in respect of the source of its components and he receives it in replacement for his shares.”

19                  What the shareholder receives is not income, it is capital.  And this is so, whether the context is trust law (see Hill at 729) or the taxation of income: see Inland Revenue Commissioners v George & Burrell [1924] 2 KB 52.

20                  These cases and the many cases which have since applied them explain the structure of ss 47 and 44 and the language which the legislature employed.

21                  First it must be said that the legislature determined to reverse, for the purposes of the Australian income tax law, decisions such as George & Burrell.  Thus s 47 was enacted to ensure that a distribution made by a liquidator (to the extent that the distribution had been made out of income) should be taxable to a shareholder who received it.  To achieve this it was not sufficient to deem the distribution to be a dividend, for that word could contemplate and was used in the Act to cover acts or transactions not necessarily of a revenue nature.  So, for example, s 6(1) of the Act now defines “dividend” as “any distribution made by a company to its shareholders”.  A distribution on capital account could thus be a dividend as defined.  It will be noted that the definition, of itself, would thus be incomplete to determine the character of the distribution to be either income or capital.  The source from which the distribution is to be made is not revealed.  The present definition of “dividend” may be compared with the definition of “dividend” first introduced in 1934 which qualified the definition by requiring that the distribution be out of profits.  The qualification was dropped in the 1936 Act, no doubt because s 44(1) of the Act, introduced at the same time, required a dividend to be paid out of profits before it was comprehended within s 44(1) and thus became, albeit depending upon the geographic source of profit, assessable income: cf Federal Commissioner of Taxation v Uther (1965) 112 CLR 630 at 636 per Kitto J.  The fact that his Honour dissented in the result does not affect the historical analysis which his Honour there sets out.  See too per Taylor J at 642 and per Menzies J at 643.  The judgment of Fullagar J in Federal Commissioner of Taxation v Blakely (1951) 82 CLR 388 at 406 repeats the history.

22                  Section 47 itself contained no territorial limitation.  Since the basal concept of the Act was to levy tax on residents of Australia deriving income wherever sourced, but to contain the impact of the tax on non-residents to Australian sourced income, s 44(1) not only explained what dividends were to be brought into assessable income, but also supplied a territorial limitation, consistent with s 25(1). Hence while taxpayers resident in Australia would pay tax on all dividends out of profits wherever sourced, non-resident taxpayers would only pay tax on dividends to the extent that the profits out of which the dividends were directly or indirectly paid were from an Australian source.

23                  The first High Court decision to address  some of the issues relevant to the present case was Blakely where shareholders obtained the assets of a company in the course of what may be described as an informal liquidation.   It was held that no amount was to be included in assessable income.  In the course of his judgment, Latham CJ at 397 accepted a submission of the Commissioner that the criterion of assessability under s 44(1) was not whether what was received by a shareholder was detached from the capital asset, namely the shares, but whether what the shareholder received came from the profits of the company.  His Honour was of the view that there had on the facts of the case been no distribution, although at least in part what was received by the shareholder represented profits derived by the company.  Fullagar J, with whom Dixon J agreed, was of the view that there had been no distribution of profits because they were profits.  Rather the shareholders had received a proportion of a net fund without distinction as to the source of the components: at 404.  Because all that happened was that the shareholders received but a proportion of the net fund without distinction as to the source of its components, there was nothing in the Act which gave the receipt the character of an income receipt, the character that it would otherwise have being capital.

24                  The second case (historically) dealt with by the High Court was Archer Bros Pty Ltd (in liq) v Federal Commissioner of Taxation (1953) 90 CLR 140.  The question which arose in that case was whether a private company which in the year of income went into liquidation should be treated as making a sufficient distribution for the purposes of the then Division 7 of the Act, which imposed a tax upon private companies which had not distributed at least as much as the Act prescribed to be a sufficient distribution.  It was held that a distribution to shareholders made by the liquidator, to the extent that that distribution was made out of income, was to be treated by force of s 47 of the Act as a dividend and that this was for the purposes of Division 7 as well as the other purposes of the Act.  Perhaps the only significance the case has in the present discussion is to illustrate some of the work which the words “for the purpose of the Act” have to perform. For this, it provides the foundation of the Commissioner’s third summary submission.  It is, however, important to note that in the joint judgment of Williams ACJ, Kitto and Taylor JJ their Honours refer to s 47 as assimilating for the purpose of the Act a distribution by a liquidator in the course of winding up for the purpose of the Act to dividends paid to shareholders by a company out of profits.  On the other hand this may be seen as no more than a paraphrase of the language of s 47(1).  However, their Honours did, at 155, emphasise that s 47(1) was not to be seen as “merely” ancillary to s 44(1).

25                  Next in time was Federal Commissioner of Taxation v W E Fuller Pty Ltd (1959) 101 CLR 403.  In that case a company made a bonus share issue to its shareholders.  The taxpayer, who had made a loss in the year of the issue, claimed to be entitled to carry forward that loss, notwithstanding the bonus issue.  It was held by majority (Fullagar and Menzies JJ, Dixon CJ dissenting) that the bonus shares were exempt income within s 80(1) of the then Act, the section governing the carry forward of losses, with the result that the amount of the loss carried forward was reduced by the amount of the value of the bonus share.  The majority reached this conclusion because s 6(1) defined dividend as including the paid-up value of shares distributed by a company to its shareholders.  Being a dividend, the bonus share issue was thus, it was held, income.

26                  The judgment of the majority need not now be explored.  It has subsequently been disapproved and the minority judgment of Dixon CJ has been said to be correct: cf Gibb v Federal Commissioner of Taxation (1966)118 CLR 628 at 632.

27                  It was the Chief Justice who held that, in the ordinary sense, a bonus share issue resulted in the shareholder receiving capital: Inland Revenue Commissioners v Blott [1920] 2 KB 657.  The fact that a bonus share was included in the definition of “dividend” for the purposes of the Act did not produce the result that the issue was income.  More was required where the share issue had the nature of capital.

28                  Parke Davis & Co v  Commissioner of Taxation (1959) 101 CLR 521 was decided in the same year.  In that case a subsidiary company of the taxpayer, incorporated in Colorado and a non-resident, was liquidated and its assets distributed on liquidation to its parent which was also not a resident of Australia.  The assets distributed included profits derived from sources in Australia as well as profits sourced outside Australia.    It was held that s 47(1) was to be treated as having an application independent of considerations of locality.  This is the Commissioner’s summary submission 2.  It was concerned only with a kind of transaction which comes within the ambit of s 44(1) relating to the assessability of dividends.  Section 44(1) provided the territorial criteria for including dividends in assessable income.  Section 44(1), so far as it concerned source was expegetical to s 23(r) which was the general provision providing the nexus between residence and source and thus confining the assessability of non-residents to income from an Australian source.  Accordingly, to the extent that there was an Australian source, s 44(1) applied to include an amount in the assessable income of the non-resident.

29                  Of s 47(1) their Honours said at 530-1 (and it is from this passage that the Commissioner’s submission 1 is derived):

“Section 47(1) is a provision dealing with a transaction according to its character.  Its purpose is obvious enough.  The section was first enacted in an earlier form to meet the situation, made clear enough by Inland Revenue Commissioners v Burrell which decided in effect that a distribution of a mass of assets, although in a colloquial sense they represented or contained profits, was a distribution of capital.  There is no ground, we think, for adding territorial restrictions to s 47.  That does not mean, of course, that in the description of the transactions s 47 gives, in reference to winding up and to a liquidator, and so on, grounds may not be found for saying that some particular procedure or process prescribed by a foreign law will be outside the terms of the section.  But it is, we think, of general application, quite independently of questions of locality.  It is when a transaction comes within the terms of s 47 that you go back to the provisions relating to locality if the case be one of a non-resident.  It is at that stage that you consider the source of imputed dividend.”

30                  An argument based upon lack of territorial connection with Australia likewise failed.

31                  Glenville Pastoral Co Pty Ltd (in liq) v Federal Commissioner of Taxation (1963) 109 CLR 199 is more to the point.  In that case the appellant company, which had received a liquidation distribution from its subsidiary, then went into liquidation and a distribution was made by its liquidator to its shareholders.  The question in issue was whether there had been a “sufficient distribution” by the appellant for the purposes of Division 7 of the Act as it then stood.  The appellant took the distribution it received from the liquidator of its subsidiary into its books as capital, as indeed it was (other than by force of s 47 with or without s 44(1)).  It was argued by the Commissioner that the only part of the distribution made by the appellant which qualified as a deemed dividend by the appellant for the purposes of s 47 was the amount by which the appellant’s funds exceeded the amount of its paid up capital which represented profit in the company law sense.  The argument was rejected.  Kitto, Taylor and Owen JJ said at 206-7:

“In no sense, apart from the provisions of this section, did any part of the distributions which the appellant received from the liquidator of Killens constitute income of the appellant.  But since, to the extent of £243,402, they represented income derived by Killens they were to that extent to be deemed, for the purposes of the Act, to be dividends paid to the appellant by Killens out of the profits derived by it.  As already mentioned, one result was that to that extent what the appellant received from Killens constituted by force of s 44(1)(a) the appellant’s assessable income for the year in question.  But the purposes of the Act are not limited to the purposes of s 44(1)(a);  they include the purposes of every provision of the Act:  see Archer Brothers Pty. Ltd. v. Federal Commissioner of Taxation.  Accordingly, in applying s 47(1) itself in relation to distributions by the appellant to its shareholders, it is necessary to treat the amount the appellant received from Killens as constituting, to the extent of £243,402, dividends paid out of Killens’ profits and therefore as being income derived by the appellant.  That being so, the whole amount of the distribution made by the appellant’s liquidator on 29^th October 1957, namely £241,669, must be deemed to have represented income derived by the appellant, and therefore to be deemed dividends paid by the appellant out of its profits except as regards any part of it which had been ‘properly applied to replace a loss of paid-up capital’.”

32                  One “significance”, at least, of Glenville for present purposes is that it demonstrates, if demonstration be necessary, that s 47(1) can have a multiple operation, at least where the companies in the chain of companies liquidated are all resident in Australia.  The real argument in the case was actually whether, since the appellant had lost its capital, a distribution which represented income had to be treated as excluded from the operation of s 47 on the ground that to the extent of the lost capital, it had necessarily been applied to make good that loss.  That argument, which has no resonance in the present case, was rejected.

33                  Next in terms of reported cases was Federal Commissioner of Taxation v Uther (1965) 112 CLR 630, a case dealing with a distribution to shareholders by way of reduction in capital.  In that case Taylor and Menzies JJ (Kitto J dissenting) found that the distribution was neither a distribution of profits or out of profits, but that what the shareholders received was capital.  The majority took the view that the decision in Blakely, to which reference has already been made, governed the case.  While an ordinary  dividend necessarily had to be paid out of profits, that was not the case with a reduction of capital.  As Taylor J said, the fact that the distribution came within the definition of “dividend” told nothing.  His Honour, however, distinguished s 47, where the dividend as defined was deemed to have been paid out of profits.  Likewise, Menzies J commented at 645:

“To deal with the same sort of problem as here confronts us when there is the liquidation of a company rather than a reduction of its capital, a special provision was found to be necessary, viz s 47”

34                  Gibb v Federal Commissioner of Taxation (1966) 118 CLR 628 again concerned the making of a bonus issue.  The bonus shares were sold and the company subsequently placed into voluntary liquidation.  The issue for decision was whether the proceeds of the bonus shares were to be treated as income.  Were Fuller to be accepted as correct, the bonus shares were income by force of the fact that they constituted a “dividend” under s 6(1) and the subsequent liquidation was to that extent thus made from “income”.  However the High Court by majority, Barwick CJ, McTiernan, Taylor and Windeyer JJ (Owen J dissenting), refused to follow the majority view in Fuller but rather approved the dissenting view of Sir Owen Dixon.  The bonus shares had, the Court said, the character of capital and were not, by force of s 6, made income.  All the definition of s 6 did was give a meaning to the word “dividend” as it was used in the Act.  Their Honours continued at p 635:

“Consequently, any distribution of the character mentioned in the definition is for the purposes of the Act a ‘dividend’ whether it constitutes an income or a capital receipt to the shareholders.  The line of reasoning employed to support the respondent’s contention does not, of course, suggest that the effect of the definition is to convert every distribution of the nature described in the definition into dividends in the ordinary sense of that term.  It asserts that, since dividends, in the ordinary and natural sense of that term, are income it follows that when the Act defines the term in a different and artificial sense – that is, to include distributions which are not dividends or income in the ordinary sense – it operates to invest dividends as so defined with the character of income.  In our view, and with respect to those who think otherwise, this line of reasoning is fallacious.  The function of a definition clause in a statute is merely to indicate that when particular words or expressions the subject of definition, are found in the substantive part of the statute under consideration, they are to be understood in the defined sense -  or are to be taken to include certain things which, but for the definition, they would not include…

We agree with Dixon CJ when he said:

‘the conception of ‘dividend’ does not affect the meaning or application of the word ‘income’; at all events so it appears to me.  The Act is not expressed to bring the defined conception of dividend within the word ‘income’.”

35                  In a passage which, while not directly relevant to the facts in issue in Gibb, has great persuasion since it represented the refutation by Barwick CJ, McTiernan and Taylor JJ of the view of the majority in Fuller, which their Honours did not follow, their Honours commented at 637:

“One further matter may be mentioned which, it seems to us, operates to confirm the views which we have expressed.  By force of s 47 of the Act, distributions to shareholders of a company by a liquidator in the course of winding up the company, to the extent to which they represent income derived by the company other than income which has been properly applied to replace a loss of paid-up capital, are deemed to be ‘dividends’ paid to the shareholders by the company out of profits derived by it.  But although in the language of the Act they may to this extent be properly described as ‘dividends’ they do not, by force of their character as such, further assume the character of assessable income, or, for that matter, of income.  This we think, is clear enough from the observations in Glenville Pastoral Co Pty Ltd v Commissioner of Taxation of the Commonwealth and Commissioner of Taxation of the Commonwealth v Uther.  The ‘statutory fiction’ (see Muller v Dalgety & Co Ltd) introduced by s 47 merely provides a basis for the operation of s 44 which is concerned exclusively with what dividends shall or shall not form part of a taxpayer’s assessable income.  It would, in our view, be anomalous to hold that distributions which constitute ‘dividends’ because they are comprehended by the definition of that term thereby, necessarily, achieve the character of income whilst distributions of a character which are not comprehended by the definition but which are deemed by s 47 to be ‘dividends’, do not by force of that provision achieve that character.”

36                  It is this passage which is relied upon by BIL in refutation of the Commissioner’s third proposition.  Indeed, it is submitted that this passage is part of the ratio of Gibb and provides a complete answer to the Commissioner’s submissions, it being conceded that the Commissioner’s argument must fail if the third summary submission is incorrect.  I shall return to this question later.

37                  Windeyer J expressed the view that the word “income” where appearing in s 47(1) of the Act included both that which was income by ordinary concepts as well as that which was assessable income.  However, the bonus dividend was neither income in ordinary concepts, nor was it made assessable income under s 44.  In consequence s 47(1) did not operate to include any amount in the assessable income of the taxpayer.  His Honour labelled as “fallacious” an argument that might be expressed in the following terms: any dividend paid out of profits has the character of income in ordinary concepts; therefore anything which is comprehended within the definition of “dividend” is given the character of income by the Act.  By this argument, the word “income” as used in the Act comprehends anything which is a “dividend” in the defined sense.  I should say in anticipation of the later discussion that it would not be fallacious, however, to say that because any dividend paid out of profits has the character of income in the ordinary sense, anything which the Act deems to be a dividend paid out of profits is given the character of income for the purposes of the Act so that when the Act uses the word “income” in s 47 it is capable of comprehending anything which the Act deems to be a dividend paid out of profits.

38                  Gibb was shortly afterwards followed by Harrowell v Commissioner of Taxation (1967) 116 CLR 607.  That case concerned successive distributions, the first the distribution to a parent company on the winding up of its subsidiary and the second the distribution to the parent company’s shareholders of its assets on liquidation.  The argument for the taxpayer was that what the parent received on liquidation of its subsidiary, while deemed by s 47 to be a dividend paid out of profits, was actually of a capital and not an income nature.  Accordingly the liquidation distribution by the liquidator of the parent was made out of capital, not out of income, and so did not enliven s 47.   In arriving at this conclusion the Court expressed the view that Gibb did not require a contrary view.  It had been submitted that the deeming of the distribution, to which s 47 referred, to be a dividend paid out of profits did not enable the Commissioner to treat any part of the amount of the distribution as income, just as the deeming of the bonus shares to be a dividend by force of s 6 did not have the effect of making the proceeds of the bonus shares income. The judgment is brief and significant to the present appeal.  That part of it (at 611-2) which addresses the submission made may usefully be repeated:

“The argument, however, is fallacious and the reasoning applied in Gibb’s Case has no application to the circumstances of this case.  Distributions by a liquidator to the shareholders of a company can be made only after the debts of the company have been paid or provided for and this must be borne in mind when we come to consider s 47(1).  Accordingly, when the sub-section expresses the initial condition for its operation – ‘Distributions … to the extent to which they represent income derived by the company’ – it proceeds on the basis that if a distribution is made to shareholders in the course of a winding up out of a fund which, either in whole or in part, represents income derived by the company, such distribution, or such part thereof, is to be regarded as a revenue profit and it is in that context that the distribution ‘is deemed to be dividends paid to the shareholders by the company out of profits derived by it’.  Within this framework there is, therefore, no room for the view that a deemed dividend under s 47 may in some circumstances consist of or include a capital profit and the reasoning in Gibb’s Case can have no application.  It seems to us that the concluding words of the sub-section were introduced to accommodate its provisions to the language of s 44.  But it is clear enough that when the legislature used these words it was speaking of a profit derived by a company on its revenue account and not otherwise.  Accordingly the sub-section, for the purposes of the Act, deems the distribution, to the extent to which the distribution is made out of income derived by the company, as dividends paid to the shareholders out of such profits.  The distribution by the liquidator of Killens, representing, as it did, income derive by Killens, the conclusion is inescapable that the effect of the sub-section was to invest the distribution with the character of a dividend paid to Glenville out of, and only out of, profits derived by Killens on revenue account and, therefore, income in Glenville’s hands.  That being so the appellant’s contention must be rejected ...” [original emphasis]

Conclusions to be drawn from the case law

39                  The following propositions are critical to the Commissioner’s case:

1.         That when s 47(1) deems a liquidation distribution to be a dividend paid out of profits, that deeming has the consequence that the distribution is to be treated as “income” for the purposes of the Act.

2.         That when the language of s 47(1) is satisfied, it has operation “for the purposes of the Act”, notwithstanding that the distribution to the shareholder is not made assessable income by the operation of s 44(1).

3.         That it follows that s 47(1) can operate successively through the liquidation of a chain of companies where some of the companies in the chain are non-resident and the source of the initial distribution is income, whether or not having a source in Australia.

40                  The first of the three propositions stated above is submitted by senior counsel for BIL to be contrary to the decisions of the High Court in cases such as Uther and, more particularly, Gibb.  Just as these cases stand for the proposition that merely because a distribution is deemed by s 6 of the Act to be a dividend it does not thereby have the character of income, so too the deeming in s 47 does not give the distribution the character of income.  It is only when s 47(1) operates together with s 44(1) to bring an amount into assessable income that the distribution is given a revenue character.  Particularly, it was submitted, the propositions set out above run directly counter to what the majority in Gibb said at 637 when propounding a further reason why the majority of the High Court in Fuller, should not be followed.

41                  The argument of BIL fails, in my view, to distinguish between, on the one hand, the way s 6 treats the distributions there referred to as being a dividend and no more, and, on the other, the deeming in s 47(1) of a liquidation distribution, to the extent it is made out of income, not only to be a dividend but a dividend which is paid out of profits.  To treat a distribution as a dividend gives that distribution no characteristic as capital or income.  Some distributions, for example of bonus shares, are clearly capital.  Others, such as the normal company law dividend, will have the character of income.  But that is not the case of a distribution which is made by a company out of profits.

42                  As I have sought to explain,  a company law dividend which of necessity is paid out of profits will be income in ordinary concepts.  The fact that it is so treated for trust law purposes without statutory deeming demonstrates this.  There are passages in the cases discussing s 47 in particular which say as much, for example the passage in the joint judgment of Kitto, Taylor and Owen JJ in Glenville which I have earlier set out.  What s 47(1) lacks as a taxing provision in a territorial system of taxation is any suggestion of source, for the basic structure of the Act is that, while resident taxpayers will pay tax on ordinary income from all sources, non-residents will only be assessed to tax on income which has an Australian source: see, for example, s 25(1) and s 23(r).  Without more, a question would arise whether the source of a dividend paid out of profits would need to be determined by reference to the location of the share or some other matter.  For that reason it is necessary that s 47(1) operate together with s 44(1) to determine whether any amount is to be included in assessable income.  Section 44(1) adopts, as the relevant territorial test, the source of the profits out of which the dividend is paid.

43                  It does not, however, follow from the fact that ordinarily s 47(1) will operate together with s 44(1) to include an amount in assessable income, that s 47(1) can not operate to deem a liquidation distribution to have the character of income.  Section 47(1) is expressed to apply  for the purposes of the Act and may do so, notwithstanding that s 44(1) may have no operation to include amounts in assessable income.  For example, as Archer Bros demonstrates, when Division 7 was in the Act, for the purposes of that Division, s 47(1) operated to determine whether a company had made a sufficient distribution and so was not to be assessed for tax under that Division.

44                  There remains, as the learned primary Judge observed, the question whether what was said in Gibb requires a contrary conclusion, particularly having regard to the decision in Harrowell.  It will be recalled that in Harrowell Taylor, Windeyer and Owen JJ expressed the view that the reasoning in Gibb had no application to the facts of Harrowell.  Likewise the decision in Gibb was concerned with facts and issues completely different from those raised by the present facts.  There the question in issue was whether the proceeds of the sale of bonus shares was income so as to make applicable s 47(1) to a liquidator’s distribution.  The case had nothing at all to do with the question whether the Act treats distributions literally falling within s 47(1) as being income so as then to enliven the operation of that section when a subsequent liquidation distribution is made from what represented the proceeds of an earlier s 47(1) distribution. 

45                  As the learned primary Judge herself observed, it is possible that not all that is said in Gibb is easy to reconcile with what was said in Harrowell, in spite of the fact that the members of the Court who participated in each decision were the same.  Clearly they did not think, when they decided Harrowell, that Gibb was wrongly decided or, for that matter, that what they said in Harrowell was inconsistent with what they had only shortly before said in Gibb.  It is, thus, necessary to consider with greater care what their Honours said in Gibb in the passage which I have earlier set out.  In so doing it is important to bear in mind that their Honours were addressing their remarks to the reasoning of the majority of the Court in Fuller with which they did not agree.

46                  It will be recalled that in Fuller Fullagar J had first expressed the view that the underlying company law principles on which issues of bonus shares proceeded dictated that the taxpayer in that case had received a share of capitalized profits which had been applied in payment for the bonus shares in fact issued.  This meant that, contrary to Inland Revenue Commissioner v Blott [1921] 2 AC 171, what was received was income in ordinary concepts and thus capable of being exempt income.   This was not a view which commended itself to Menzies J.  Both of their Honours were, however, of the view that the structure of the Act, by treating bonus share issues as dividends, gave the distribution effected by the bonus share issue the character of income.  This was so because dividends had the character of income in ordinary concepts.  What s 44(2) did was to remove such distributions from assessable income so as not to fall within s 44(1).  But the distribution remained income and as such had to be taken into account in determining the carry forward of losses.  As I have already noted, this was not the view which commended itself to Dixon CJ who dissented.

47                  Gibb, it will be recalled, was likewise concerned with bonus shares.  However, the issue was not the impact of the bonus share distribution on the computation of exempt income so as to determine whether there was a loss to be carried forward under the provisions of s 80 of the Act, but rather whether the proceeds of those shares were “income” for the purposes of s 47(1).  There was, as Barwick CJ, McTiernan and Taylor JJ in Gibb perceived, a parallel in that s 47(1) (like the definition of dividend in s 6(1)) also operated to deem a distribution of a particular kind to be a dividend.  So it was in this context that their Honours said, with respect correctly, that in their capacity just as dividends, liquidation distributions were not income.  That explains what their Honours meant when they said at 637:

“But although in the language of the Act they [ie the liquidation distributions] may to this extent be properly described as ‘dividends’ they do not, by force of their character as such, further assume the character of assessable income, or, for that matter, of income.” [emphasis added]

To that extent the passage in Gibb does not stand in the way of the Commissioner.  It is the balance of the passage after the reference to Glenville  and Uther to which senior counsel for BIL points.  It is clear that in this passage their Honours wished to convey and did convey the view that, for the purpose of determining whether a liquidation distribution was assessable income, s 47(1) operated in conjunction with s 44(1).    It did not operate on its own to make the distribution assessable income.  Such a distribution, as their Honours observed, did not become assessable income merely because the distribution was to be treated as a dividend.  So to say, is clearly, with respect, unexceptionable.  If the passage gives any assistance to BIL it is, in my opinion, only in the one sentence, viz:

“The ‘statutory fiction’ … introduced by s 47  merely provides a basis for the operation of s 44 which is concerned exclusively with what dividends shall or shall not form part of a taxpayer’s assessable income.” [emphasis added]

48                  It is a slender reed upon which to rely to give too much emphasis to the word “merely” as it appears in this sentence.  It can not be the case that their Honours intended to say that the only purpose of the Act to which the deeming of s 47 had application was s 44(1) for that would be inconsistent with Archer Bros.  Rather, I think, all their Honours intended to convey in that sentence and the balance of the paragraph in which it is contained was the fundamental point that s 47 did not operate on its own to make a distribution assessable income; that was the role of s 44(1).  That fundamental point is consonant with Parke Davis and Glenville.  It can also be arrived at by noting that section 47(1) has no territorial limitation at all.  That territorial limitation is to be found in s 44(1).  To use the language of Parke Davis and the Commissioner’s second submission, s 47(1) is independent of considerations of locality.  That locality is supplied by s 44(1) and for this purpose the deeming in s 47(1) provides the touchstone for the application of s 44(1).  This, however, says nothing about the deeming required by s 47(1) so far as it is sought to be applied to a series of liquidation distributions such as the present case raises.

49                  While it is true that s 47(1) can not operate by itself to make a distribution assessable income, it can operate “for the purposes of the Act” on its own as we have seen.  So, when Clarkson was liquidated, the resultant distribution was treated for the purposes of the Act as a dividend paid out of profits and thus income.  Thus when Rowsom was liquidated the resultant distribution was, by the application of s 47(1) to the Clarkson distribution, to be treated as income and in the result there was deemed to be a distribution to BIL, being a dividend paid out of profits to which the terms of s 44(1) then fell to be applied.

50                  It must be said that it would be strange indeed if there were any other result.  Let us take two examples.  In both examples there are three companies.  The top company, A, a resident of Australia, is wholly owned by the intermediate company, B, which is wholly owned by the bottom company, C.  C is a resident of Australia in each example.  In the one example B is a resident of Australia;  in the other B is not a resident of Australia.  A is liquidated and a distribution is made by its liquidator to its shareholder, B, out of profits.  In turn, B is liquidated and a liquidation distribution is made to C.  It is immaterial for the purposes of the example whether the profits of A have an Australian or an ex-Australian source.

51                  In the example where company B is a resident of Australia, Archer Bros has the consequence that the whole of the liquidation distribution of company B is included in the assessable income of company C by successive applications of s 47(1).  It is irrelevant that the original source of A’s profits is in Australia.

52                  In the other example, on the submission of BIL, company B receives a distribution which for the purposes of company law has the character of capital.  Section 47(1) is ignored for this purpose.  When company B is liquidated s 47(1) can have no application for the liquidation distribution by company B is made out of capital and not income.  In consequence no amount is treated as being included in the assessable income of company C, even if the whole of the profits of company A which are the original source of the ultimate distribution has an Australian source.

53                  An interpretation of the Act which would equate the two examples is to be preferred, if it is open as, with respect to the learned primary judge, it is, over one which would produce different results depending upon the residence of the intermediate company.

54                  Of course anomalies do arise in taxation.  So much may be conceded.  But when the language of the Act can be interpreted so as not to give rise to anomalies there is no reason to reject that interpretation.  It is not, with respect to the learned primary Judge, an anomaly which the language of the Act itself suggests.  Nor, when one examines, as I have, the history of the legislation and the cases decided on it, is it an anomaly which that history suggests.  To the contrary, once it is recalled that a dividend paid out of profits is treated for ordinary purposes as income in ordinary concepts, the language of s 47(1) becomes clear, as does its operation in the present case.

55                  I would, for these reasons, allow the appeal, set aside the judgment appealed from and, in lieu thereof, affirm the objection decision and dismiss the application with costs.  BIL shall pay the costs of the appeal.

Associate:

Dated:                          10 July 2000

ON APPEAL FROM A JUDGE OF THE FEDERAL COURT OF AUSTRALIA

REASONS FOR JUDGMENT

HEEREY J

56                  I agree with Hill J.

Associate

Dated:              10 July 2000

ON APPEAL FROM A JUDGE OF THE FEDERAL COURT OF AUSTRALIA

REASONS FOR JUDGMENT

SUNDBERG J

57                  For the reasons given by Hill J I would make the orders his Honour proposes.

Associate:

Dated:              10 July 2000