Commissioner of Taxation v Energy Resources of Australia Limited

REASONS FOR JUDGMENT RYAN J and FINKELSTEIN J: 1 This appeal is concerned with the valuation of inventory (trading stock) for income tax purposes. When computing the assessable income to include in its returns for the financial years ended 30 June 1993 and 30 June 1994 the respondent undervalued its trading stock at both the beginning and end of each year. When the error was discovered, it requested the Commissioner to amend the assessments so that it would be assessed on its actual assessable income. The Commissioner accepted that the closing figure for the year ended 30 June 1993 was understated and agreement was reached on the correct figure. The opening figure for the 1994 year was consequently increased so that it was equal to the closing figure for the 1993 year. The closing figure for the 1994 year was also increased. Then the assessment for that year was amended accordingly. The Commissioner acknowledges that the opening figure for the 1993 year was undervalued. Nevertheless, the Commissioner did not alter this figure because it would not then be the same as the closing figure for the previous year. The Commissioner correctly says that s 170 of the Income Tax Assessment Act 1936 (Cth) precludes amending the assessment for the year ended 30 June 1992. For this reason the Commissioner contends that the value of trading stock as at 30 June 1992 cannot be altered and therefore it is not possible to change the opening figure for 1 July 1993. That contention was rejected by the trial judge who saw no reason why the correct value should not be taken into account. The question in issue is whether the trial judge or the Commissioner is correct. 2 In Australia income tax is not levied on the profits of a business. Calculating assessable income is more complicated than that. Consequently, the method by which accountants determine whether a business has made a profit or loss often will have no application in determining the assessable income of a business: The Commissioner of Taxation of the Commonwealth of Australia v St Hubert's Island Pty Ltd (in liq) (1978) 138 CLR 210, 228. Nevertheless, some accounting background is helpful in understanding the issues that arise on this appeal. 3 Sometimes courts have been required to explain the appropriateness of accounting for trading stock (that is assets held for sale in the ordinary course of business) when determining the amount of profit derived (or loss incurred) by a business. In Ostime (Inspector of Taxes) v Duple Motor Bodies Ltd [1961] 1 WLR 739, 751 Lord Reid said: "It appears that at one time it was common to take no account of stock-in-trade or work in progress for income tax purposes; but long ago it became customary to take account of stock-in-trade, and for a simple reason. If the amount of stock-in-trade has increased materially during the year, then in effect sums which would have gone to swell the year's profits are represented at the end of the year by tangible assets, the extra stock-in-trade which they have been spent to buy; and similar reasoning will apply if the amount of stock-in-trade has decreased. So to omit stock-in-trade would give a false result." See also Carden's case (The Commissioner of Taxes (South Australia) v The Executor Trustee and Agency Company of South Australia Ltd)(1938) 63 CLR 108, 155 where Dixon J said that "[t]he computation of profits from manufacture and trading has always proceeded upon the principle that the profit may be contained in stock-in-trade and 'outstandings'". 4 Accountants put the matter rather differently, though to much the same effect. According to members of that profession, determining the net profit (or loss) of a retail or manufacturing business is a two-stage process. First, it is necessary to calculate the gross profit of the business. When that figure has been ascertained the second step is to deduct from the gross profit the operating expenses of the business. It is for the purpose of establishing the gross profit that inventories must be taken into account. A common method of calculating gross profit for a given period is to subtract from the revenue produced by sales during that period the cost of goods sold (ie the cost of sales). To calculate the cost of sales it is necessary to account for inventory. Put simply, the cost of sales is the cost (value) of inventory held at the beginning of the period, plus the cost (value) of inventory purchased during the period less the cost (value) of inventory at the end of the period. The ending inventory for the period then becomes the opening inventory for the following period. There are various methods for valuing inventory, the most common being the lower of cost (which may be direct cost, absorption cost or standard cost) and net realisable value. 5 The determination of the correct value of inventory is very important, for an error in the valuation will result in a distortion of the gross profit. If the cost (value) of the closing stock is understated there will be an overstatement of the cost of goods sold and an understatement in both the gross profit and net profit. (The opposite will happen if the closing stock is overstated.) Moreover, if the error in the closing stock is not discovered there will be a distortion in the following year's figures because the cost (value) of closing stock is reproduced as the opening stock for the next period. In that year, however, the cost of goods will be understated and the gross profit and net profit will be overstated. Although the figures for each year are wrong, the total net profit over the two years will be correct because the inventory errors offset one another. 6 The first Commonwealth income tax statute, the Income Tax Assessment Act 1915 (Cth) made provision for taking into account the value of trading stock on hand at the beginning and end of an income year for the purpose of determining a taxpayer's assessable income. The trial judge reviewed the history of the legislation relevant to this appeal. That history need not be repeated. For present purposes it is sufficient to note that the relevant provisions are ss 28, 29 and 31 of the 1936 Act. These sections are found in Sub-Div B of Div 2 Part III of the 1936 Act, comprising ss 28 to 37. Relevantly, ss 28, 29 and 31 provide: "28. (1) Where a taxpayer carries on any business, the value, ascertained under this subdivision, of all trading stock on hand at the beginning of the year of income, and of all trading stock on hand at the end of that year shall be taken into account in ascertaining whether or not the taxpayer has a taxable income. (2) Where the value of all trading stock on hand at the end of the year of income exceeds the value of all trading stock on hand at the beginning of that year, the assessable income of the taxpayer shall include the amount of the excess. (3) Where the value of all trading stock on hand at the beginning of the year of income exceeds the value of all trading stock on hand at the end of that year, the amount of the excess shall be an allowable deduction. 29. (1) The value of live stock and of each article of other trading stock to be taken into account at the beginning of the year of income shall be its value as ascertained under this or the previous Act at the end of the year immediately preceding the year of income. … 31. (1) Subject to this section, the value of each article of trading stock (not being live stock) to be taken into account at the end of the year of income shall be, at the option of the taxpayer, its cost price or market selling value or the price at which it can be replaced." 7 The Commissioner says that on its proper construction the effect of s 29 is (i) to require the opening value of inventory for any income year to be the same as the closing value in the previous year; and (ii) if there is any error in the opening value that error can only be corrected if it is also possible to correct the closing value in the previous year so that the correct amount of income tax can be paid for that year. The effect of this argument, if it is accepted, is that past errors can only be rectified until a year of income is reached in respect of which the Commissioner is unable to reopen the assessment. 8 There are two decisions of the Commonwealth Taxation Board of Review which are contrary to the Commissioner's approach. The decisions are (1947) 14 CTBR Case 10 and (1956) 6 CTBR(NS) Case 12. In each case the Board decided that income tax should be assessed taking into account the actual value of inventories, irrespective of whether it is possible to amend the assessment for the previous year. 9 Since these cases were decided a number of things have happened. First, the provisions in Sub-Div B of Div 2 of Part III of the 1936 Act have been amended on at least sixteen occasions. The amendments were effected by Acts numbered 48 of 1950, 44 of 1951, 90 of 1952, 28 of 1953, 62 of 1955, 94 of 1961, 69 of 1963, 164 of 1973, 51 of 1973, 50 of 1976, 57 of 1977, 146 of 1979, 108 of 1981, 57 of 1990, 4 of 1991 and 190 of 1992. There have also been at least eleven amendments to the 1936 Act that indirectly affect the operation of Sub-Div B. The amending Acts are numbered 164 of 1973, 165 of 1973, 57 of 1977, 172 of 1978, 147 of 1979, 149 of 1979, 19 of 1980, 108 of 1981, 110 of 1981, 107 of 1989 and 78 of 1996. 10 Secondly, the 1936 Act is being progressively rewritten and replaced. The first statute in this process is the Income Tax Assessment Act 1997 (Cth). The 1997 Act is more than a consolidating statute; it has made significant changes to the 1936 Act. In particular, changes have been made to the method of accounting for inventories. Division 70 of the 1997 Act is where the relevant provisions are found. Specifically, s 70-40(1) of the 1997 Act provides: "The value of an item of trading stock on hand at the start of an income year is the same amount at which it was taken into account under this Division or Subdivision 328-E [which is not presently relevant]…at the end of the last income year." The object of s 70-40(1) is to reverse the decisions of the Board of Review on s 29. The Explanatory Memorandum to the 1997 Act (at 162) states: "Section 29 of the 1936 Act says that an item's opening value is the value ascertained under the Act at the end of the previous year. There are some Board of Review decisions concluding that those words mean that the opening value must be what the previous year's closing value should have been. If the previous year's assessment cannot be amended because of time limits, its closing value will be different from the next year's opening value. This will produce either a windfall gain or an unexpected loss for the taxpayer. The rewrite avoids the possible problem. If one year's closing value is amended, then the next year's opening value will change to reflect that amendment. If the closing value cannot be amended, the next year's opening value will still be what was recorded as the closing value. Subsection 70-40(2) supports this by ensuring that an item's opening value is nil if the item's closing value in the previous year was not taken into account at all." 11 The statutory history to which we have referred immediately raises the question whether there has been a legislative recognition of the construction which the Board has placed on s 29. Put differently, the question is whether it can be said that Parliament has in effect endorsed the Board's decisions so that it would be wrong for us now to consider the correctness of those decisions. 12 To answer this question it is first necessary to characterise the function of the Board. The Board (initially known as the Board of Appeal) was established by the Income Tax Assessment Act 1921 (Cth). Its name was changed to Board of Review by s 9 of the Income Tax Assessment Act 1925 (Cth). The existence of the Board was continued by the 1936 Act: s 178. In 1986 its functions were taken over by the Administrative Appeals Tribunal. In the 1940s and 1950s (the period in which the Board handed down the two decisions) a taxpayer who was dissatisfied with an assessment under the 1936 Act could lodge an objection to the assessment. If the Commissioner disallowed the objection in whole or in part the taxpayer might refer the Commissioner's decision to the Board for review or appeal the decision to the High Court or a Supreme Court: s 187. The Board had power to review the decision of the Commissioner (s 192) and for that purpose could exercise the powers and functions of the Commissioner (s 193). An appeal from the Board's decision could only be taken to the High Court or a Supreme Court on a point of law: s 196. The effect of these provisions was that, subject to an appeal on an error of law, the Board's decision was a final determination of the validity of a taxpayer's objection to an assessment. It is evident, therefore, that the Board exercised judicial power. Its power was no less judicial than the power exercised by the Workers' Compensation Board, as to which see: Babaniaris v Lutony Fashions Pty Ltd (1987) 163 CLR 1. 13 A consequence of finding that the Board exercised judicial power is that Parliament is to be treated as having knowledge of the construction placed by the Board on s 29. That is because Parliament is deemed to know the law: Williams v The Official Assignee of the Estate of William Dunn (1908) 6 CLR 425, 441. Of course the knowledge is indirect. It exists in the sense that those responsible for drafting taxation legislation (parliamentary counsel and other advisers) must be taken to be familiar with the law relating to that topic: Greaves v Totfield (1880) 14 Ch D 563; Dennehy v Reasonable Endeavours Pty Ltd; in the matter of Dennehy (A Bankrupt) [2003] FCAFC 158 at [16]. 14 In Re Alcan Australia Ltd; ex parte Federation of Industrial Manufacturing and Engineering Employees (1994) 181 CLR 96, 106 the High Court said that "[t]here is abundant authority for the proposition that where the Parliament repeats words which have been judicially construed, it is taken to have intended the words to bear the meaning already 'judicially attributed to [them]', although the validity of that proposition has been questioned." This proposition applies not only in cases where Parliament has repeated the words which have been construed, but also where the relevant provision has been amended and Parliament has not seen fit to override the effect of judicial decisions which construe that provision: Platz v Osborne (1943) 68 CLR 133 at 141, 146-7. The principle should also apply to amendments made to one or more of a number of provisions which, when taken together, deal exhaustively with a particular subject. Sub-Division B is in this category because exclusive provision was made for the manner in which inventories were to be brought to account. 15 In the light of the many amendments to Sub-Div B, being amendments which left undisturbed the language on which the construction by the Board of s 29 depends, it is to be presumed that Parliament has approved that construction. It is true that there is nothing in the amending legislation which expressly recognises that construction. Nevertheless, it is impossible to imagine that when Parliament enacted the amendments it advisers were unaware of the Board's decisions. The situation might be different in the case of re-enactment by a consolidating statute, where other considerations will apply. In this case our approach should be governed by the fact that the relevant words in s 29 have not changed for nearly fifty years, whereas many words in the surrounding sections have. This is sufficient for us to conclude that Parliament has accepted the meaning given by the Board to s 29. 16 There is another reason why we should not depart from the Board's construction of s 29. There are cases which hold that it is inappropriate for a court to disturb a decision which is taken to have settled the law for many years. In Bourne v Keane [1919] AC 815 Lord Buckmaster mentioned the applicable principles. He said (at 874): "Firstly, the construction of a statute of doubtful meaning, once laid down and accepted for a long period of time, ought not to be altered unless your Lordships could say positively that it was wrong and productive of inconvenience; Secondly, that decisions upon which title to property depends, or which by establishing principles of construction or otherwise form the basis of contracts, ought to receive the same protection; Thirdly, decisions that affect the general conduct of affairs, so that their alteration would mean that taxes had been unlawfully imposed, or exemption unlawfully obtained, payments needlessly made, or the position of the public materially affected, ought in the same way to continue." 17 In Babaniaris v Lutony Fashions Pty Ltd (1987) 163 CLR 1, 29 Brennan and Deane JJ explained that this principle of non-intervention is followed to "keep the scale of justice steady" (citing from Broom's Legal Maxims, 10th edn (1939) at 90). That is to say, it is an approach which promotes the certainty of the law and protects transactions which have taken place in the faith of the rule as it has been declared: Morgan v Crawshay (1871) LR 5 HL 304, 319-320; Williams v The Official Assignee of the Estate of William Dunn (1908) 6 CLR 425 at 441-442, 451-453; Re Bolton; Ex parte Beane (1987) 162 CLR 514 at 520, 531-532; Babaniaris v Lutony above at 13 per Mason J (as he then was), at 23 per Wilson and Dawson JJ; Brooks v Commissioner of Taxation (2000) 100 FCR 117, 136 (but see John v Commissioner of Taxation (1989) 166 CLR 417, 440 where the principle was not applied, notwithstanding that identified individuals had relied upon a particular decision as the basis for ordering their affairs). The principle is not, however, without exception. The restriction should not apply if the decision is plainly erroneous: The Sara (Hamilton v Baker) (1889) 14 AC 209, 222; West Ham Union v Edmonton Union [1908] AC 1, 4-5; Bourne v Keane [1919] AC 815, 874; John v Commissioner of Taxation (1989) 166 CLR 417, 440; Prebble v Commissioner of Taxation [2003] FCAFC 165. This qualification was explained by Mason J in Babaniaris v Lutony, above, at 13-14: "The fundamental responsibility of a court when it interprets a statute is to give effect to the legislative intention as it is expressed in the statute. If an appellate court…is convinced that a previous interpretation is plainly erroneous then it cannot allow previous error to stand in the way of declaring the true intent of the statute [citations omitted]…The injustice or inconvenience which will result from displacement of a long-standing decision is certainly a very important factor to be considered, but there is no support in principle or authority for the proposition that the court should persist with a manifestly incorrect interpretation on the ground that it would cause injustice or inconvenience." Nor (if this be a different exception) should the principle be employed when the meaning of the statute is plain and unambiguous: Hanau v Ehrlich [1912] AC 39; Babaniaris v Lutony aboveat 23-24 per Wilson and Dawson JJ; Craies on Statute Law, 7th edn (1971) at 156. 18 On the view of s 29 most favourable to the Commissioner, the section is at least ambiguous. Having given the matter close consideration we are not satisfied that the Board's decisions are plainly erroneous. Moreover, it would clearly be unacceptable to deny to this taxpayer the construction accepted by the Board where other taxpayers for the last fifty years have had the advantage or disadvantage (as the case may be) of this construction. 19 We also take it to be settled law that in construing a provision in a statute it is legitimate to have regard to any subsequent amendments, provided the provision under consideration is ambiguous. In Attorney-General v Clarkson [1900] 1 QB 156, 165 Sir Francis Jeune P said: "But, having regard to that Act, it seems to me that it is impossible for us to take any other view of the construction of s. 5 than that which, in my opinion, the Legislature have imposed upon us. Our duty is to interpret the meaning of the Legislature, and if the Legislature in one Act have used language which is admittedly ambiguous, and in a subsequent Act have used language which proceeds upon the hypothesis that a particular interpretation is to be placed upon the earlier Act, I think the judges have no choice but to read the two Acts together, and to say that the Legislature have acted as their own interpreters of the earlier Act." In Cape Brandy Syndicate v Inland Revenue Commissioners [1921] 2 KB 403, 414 Lord Sterndale said: "I think it is clearly established in Attorney-General v Clarkson…that subsequent legislation on the same subject may be looked to in order to see what is the proper construction to be put upon an earlier Act where that earlier Act is ambiguous. I quite agree that subsequent legislation, if it proceed upon an erroneous construction of previous legislation, cannot alter that previous legislation; but if there be any ambiguity in the earlier legislation then the subsequent legislation may fix the proper interpretation which is to be put upon the earlier." See also Deputy Federal Commissioner of Taxes (South Australia) v Elders' Trustee and Executor Company Ltd (1936) 57 CLR 610; Grain Elevators Board (Victoria) v Dunmunkle Corporation (1946) 73 CLR 70. 20 It follows that in considering the meaning of s 29 it is permissible to take s 70-45 into account. In this regard it would indeed be strange to construe s 29 as implicitly forbidding that which s 70-45 of the 1997 Act expressly forbids. To adopt such an approach would be to assume that Parliament misunderstood the operation of s 29 and that s 70-45 serves no purpose. While it is possible for Parliament to mistake the meaning of an earlier provision, that is not the position here. Section 70-45 of the 1997 Act has work to do. Its purpose is to overturn the position that prevailed under s 29. 21 Even in the absence of any guidance from Parliament, we would reject the Commissioner's construction of s 29. As already indicated, the opening value of trading stock "shall be its value as ascertained under [the 1936 Act] at the end of the [immediately preceding] year of income". According to the Commissioner this means that (i) the opening value must be the stock's value "as it was ascertained" at the close of the previous year; and (ii) the value is "ascertained" by the Commissioner. The Commissioner may "ascertain" that value either by accepting the value adopted by the taxpayer (provided that it is undertaken in accordance with one of the methods mentioned in s 31(1)) or by undertaking his own determination, regardless of whether the result is correct. We do not agree. The word "ascertain" means to determine or establish with certainty. Section 29 does not provide for the determination to be made by the Commissioner. The section only provides for the value to be determined "under the [1936] Act". Problems are often encountered when one uses the passive voice: R Dickerson, The Fundamentals of Legal Drafting (1965), 116. Such a problem exists here. Speaking strictly, value cannot be determined by legislation; only a person can make the determination. On the other hand, s 31(1) specifies how an article of trading stock is to be valued that is, at the taxpayer's option, by either its cost price, market selling value or its replacement cost. It follows, in our opinion, that when s 29 provides that the value is to be "ascertained" under the 1936 Act that means that the value must be determined in accordance with one of the three prescribed methods. Consequently, it is the correct value under the chosen method to which regard must be had. It makes no difference whether that value is determined by the taxpayer or by the Commissioner. Put another way, for the purposes of an income tax year the opening value of stock must be its actual value assessed by reference to one of the three available methods whether or not any change is made to its value in the previous year. This conclusion negatives the possibility that the Commissioner can "ascertain" the value of the stock by ascribing to it a value based on the chosen method but calculated incorrectly. 22 For these reasons we would dismiss the appeal with costs. I certify that the preceding twenty-two (22) numbered paragraphs are a true copy of the Reasons for Judgment herein of the Honourable Justices Ryan and Finkelstein.