Commissioner of Taxation (Cth) v Commonwealth Aluminium Corporation Ltd

The Board of Review by majority reduced the assessments to nil. From this decision of the Board, the appellant appealed to the Supreme Court of Victoria. That Court (Jenkinson J.) dismissed the appellant's appeal in respect of the tax years 1969, 1970 and 1971: it upheld the appeal in respect of the tax year 1967 and returned the objection to the Board of Review for further hearing [8] .

The appellant then appealed to the Federal Court of Australia against the dismissals of his appeals and the respondent appealed to that Court against the upholding of the appeal in respect of the 1967 tax year. The Federal Court dismissed the appellant's appeals and allowed the respondent's appeal [1] .

1. (1979) 38 F.L.R 19; 9 A.T.R. 753; 79 A.T.C. 4203.

The appellant now appeals to this Court against the decision of the Federal Court and seeks the reinstatement of his assessments for the years 1967, 1969, 1970 and 1971.

The fate of the appeals to the Federal Court turned entirely on the meaning and application of s. 136, which it is convenient immediately to set out in full:

(b) is carried on by a company a majority of the shares in which is held by or on behalf of non-residents; or

(c) is carried on by a company which holds or on behalf of which other persons hold a majority of the shares in a non-resident company,

and it appears to the Commissioner that the business produces either no taxable income or less than the amount of taxable income which might be expected to arise from that business, the person carrying on the business in Australia shall, notwithstanding any other provision of this Act, be liable to pay income tax on a taxable income of such amount of the total receipts (whether cash or credit) of the business as the Commissioner determines.

It is conceded by the appellant that the circumstances of the tax years in question cannot be brought within the terms of pars (b) and (c) of s. 136. The appellant relies exclusively on the terms of par. (a) to justify his determination of a taxable income on which he has assessed the respondent to income tax.

Subsequent to the incorporation of Comalco, Kaiser Aluminium and Chemical Corporation ("Kaiser"), a company incorporated in America and a non-resident for the purposes of the Act, acquired 50 per cent later reduced to 45 per cent, of the issued shares in the capital of Comalco, shares which were still held by Kaiser in the relevant tax years.

At all material times the Riotinto Zinc Corporation Ltd. ("Riotinto"), a public company incorporated in the United Kingdom, by reason of its shareholding in a chain of companies, three in number, could control the manner in which 45 per cent of the shares were voted at a general meeting of Comalco. The remaining 10 per cent of the shares in Comalco were held by the public.

The identity of the shareholders of Riotinto and of Kaiser is not known. In particular, it is not known whether any of them or, if so, what proportion of them were residents of Australia for the purposes of the Act. However, treating the chain of shareholding as ending with the two companies, Riotinto and Kaiser, rather than with their shareholders, it can be said that during the tax years in question, it would have been possible for Riotinto and Kaiser, both non-residents, acting in conjunction effectively through Comalco to have controlled a general meeting of the shareholders of the taxpayer.

But in none of the tax years in question did either Riotinto or Kaiser give any directions to or attempt to effect any control of the taxpayer. Nor did either of those companies in any wise interfere with the control or management by the taxpayer of its business of mining and vending bauxite.

A feature of the manner of disposal of some part of the bauxite product of the taxpayer's mining was given prominence by the appellant. It appears that the respondent, in negotiating with a Japanese buyer of bauxite, agreed with that buyer to form with that buyer a company in Hong Kong to which bauxite would be sold by the respondent, part of which would be on-sold to that Japanese buyer and the remainder to other buyers at a profit to the Hong Kong company. It does not appear whether the Japanese buyer would have been prepared to buy direct from the respondent at the price it paid the Hong Kong company without participating in the profit to be made by that company from its purchase and sale of bauxite. It was said for the respondent that the arrangement for the intervention of the Hong Kong company was to keep from general knowledge the fact that the actual price paid by the Japanese company was less than the "going rate".

But, however that might be and whatever the motivation for this manner of disposal of the bauxite, the decision so to handle the transaction with the Japanese buyer was made entirely by the respondent without any participation in that decision by either of the non-resident companies.

The competing submissions of the parties before this Court were, on the part of the appellant, that, having regard to the shareholding of the taxpayer which Riotinto and Kaiser could influence, it should be inferred that in fact during the tax years in question those two companies did exercise control of the taxpayer and therefore of its business. It was said that in this fashion par. (a) of s. 136 was satisfied and, apart from discussion of their quantum, the assessments supportable.

The respondent's submission was simply that it did control its own business throughout the whole of the tax years in question and that it could not be said that, by reason of the shareholding of Riotinto or Kaiser, the business of the taxpayer was in fact controlled in any of the tax years in question by a non-resident. Assuming a capacity in the non-residents to exercise control, it had not in fact been exercised.

The Federal Court of Australia ccepted this submission of the respondent. I may say at once that, for my part, I have no difficulty whatsoever in accepting that view myself. There is no validity, in my opinion, in the proposition that, because, due to the shareholding, the non-resident companies had the capacity if they chose to exercise it to control a general meeting of the taxpayer, therefore they did in fact control its business in the tax years in question. Further, it is important to distinguish between control of the business and the control of a general meeting of shareholders. Whilst through the latter the business might possibly be controlled, it does not in the least follow that the control of a general meeting necessarily involved or resulted in fact in the control of the company's business.

Further, we know for a fact that neither of the non-residents exercised in any of the years in question whatever capacity they might have had to control the corporate life of the taxpayer. Not having done so, it could scarce be said that the taxpayer's business was controlled in any of these years by the non-residents. The appellant's submission to the contrary is insupportable and ought not to be accepted.

The Supreme Court of Victoria, in determining the appeals brought to it from the decisions of the Board of Review, took the view that the persons who controlled the business of the taxpayer in the tax years in question were the members of the board of directors of the taxpayer. But, even so, except in the tax year 1967, the majority of the members of the board of the taxpayer were residents of Australia. That being so, the Supreme Court of Victoria, even though it took the view that the control of the business was in the hands of the board of directors, found no basis for concluding that in the tax years other than the year 1967 the business of the taxpayer was controlled by non-residents.

In relation to the year 1967, however, the majority of the board were non-residents but no decisions of the board in relation to the business of the company were made by a board the majority of whose members attending the meeting of directors were non-residents. The Supreme Court thought, however, that, because the non-residents who did not attend the board were kept apprised of the decisions of the board and did not attempt to reverse any of those decisions, they must be treated as participating in the decisions, so making these decisions decisions of a board of directors of whom the majority were non-residents though in fact the resolutions were passed by resident directors only. I am unable to agree with this approach of the Supreme Court. The proper conclusion to be drawn from the facts to which I have referred is that the non-resident directors were content to leave the control of the business in the hands of the resident directors. It is not, in my opinion, correct to say that their silence or lack of endeavour to change the decisions of the resident directors warrant the conclusion that they had become participants in those resolutions. It is an approach which, in my opinion, is not only artificial in point of fact but unwarranted in point of law. Thus, in my opinion, even if the view of the Supreme Court of Victoria that the relevant control of the business of the taxpayer was in the hands of the board of directors were correct, the facts did not warrant the allowance of the appeal by the appellant in respect of the tax year 1967.

But, in my opinion, the Supreme Court was in error in concluding that the directors, viewed as it were apart from the company, were in control of the business of the company. It is quite true that they were invested by the articles of the taxpayer with the power to manage and control the business of the company. But they did so for the company, not as it were for themselves as a separate and distinct entity from the company itself. A consequence of the view of the Supreme Court, which it seems to me must have been overlooked, is worthy of mention. For, if the Supreme Court's view were right, a business owned by a non-resident corporation which had a majority of resident directors would not be controlled by a non-resident within the meaning of par. (a) of s. 136. I cannot myself think that such a situation would be consonant with either the text or the policy of the section.

The appellant, in a last minute endeavour during his reply to the respondent's argument, sought to rely upon the view of the Supreme Court of Victoria but with the gloss that where there had been a majority of non-resident directors for any part of a year the Supreme Court ought to have treated the matter, as the appellant's junior counsel said, "distributively" and supported an assessment for part of that year. He treated the Commissioner as being able to make an assessment for part of the year though unable, because of the state of the directorate in the balance of the year, to do so either for the year as a whole or for the remaining part of the year. In my opinion, this suggestion lacks substance. Analogy was sought to be drawn between the case of the taxation of a person who is a resident for part of a year and a non-resident for the other part of a year, but in truth there is no such analogy. What the Commissioner is able to do under s. 136, if the necessary condition of control is fulfilled, is to take the total receipts of the taxpayer for the year and apply to that total some percentage in order to arrive at a taxable income which is an annual income. But, in any case, as I have indicated, in my opinion, the view of the Supreme Court as to the board of directors being relevantly in control of the taxpayer's business ought not to be sustained.

A question mentioned in the course of argument was whether par. (a) was applicable at all in the case of a corporate owner of an Australian business. It is quite true that par. (a) must be read in the light of the presence of pars. (b) and (c), although in what I have already written I have not treated either par. (b) or (c) as having any restrictive effect on par. (a).

Paragraphs (b) and (c) provide the extent to which the Parliament has seen fit to treat the shareholding of a taxpayer as a ground for the exercise by the Commissioner of the powers given by s. 136. But there is an evident distinction between par. (a) which deals with the actual control of the business of the taxpayer and pars. (b) and (c) which deal with the capacity to control the corporation through shareholding. Paragraph (a) takes a radically different criterion for the availability of the Commissioner's powers under the section to those chosen in the other two paragraphs.

I can see no reason why par. (a) should not extend to the case of a business corporately owned. The assumption of that paragraph is that the business may be carried on by an owner but, none the less, be controlled by others. In my opinion, par. (a) is applicable to the case of a corporate owner carrying on a business in Australia if the control of that business is during the tax year in the hands of non-residents. Here, however, as I have said, the only control in fact of the taxpayer's business was by the taxpayer itself.

There is necessarily some influence in the presence of pars. (b) and (c) on the submission of the appellant, though I have been able to deal with it without reflecting any such influence. It would be strange indeed, in my opinion, if par. (a) should allow of an exercise of the Commissioner's power in a case where the capacity to control through shareholding was not limited to the instances given by pars. (b) and (c). As I have said, these paragraphs seem to me to exhaust the extent to which the shareholding in the taxpayer's capital should be treated as the decisive circumstance attracting the Commissioner's power under the section. But I find it unnecessary for the decision of this appeal to decide whether or not the presence of pars. (b) and (c) are fatal to the appellant's attempt to trace the control through shareholding beyond the limits set for these purposes by those paragraphs.

I would therefore agree with the conclusion and reasoning of the Federal Court and dismiss these appeals.

These four appeals relate to assessments made by the appellant Commissioner against the respondent taxpayer under s. 136 of the Income Tax Assessment Act 1936, as amended ("the Act"). The taxpayer objected to assessments for the years 1967 to 1971 inclusive. The taxpayer appealed to the Board of Review against the assessments for the five years except the 1968 year. In relation to the 1968 year the taxpayer appealed to the Supreme Court of Victoria.

The Board of Review allowed the taxpayer's appeals by majority, reducing each asessment to nil. The Commissioner thereupon appealed to the Supreme Court of Victoria which dismissed the appeals in respect of the years 1969 to 1971 inclusive and allowed the appeal in relation to the 1967 year. The appeal in relation to the 1968 year was adjourned [2] . The taxpayer appealed to the Federal Court in respect of the 1967 year and the Commissioner appealed in relation to the 1969 to 1971 years inclusive. The Federal Court dismissed the Commissioner's appeal and allowed the taxpayer's appeal [3] .

1. (1978) 8 A.T.R. 816; 78 A.T.C. 4371. 2. (1979) 38 F.L.R. 19; 9 A.T.R. 753; 79 A.T.C. 4203.

(b) is carried on by a company a majority of the shares in which is held by or on behalf of non-residents; or

(c) is carried on by a company which holds or on behalf of which other persons hold a majority of the shares in a non-resident company,

and it appears to the Commissioner that the business produces either no taxable income or less than the amount of taxable income which might be expected to arise from that business, the person carrying on the business in Australia shall, notwithstanding any other provision of this Act, be liable to pay income tax on a taxable income of such amount of the total receipts (whether cash or credit) of the business as the Commissioner determines.

Throughout, the Commissioner's case has been based on par. (a) of s. 136. It is common ground that pars. (b) and (c) have no application to the facts of this case. The issue, then, is whether the taxpayer's business, that of mining and selling bauxite, a business carried on in Australia, was "controlled principally by non-residents", during the years in question.

In the Supreme Court of Victoria, Jenkinson J. held that, in light of the history of s. 136 (a), the paragraph applied to a business carried on by a company and that the presence of pars. (b) and (c) did not warrant the conclusion that par. (a) was restricted to non-corporate businesses. His Honour considered that art. 62 of the taxpayer's articles of association, which vested in the board of directors the power of management of the company's business, led to the conclusion that the business carried on by the taxpayer was controlled by its directors. He held that in the year ended 31st December 1967 the majority of the taxpayer's directors were non-residents and that the business could be said to have been controlled principally by those non-residents. However, on considering the number of meetings held in the 1969 year and the predominant influence of the resident directors, his Honour came to a different conclusion in relation to this year. Again, in the 1970 and 1971 years he found that the business was controlled principally by resident directors.

A different view prevailed in the Federal Court where Deane J. (with whom Brennan and Fisher JJ. agreed) concluded that the business of mining and selling bauxite which was carried on in Australia during the relevant tax years by the taxpayer was, for the purposes of s. 136 (a), controlled principally by the taxpayer which was not a non-resident. It followed that s. 136 had no application to the taxpayer in any of the tax years.

An outline of the basic facts is an essential preliminary to an understanding of the case which is presented by the Commissioner and of the questions which arise for determination. The share capital in the taxpayer was throughout the relevant period wholly owned by Comalco Ltd. ("Comalco"), a company incorporated in Australia. The share capital in Comalco was owned in the years 1965 to 1969 as to 50 per cent by Kaiser Aluminium and Chemical Corporation ("Kaiser"), a company incorporated in the United States of America, and as to 50 per cent by Conzinc Riotinto of Australia Ltd. ("C.R.A."), a company incorporated in Australia. In the years 1970 to 1971 the interest of Kaiser and C.R.A. in the share capital of Comalco was reduced to 45 per cent each, the public in Australia and New Zealand holding the balance of the share capital.

In the years 1965 to 1971 the share capital in C.R.A. was held as to 80 per cent to 85 per cent by C.R.A. Holdings Pty. Ltd., an Australian company, with other Australian residents holding most of the balance of the shares. The share capital in C.R.A. Holdings Pty. Ltd. was held as to 99.9 per cent by R.T.Z. Australia Holdings Ltd., a company incorporated in the United Kingdom, which in turn was wholly owned by the Riotinto Zinc Corporation Ltd. ("Riotinto"), a company incorporated in the United Kingdom. Neither Kaiser nor R.T.Z. Australia Holdings Ltd. nor Riotinto answers the statutory definition of resident in s. 6 of the Act.

The Commissioner's primary argument is that in the relevant period the capacity to control the taxpayer and its business activities lay with the overseas companies Kaiser and Riotinto, and that the inference to be drawn is that this capacity was exercised. For the drawing of this inference the Commissioner relies on two additional matters to which we shall refer later. With the exception of a direction given by Comalco to the taxpayer in relation to transactions involving the setting up of a Hong Kong company, Comalco Bauxite (H.K.) Ltd. ("the Sales Company"), the Commissioner concedes that the evidence does not establish outside interference by Kaiser or the other overseas companies or their representatives in the management and control of the taxpayer's business activities. However, he submits that Riotinto at all times had the capacity to control general meetings of the chain of companies which intervened between Riotinto and Comalco and that therefore Kaiser and Riotinto together had the capacity to control general meetings of Comalco and the taxpayer and consequently the business of the taxpayer. It is then submitted that the inference to be drawn is that the business of the taxpayer was conducted in accordance with the policies and wishes of the two overseas corporations.

Section 136 is a re-enactment of s. 28 of the Income Tax Assessment Act 1922 with the addition of the present pars. (b) and (c). In turn, s. 28 was based on s. 23 of the Income Tax Assessment Act 1915-1921 which had its origin in and was in the same terms as s. 31 of the Finance (No. 2) Act, 1915 U.K., later rr. 7-12 of the General Rules of the Income Tax Act, 1918 U.K. - see Case No. N. 69 [4] . The first point to be gleaned from this legislative history is that par. (a) is the paramount provision of s. 136; pars. (b) and (c) are designed to amplify par. (a) and extend its reach to the situations which they describe. The legislative history in our view negates the suggestion that the content of par. (a) should be restricted by reference to pars. (b) and (c).

The second point to be made is that "controlled principally" means "controlled chiefly or in the main". It does not mean "controlled as a principal, not as an agent".

The third point to be made, one which emerges from the language of the section itself, is that par. (a) extends to a business carried on by a company. This is the effect which should be given to the words of the paragraph read according to their ordinary meaning. This view is reinforced by the legislative history of the section mentioned earlier. Sub-section (1) of s. 28 applied to businesses carried on by companies: this was made quite clear by the terms of sub-s. (2). One would require compelling reasons, which are here absent, before concluding that, when substantially re-enacted in 1936, what is now par. (a) of s. 136 did not have the like operation. Further, as pars. (b) and (c) are directed to particular and special situations, the fact that they refer to a business "carried on by a company" is not in itself a sufficient ground for confining the very general language in which the opening words of the section and par. (a) are expressed.

Moreover, the word "controlled", when used passively, in its ordinary meaning refers to de facto control rather than to capacity to control, a concept which is reflected in pars. (b) and (c). There is nothing in s. 136 to suggest that "controlled" bears a special meaning in par. (a). What is more, the notion of de facto control is appropriate when we consider that it is to the business carried on by a company, not to the company itself, that the word relates. In the absence of evidence of acts of control on the part of outsiders, the question then arises as to who it was who "controlled" in the sense described the business of the taxpayer in the years in question. Article 62 of the taxpayer's articles of association, in accordance with usual practice, vests the management of the company's business in the board of directors and there was no delegation of the board's powers to a managing director or any other person. Although control and management are not synonymous terms, any distinction that may be made between them does not lead to the conclusion that an article such as art. 62 constitutes the directors as mere managers and leaves the shareholders in de facto control of the business of the company.

It is idle to suggest that it is the shareholders who ordinarily control the business and the business activities of a company. Their participation is generally limited to the receipt of annual accounts and reports of directors, to the approval of these accounts and reports and to the election of directors at the annual general meeting. Important decisions, whether involving questions of policy or not, are invariably taken by the directors who are ultimately responsible to the company in general meeting for the conduct of the company's business operations. The shareholders, through their power to control the company in general meeting and perhaps through their power to elect directors, may be said to "control" the company, but as a general rule they do not exercise de facto control of the company's business. Of course, there will be cases in which the evidence establishes that some shareholders or outsiders do exercise de facto control of the business and possibly there may be cases in which shareholders exercise control of the business at general meetings. This is not such a case.

It may be noted that the distinction between management and control on the one hand and the voting power of shareholders on the other hand is drawn in the definition of resident in s. 6.

In the context of s. 136 little assistance is to be derived from the many cases which examine such concepts as control of a company and "controlling interest" in a company, viz. Mendes v. Commissioner of Probate Duties (Vict.) [5] ; Kolotex Hosiery (Australia) Pty. Ltd. v. Federal Commissioner of Taxation [6] ; British American Tobacco Company Ltd. v. Inland Revenue Commissioners [7] and Barclays Bank Ltd. v. Inland Revenue Commissioners [8] . They reflect in varying degrees the different notion of capacity to control which, on the view which we have expressed, is not an element in par. (a).

1. (1967) 122 C.L.R. 152. 2. (1975) 132 C.L.R. 535. 3. [1943] A.C. 335. 4. [1961] A.C. 509.

It has been suggested that control of a company may be inferred from the existence of a power to control its business (British American Tobacco Company Ltd. v. Inland Revenue Commissioners [9] , but this is another question. As Kitto J. observed in Mendes [10] , when "a statute makes the fact that "the company" was controlled by a deceased person a critical fact in respect of duty upon the estate of the deceased the prima facie meaning must be that the company in general meeting was controlled by the deceased in the only way in which it could be so controlled", namely, by the exercise of majority power at such a meeting. Section 136 (a) is directed not to the control of the company, that is, of the company in general meeting, but to the control of the business carried on by the company with respect to which the decisions are made by the directors " for the company, not by it", to use the words of Kitto J. (Mendes [11] : see also per Windeyer J. [12] ).

1. [1943] A.C., at p. 339. 2. (1967) 122 C.L.R., at p. 161. 3. (1967) 122 C.L.R., at p. 160. 4. (1967) 122 C.L.R., at pp. 169-170.

That the directors make their decisions as fiduciary agents of the company is not to the point. In Gramophone and Typewriter Ltd. v. Stanley [13] Fletcher Moulton L.J., speaking of the control of a legal corporator, said:

Such control is inseparable from his position as a corporator and is a wholly different thing both in fact and in law from carrying on the business himself. The directors and employees of the corporation are not his agents, and he has no power of giving directions to them which they must obey. It has been decideed by this Court, in the case of Automatic Self-Cleansing Filter Syndicate Co., Ld. v. Cunninghame [14] , that in an English company, by whose articles of association certain powers are placed in the hands of the directors, shareholders cannot interfere with the exercise of those powers by the directors, even by a majority at a general meeting. Their course is to obtain the requisite majority to remove the directors and put persons in their place who agree to their policy. This shows that the control of individual corporators is something wholly different from the management of the business itself. See generally K. A. Aickin, Division of Power Between Directors and General Meeting as a Matter of Law, and as a Matter of Fact and Policy, Melbourne University Law Review, vol. 5 (1967), p. 448. 1. [1908] 2 K.B. 89, at p. 98. 2. [1906] 2 Ch. 34.

It is not an acceptable reason for saying, in the context of s. 136 (a), that ordinarily it is the company itself, not its directors, who actually control the business. The company is the proprietor of the business which it carries on through its officers. But s. 136 (a) in speaking of de facto control is looking, not to the proprietor, whether he be an individual or a company, who carries on the business. Section 136 (a) is aimed, though not exclusively, at a situation in which the business is controlled by a non-resident, not being the person who carries on the business, with the consequence that the person who carries on the business becomes liable to pay income tax under the section. This construction of the section gives effect to its evident purpose - that is, to give the Commissioner a discretion to determine the taxable income of a business when it is actually controlled by non-residents' being persons whose decisions with respect to the business may result in the incurring of inflated liabilities and outgoings to the advantage of overseas interests.

It may be, as was suggested in argument, that a company resident in New York whose directors are all residents in Australia would not fall within the section. Whether it does so or not depends on a variety of factors. There may be evidence of control by non-residents who are not directors (par. (a)); there may be evidence of shareholding to bring it within par. (b) or of facts establishing par. (c). We do not regard these contingencies as a sufficient reason for discarding the construction which we favour.

In support of the Commissioner's primary argument, Mr. Liddell points out that in Texas Co. (Australasia) Ltd. v. Federal Commissioner of Taxation [15] , it was admitted that the business carried on by the taxpayer was at all material times controlled by persons resident outside Australia. But we have no knowledge of the primary facts or of the available evidence which induced the making of the admission. Certainly no issue arose on the point for the Court to decide and nothing of present significance appears in the judgments. The case therefore provides no support for the Commissioner.

The judgment of Deane J. in the Federal Court was directed to rejecting the far-reaching interpretation which the Commissioner endeavoured to place upon s. 136 (a). We agree with his Honour in thinking that it is an interpretation which cannot be supported, though we do not agree with the interpretation which his Honour expounded. The construction placed upon s. 136 (a) by Jenkinson J., which generally accords with the view which we have expressed, was not dealt with by the Federal Court, because it was not argued by Mr. Liddell. On the case which he presented, the Federal Court thought that the critical issue was whether or not the paragraph required investigation beyond the taxpayer's control of its business. It was evidently assumed that if the paragraph did not require investigation beyond the taxpayer's control of its business, that would be the end of the matter. The Federal Court considered this as a preliminary question and, having decided it against the Commissioner, it decided the appeals in favour of the taxpayer.

It is not surprising that the Federal Court took this course because even in this Court an argument in support of Jenkinson J.'s judgment was presented, not in chief, but in reply by Dr. Spry, an unusual course to say the least of it.

The Commissioner drew attention to evidence relating to the formation of the Sales Company in Hong Kong in which the share capital was owned as to 52 per cent by Comalco and as to 48 per cent by two Japanese companies, Showa Denko and Sumitomo Chemicals, in equal shares. The company was formed as a sales company after negotiations had taken place for the sale of bauxite by the taxpayer to the two Japanese companies. The taxpayer claimed that the Japanese companies were willing to purchase large tonnages on the footing that they were given a discount on current prices. There was, however, a risk that they might enter into contracts with Alcan, a Canadian company which had a right to draw on the taxpayer's production of bauxite and which threatened to sell at a very substantial discount. The taxpayer claimed that it was unwilling to allow this development to occur because it would prejudice its future Pacific trade. As a result the parties, including Comalco, agreed on a proposal whereby the Sales Company was incorporated in Hong Kong, long-term contracts were entered into in 1965 for very substantial tonnages of bauxite under which the bauxite was sold by the taxpayer to the Sales Company at a discount and the Sales Company sold the bauxite on to the Japanese companies at current prices. In fact the bauxite was not delivered by the taxpayer to the Sales Company but was delivered direct to the Japanese buyers. The profit made by the Sales Company was then distributed by way of dividend to its three shareholders.

Comalco played a part in negotiating these developments and there is no doubt that the taxpayer acted on directions given to it by Comalco in connexion with these transactions. As Jenkinson J. found, the decision that the majority of the shares in the Sales Company would be held by Comalco and not by the taxpayer was forced upon the taxpayer's directors in consequence of Comalco's general policy that its subsidiaries would not have subsidiaries and that the share capital in all subsidiaries in the group of companies controlled by Comalco would be held by Comalco. Jenkinson J. found that the decision of the taxpayer's directors that the taxpayer would sell bauxite to the Sales Company was substantially influenced by a direction by Comalco. His Honour also found that decisions by Comalco concerning the advance of funds to the taxpayer also influenced decisions made with respect to the taxpayer's business by its directors. However, on the whole of the evidence Jenkinson J. found that the business was controlled by the taxpayer's directors. With these findings we agree and we do not think that the evidence relating to the Sales Company and the transactions entered into with it justify the general inference that the taxpayer's business was controlled by the two overseas companies behind Comalco.

The other matter on which the Commissioner relied to support the inference he sought to draw was that Kaiser and Riotinto combined to incorporate the taxpayer and to set it up in business. On the assumption that this fact is established by the evidence, it is not a fact which taken in conjunction with other facts proved, namely, the shareholding structure of the taxpayer and the inferences yielded by the transactions relating to the Sales Company, warrants the drawing of the suggested inference that the two companies actually controlled the taxpayer's business in the relevant periods.

There remains for consideration the question whether the control, which was exercised in fact by the directors, amounted to control by non-residents. We have not been able to perceive any basis for disagreeing with the conclusion reached by Jenkinson J. for the years 1969 to 1971 that, as the majority of the directors then holding office were residents of Australia, par. (a) has no application.

However, we cannot agree with his Honour's conclusion in relation to the 1967 year. In that year, generally speaking, the decisions of the directors were taken by a majority of directors resident in Australia. Moreover, for the most part the more important decisions with respect to the taxpayer's business were taken at meetings attended only by resident directors and the proposals resulting in such decisions were "conceived and formulated", to use the words of Jenkinson J., by residents. Nevertheless his Honour inferred that directors resident overseas were kept informed of decisions taken at directors' meetings and that they did not subsequently dissent from them. However, one should only conclude that the silence of directors resident overseas amounts to agreement with and participation in the decisions of the board if there is evidence of those non-resident directors being informed in detail of the meetings and at times having exhibited from abroad some overt participation. There was no such evidence here. For this reason we consider that it was incorrect to count them in and thereby conclude that the business was controlled by a majority of non-residents. In truth the position was that in the 1967 year the business was controlled by the directors who attended board meetings and were in Australia of whom a majority were residents.

These four appeals by the Commissioner of Taxation against the Federal Court's decision in the taxpayer's favour concern the interpretation and application of s. 136 (a) of the Income Tax Assessment Act 1936, as amended, and the question is whether the taxpayer's business carried on in Australia is controlled principally by non-residents. Section 136 provides:

(b) is carried on by a company a majority of the shares in which is held by or on behalf of non-residents; or

(c) is carried on by a company which holds or on behalf of which other persons hold a majority of the shares in a non-resident company,

and it appears to the Commissioner that the business produces either no taxable income or less than the amount of taxable income which might be expected to arise from that business, the person carrying on the business in Australia shall, notwithstanding any other provision of this Act, be liable to pay income tax on a taxable income of such amount of the total receipts (whether cash or credit) of the business as the Commissioner determines. Section 136 (a) extends to corporate business and is not qualified by sub-sections (b) or (c). It appeared to the Commissioner that, in the tax years in question (1967, 1969, 1970 and 1971), the taxpayer's business produced less than the taxable income which might be expected to arise from the business. The presumption is that the Commissioner's assessment is correct, and the taxpayer has the burden of showing that it is excessive (see s. 190 (b) of the Act; McCormack v. Federal Commissioner of Taxation [16] ;Macmine Pty. Ltd. v. Federal Commissioner of Taxation [17] . The taxpayer sought to do so by showing that its only business was the mining and sale of bauxite in accordance with its objects and the Commonwealth Aluminium Corporation Agreement Act, 1957 Q.; it acquired leases from the Queensland Government at Weipa and its mining activities were in accordance with an agreement made with that Government and dealt with in that Act. 1. (1979) 143 C.L.R. 284. 2. (1979) 53 A.L.J.R. 362; 24 A.L.R. 217; 9 A.T.R. 638; 79 A.T.C. 4133.

The taxpayer claims that s. 136 (a) was inapplicable, as the company was itself resident in Australia and carrying on the business. The Federal Court accepted this. In my opinion, they were wrong to do so. Section 6 of the Act provides that, in the absence of a contrary intention, the word "resident" or the phrase "resident of Australia" when used in the Act means, in the case of a company, a company which is incorporated or carries on business in Australia and has either its management and control in Australia or its voting power controlled by shareholders who are residents of Australia. The phrase "non-resident" means "a person who is not a resident of Australia". Section 136 was intended to be an effective instrument for the Commissioner to deal with non-residents controlling businesses in Australia in such a way that they were able to reduce taxable income by shifting available profits elsewhere or by other devices. The section would be virtually useless if it could not be applied whenever the immediate corporate owner was incorporated in Australia. It would, for example, make nonsense of s. 136 (a) to read it so that a resident company, all of whose directors and shareholders were non-resident, was outside its scope.

The taxpayer also claimed that its directors, officers and employees conducted the affairs of the company themselves, and that no outside interference had been proved. The Commissioner contended that the ownership of the company was ultimately predominantly in the hands of non-residents who had the capacity to control in the relevant years and that it should be inferred that they in fact controlled the company and therefore the business. The history of the company shows that it was established by The Riotinto Zinc Corporation Ltd. ("Riotinto") and Kaiser Aluminium and Chemical Corporation ("Kaiser") and is part of a well-known transnational group. The taxpayer was owned totally by Comalco Ltd. ("Comalco"), 50 per cent of which was owned by Conzinc Riotinto of Australia Ltd (C.R.A.) (which is resident in Australia) and 50 per cent by Kaiser (which is resident in the U.S.A.) in 1965-1969. In 1970-1971, C.R.A. and Kaiser's holdings were reduced to 45 per cent each, with the public in Australia and New Zealand holding the other 10 per cent. C.R.A. was itself owned during 1965-1971 (85-80 per cent) by C.R.A. Holdings Pty. Ltd. (resident in Australia), 99.9 per cent of which was owned by R.T.Z. Australia Holdings Ltd. (resident in the United Kingdom), which in turn was totally owned by Riotinto (resident in the United Kingdom). The rest of C.R.A. was owned by Australian residents (other than C.R.A. Holdings Pty. Ltd.) who held 15-18 per cent and non-residents who held a maximum of 1 per cent.

The history of the aluminium industry is short and well-documented (see Consideration of International Measures on Bauxite (1978), Report by the United Nations Conference on Trade and Development; G. J. Crough, Foreign Ownership and Control of the Australian Mineral Industry, Transnational Corporations Research Project, Occasional Paper No. 2, University of Sydney (1978); C. M. Varsavsky, Industry Studies of Non-Ferrous Industries: Aluminium, Institute for Economic Analysis, New York University; E. R. Richardson, History of Multinational Corporations and Bauxite Producing Countries, Proceedings of Bauxite Symposium, No. 111, 17th July 1975). It is possible to account for practically every ton of aluminium that has ever been produced, and it is possible to say exactly who produced it (see Varsavsky, p. 47). The main features of the market structure appear to be a high degree of concentration in a few large commercial enterprises, an associated high degree of vertical integration in the industry, the transnational character of the main enterprises concerned, and the high degree of foreign ownership and control of productive activities in the raw material producing companies.

The industry is dominated by a group known as "The Big Six", the Aluminium Company of America (ALCOA), Reynolds Metal Company, Kaiser, Alcan Aluminium Ltd., Pechiney Ugine Kuhlmann and Swiss Aluminium Ltd. At least one of these firms is involved in virtually every major bauxite, alumina and aluminium project undertaken in the world. Riotinto is one of the smaller transnationals who have joined The Big Six in the last twenty-five years. In Australia bauxite is mined by three companies, Comalco, Nabalco and Alcoa of Australia. Comalco is the largest producer.

The market structure of the industry has resulted, among other things, in a virtual monopoly of advanced technology in the sector by the major companies, and their vertical integration makes it difficult to establish bauxite and alumina enterprises that are independent of these companies as they control such a large proportion of the market for these commodities (see UNCTAD Report, p. 17). The UNCTAD Report also stresses another feature of the industry, that of "transfer pricing", the technique of shifting profits and taxable income away from the country where the business is carried on to a country which imposes less taxation. This is a well-known technique of transnational corporations (see Dominant Positions of Market Power of Transnational Corporations: Use of the Transfer Pricing Mechanism, Study by the United Nations Conference on Trade and Development, U.N. New York (1978); Overseas Investment in Australia, Treasury Economic Paper No. 1, Ch. 4, May 1972; C. J. Crough, Foreign Ownership and Control of the Australian Mineral Industry). It lessens the taxable income of the persons or companies in the transnational group conducting business in the country where real profits are made. The existence of such a tax avoidance operation in favour of non-residents is a strong indication that the taxpayer is controlled by non-residents. The UNCTAD study concluded that: "Manipulation of transfer prices arises from the control exercised by a parent company over the subsidiaries and from the dominant positions of transnational corporations in the market for particular products" (p. 2). The Study states that a prime objective of corporate strategies, to which all others are directly or indirectly related, is the attainment and growth of market power. A key element of this is pricing policy, particularly the determination of transfer-pricing:

Inevitably, the more a company holds a dominant market position, nationally and internationally, and the greater the value of intra-corporate transactions, the more likely it is that the transfer pricing policies adopted will be those which aim at maximizing global corporate market power and the profits of the corporation as a whole rather than those of particular units. In order to increase profits globally, corporations try to minimize over-all tax burdens by manipulating the prices of intra-corporate transactions in goods and services so that profits are raised in countries where there are lower tax rates and reduced in countries with higher rates. Thus, a transnational corporation may decide to establish an international base company as an intra-firm trade intermediary in a tax haven, with the objective of accumulating profits in that company. Tax havens are usually located in relatively small countries or territories which have low tax rates and frequently exempt from taxes either all or most income derived from foreign sources (see par. 29, p. 7). [t]ransfer pricing decisions are usually taken at the highest levels, and higher than in the case of market price decisions. It has been said that "Executives view transfer pricing as an extremely important area of decision-making, and many of them regard it as the most critical short-run decision" (par. 53, p. 11).

In my opinion, the taxpayer was at all relevant times controlled by non-residents and its business was controlled by non-residents. There is no point in this case in distinguishing between the company and the business, which was the corporate business. I do not think that any rigid formula is appropriate for determining whether a company is controlled by non-residents. The test of whether there are more non-resident than resident directors is superficial, and in practice, would often be an absurd formalism. Business can be controlled in many ways. Traditionally, ownership was the key to control but in recent times, numerous sophisticated techniques of control such as franchise and loan agreements have evolved in response to national laws aimed at protecting revenue and maintaining national businesses. A classical method of control is the election or appointment of those who manage. In this case, the business was managed by directors and the corporation was ultimately owned by non-residents who were able, through the intermediate resident shareholders, to elect the executive and other directors (resident and non-resident). Other provisions for reversal of decisions (alternate directors, etcetera) were quite sufficient to vest the control, as distinct from the management, of the company in the non-residents.

During the tax years in question, a substantial part of the taxpayer's product, bauxite, was shipped directly to Japan to two Japanese companies (Showa Denko and Sumitomo Chemical) although it had been sold to a Hong Kong company, Comalco Bauxite (H.K.) Ltd., 52 per cent of which is owned by Comalco and 48 per cent equally by the two Japanese companies. The Hong Kong company was intended to and did relatively little except for book entries. There was no physical delivery to it. It simply purchased the bauxite from the taxpayer, for example at 33/- per ton, and sold it to the Japanese companies at 40/- per ton. The taxpayer acted on Comalco's directions and it is very relevant to the question of control that it sold to a company controlled by Comalco which was ultimately owned by those whom the Commissioner claims control the taxpayer. The taxpayer in this way conferred a very substantial benefit on the Hong Kong company and therefore on Comalco and its ultimate owners. The Chairman of the Board of Review found:

I am unable to detect any business exigency which so far as the taxpayer itself was concerned required the inter-position of a company between it and the Japanese customers. Moreover, the decision to supply the Hong Kong company with bauxite at 33/- per ton was one which was voluntarily undertaken and which in my view operated to the detriment of the taxpayer and to the advantage of its parent. I am of the opinion that the taxpayer got less for its bauxite that went to Japan than might be expected, and that in consequence the amount of taxable income which arose from the taxpayer's business was also less than might be expected.

In the Supreme Court of Victoria, Jenkinson J. found on the whole of the evidence that the business was controlled by the respondent taxpayer's directors. He stated [18] , however, that:

The evidence did establish some direction of the respondent's business by Comalco Ltd. For example, when the respondent's directors decided that a company should be incorporated in Hong Kong, that the company should buy bauxite from the respondent and re-sell the bauxite to prospective purchasers of the respondent's bauxite, and that the prospective purchasers should be minority shareholders in the Hong Kong company, the decision that the majority of the shares would be held by Comalco Ltd. and not by the respondent was forced upon the respondent's directors by reason of the general policy which had been previously laid down by Comalco Ltd. that its subsidiaries would not have subsidiaries and that the shares in all subsidiaries in the group of companies controlled by Comalco Ltd. would be held by the latter. Thus the decision of the respondent's directors that the respondent would sell bauxite to a company controlled by Comalco Ltd. was substantially influenced by a direction by Comalco Ltd. Decisions by Comalco Ltd. concerning the advance of funds to the respondent also influenced the direction of the business by the respondent's directors. 1. (1978) 8 A.T.R., at pp. 822-823; 78 A.T.C., at p. 4378.

In my opinion Jenkinson J. did not give the proper weight to these critical findings. It is necessary to investigate the structure and operations of the industry within which the company is a unit. While it is possible to analyse the formal structure of corporate authority, it is more difficult to look at the influence of organization or concentration of shareholders and the pattern of formal and informal inter-corporate relationships in which the corporation is enmeshed. The type of power exercised in these areas can rarely be sharply segregated or clearly defined (see Warwick Richards, Towards an Analysis of Corporate Power, Transnational Corporations Research Project, Working Paper No. 2, University of Sydney). In an analysis of corporate power and control of a business it is a mistake to focus on any particular business or subsidiary corporation as if it were separate. These must be considered in context as part of the industry and of the transnational group. As Richards points out (pp. 49-50),

Transnational corporations have little affinity with the notional "firm" of much of the control literature. If we begin from the operating Australian section of the base of the corporate pyramid, the analysis of the "juridical" structure of transnational corporate capital must confront an intricate web of equity relationships between corporate entities which include, in addition to wholly-owned subsidiaries and joint ventures, other forms of the direct and indirect coalescence of local and foreign capital (such as patents and licences). Analysis which focuses on "the firm" as an independent entity at any given level of this structure is chimerical

The corporate entities responsible for local phases of operations within a vertically-integrated transnational group can be meaningfully examined only within the structure of the group. Frequently, however, hybrid operating entities are spawned by joint-venturers exclusively for a specific local phase of such operations. In this situation, the analysis of latent control factors is complicated by the integration of the corporate operating unit into a complex pattern of transnational corporate groupings which are integrated at some levels of their operations and "independent" at others.

These observations are applicable to the present case.

Where the business is conducted by a subsidiary of transnationals, whose headquarters is not Australia, it would require very strong evidence to show that it was not "principally controlled by non-residents". The purpose of the transnational organization would be frustrated if there were not such control.

I am not satisfied that the full picture of control of the taxpayer has emerged in the evidence. So far as it has, it shows that the taxpayer's business is part of a transnational operation, and is controlled by transnational corporations which are non-residents of Australia. In my opinion the Chairman of the Board of Review was correct in concluding that "By the exercise of control from one company to another, and by co-operation where necessary, these two non-residents (Kaiser and Riotinto) can ensure that the business of the taxpayer is conducted as they desire". During the relevant years the taxpayer's business was controlled principally by non-residents within the meaning of s. 136. However, it is sufficient to state that the taxpayer has not discharged the onus of proving that its business was not principally controlled by non-residents.