Australian Competition and Consumer Commission v Metcash Trading Limited

Date of hearing: 24 - 26 October 2011

Division: GENERAL DIVISION

Counsel for the Appellant: Mr A Myers QC with Mr J Halley SC and Mr C Arnott

Solicitor for the Appellant: Australian Government Solicitor

Counsel for the First Respondent: Mr P Brereton SC with Mr D Roche

Counsel for the Second Respondent: Mr J Griffiths SC with Mr C Moore SC and Mr R Yezerski

IN THE FEDERAL COURT OF AUSTRALIA NEW SOUTH WALES DISTRICT REGISTRY GENERAL DIVISION NSD 1533 of 2011

BETWEEN: AUSTRALIAN COMPETITION AND CONSUMER COMMISSION

JUDGES: FINN, Buchanan, Yates JJ DATE OF ORDER: 30 NOVEMBER 2011 WHERE MADE: SYDNEY

1. The appeal be dismissed.
2. The appellant is to pay the respondents' costs. Note: Entry of orders is dealt with in Rule 39.32 of the Federal Court Rules 2011.

IN THE FEDERAL COURT OF AUSTRALIA NEW SOUTH WALES DISTRICT REGISTRY GENERAL DIVISION NSD 1533 of 2011

BETWEEN: AUSTRALIAN COMPETITION AND CONSUMER COMMISSION

JUDGES: FINN, Buchanan, Yates JJ DATE: 30 NOVEMBER 2011 PLACE: SYDNEY

FINN J 1 I have had the advantage of reading the reasons of Buchanan J and of Yates J. I agree with the reasons, and the orders proposed by, Yates J. I would add that it is unnecessary for the reasons given by Yates J to express a concluded view on the proper construction of, and standard of proof that is to be applied in cases coming under, s 50(1) of the Competition and Consumer Act 2010 (Cth). I refrain from doing so. I certify that the preceding one (1) numbered paragraph is a true copy of the Reasons for Judgment herein of the Honourable Justice Finn.

Associate: Dated: 30 November 2011 IN THE FEDERAL COURT OF AUSTRALIA NEW SOUTH WALES DISTRICT REGISTRY GENERAL DIVISION NSD 1533 of 2011

BETWEEN: AUSTRALIAN COMPETITION AND CONSUMER COMMISSION

JUDGES: FINN, Buchanan, Yates JJ DATE: 30 November 2011 PLACE: SYDNEY

Introduction 2 I agree with Yates J that the appeal should be dismissed. His Honour's analysis has relieved me of the burden of writing separate reasons to the same effect. However, there are some matters about which I wish to make some specific comments.

Hypotheses and comparisons 3 This case must be approached recognising the essentially uncertain foundation for the conclusions which the Court was asked to adopt, and the source of uncertainty. The first source of uncertainty was what would actually happen if the acquisition did not proceed. The second source of uncertainty was what would be the "likely" effect on competition if the acquisition did not proceed - i.e. by comparison with the assumptions made if it did proceed. The question of the standard of proof to be met to resolve those uncertainties is the main question I propose to address. 4 The Australian Competition and Consumer Commission ("ACCC") claimed, in the present proceedings, to be entitled to succeed in blocking the proposed acquisition by Metcash of Franklins, and its (primarily) retail assets if: (i) it could show that there was a "real chance" of the occurrence of an hypothesis advanced by it about what would happen if the proposed acquisition was blocked; and (ii) on that hypothesis (i.e. as a "real chance" of it occurring even if not probable) there was a "real chance" that there would be a substantial lessening of competition (i.e. as compared with what would (presumably) happen if the acquisition proceeded). 5 The ACCC claimed that the twin tests it proposed were supported by authority in this Court. The trial judge accepted that he should apply the "real chance" test at the second stage, but not the first. His Honour took the view that, at the first stage, the ACCC was obliged to establish its chosen hypothesis (or any hypothetical position) on the balance of probabilities. 6 The circumstances of the present case are such as to call into question the soundness of the authorities relied on by the ACCC. In any event, they do not go so far as to support the application of a "real chance" test at the first stage proposed by the ACCC. If the ACCC was correct about its two stage application of the "real chance" test, the case would have to be decided upon a position where speculation was heaped on speculation. That outcome would not, in my view, be consistent with the application of a proper judicial method. Nor is it required by statute. Further, in my view, application of the "real chance" test even at the second stage also presents problems. The circumstances of the present case, with its concentration on comparisons which are all in the future, provide an illustration of the danger of descending into the realm of conjecture. That seems to me to provide a reason to question the correctness of the statutory construction which lies behind the adoption of the test in some cases and it should be revisited. 7 Before I deal with those questions, I need to say something briefly about some other matters which were to the forefront of the ACCC's case.

Assumptions and economic theory 8 The ACCC's case theory seemed to me to incorporate the following broad assumptions: 1. Monopolies lead to abuse of market power; 2. Profits higher than "normal" are uncompetitive because they illustrate that an abuse of market power has occurred, or at least that there is an absence of desirable levels of competition; 3. The possibility that a profit higher than "normal" might be made is sufficient proof of an absence of desirable levels of competition; 4. The possibility that existing levels of profit might increase to be higher than "normal" is sufficient proof of likely lessening of competition. 9 However useful assumptions of this kind (and the economic theories on which they are based) may be as the foundation for an economic model from which to begin examination of a particular factual situation (or hypothesis) as a matter of theory, great care must be taken that such assumptions do not become embedded in the application of statutory or other legal tests as a substitute for the fact finding and analysis required by a proper application of the judicial method. 10 In the present case, the statutory tests to be applied are set out in s 50 of the Competition and Consumer Act 2010 (Cth) ("Competition Act"). The case against Metcash was obliged to satisfy those tests, rather than some more theoretical position. In that context, "market" is defined. It means "a substantial market for goods or services" (s 50(6)). The task of examining whether a substantial lessening of competition is likely in given circumstances obviously requires that a sufficiently relevant market be identified. Clearly, having regard to the statutory context, that is intended to reflect commercial reality, rather than be driven by economic theory. In the present case, there was little reason to depart very far from a consideration of the actual commercial environment in which the competing possibilities would play themselves out.

The market in the present case 11 In the present case, the ACCC's efforts were largely devoted to identifying, describing and limiting a market which would provide a sufficient illustration of the fourth assumption set out above. Accordingly, it proposed a market in which the major retailers were absent and most of Franklins' own operations were excluded. The ACCC defined the market in a way which excluded the 80 Franklins corporate stores, and included wholesale supply of only nominated packaged goods to 10 franchise stores. The market was confined to limited goods which did not represent the whole range of goods supplied by Metcash to its own customers or those supplied by Franklins to its franchisees. This proposed market excluded all services offered and supplied by Metcash to its customers which were connected with the retail operations of those customers, even though on the evidence the supply of those services was fully integrated with, and part of the service of, the supply of goods at a wholesale level - both those included by the ACCC and those which were not. 12 In my view, this approach, and the theories which were called in aid to justify it, represented a distraction from the real questions for attention. The postulated market was correctly rejected by the trial judge. The market was not identified by reference to the dynamics and constraints really at work, but by reference to the need to supply a foundation for the hypothesis which the ACCC wished to offer about the future state of the "market". As the trial judge found, however, the market forces operating as a constraint upon Metcash were not satisfactorily identified by the ACCC's thesis. Those forces were at work in a broader context that necessarily included reference to retail operations and retail sales. Once the market suggested by the ACCC was rejected its case could not succeed. That obstacle is insurmountable. It is a reason why the appeal cannot succeed on any view.

The "wholesale assets" 13 There was a further problem with the approach taken by the ACCC. Its hypothesis for "likely" competition, if the acquisition by Metcash was prevented, depended on the proposition that Franklins' "wholesale assets" would be acquired by a third party and on the suggestion that a new, successful, wholesale supply business, of a different kind from that operated by Franklins, would spring up if Metcash was prevented from acquiring Franklins' retail business. However, the way that the so-called "wholesale assets" were identified immediately called into question the ACCC's attempt to define the market in the way it did (which theoretically excluded retail activity). 14 The "wholesale assets" the ACCC identified in its pleadings were only deployed by Franklins itself in the suggested wholesale market to a very limited degree. The three types of "wholesale assets" identified were: (a) agreements with IT licensors, logistics providers, manufacturers and other suppliers under which Franklins acquires and supplies 'wholesale packaged groceries' … to its Eighty Corporate Stores and Ten Franchise Stores, for retail sale to the public; (b) associated assets owned or controlled by Interfrank which assist in acquiring and supplying wholesale packaged groceries to its Eighty Corporate Stores and Ten Franchise Stores; and (c) retail support assets including pricing systems, expertise in promotions and advice provided to its Eighty Corporate Stores and Ten Franchise Stores. 15 It is apparent from the way these assets are identified that it is inaccurate to describe them as "wholesale assets". In each case those assets were deployed to service the 80 corporate stores (which the ACCC excluded from its defined market) as well as the franchise stores. The so-called "wholesale assets" represented the infrastructure and arrangements whereby Franklins conducted the whole of its vertically integrated retail business and, out of which, as a minor part of its own activities, and a very small part of the suggested wholesale market, also serviced its franchise stores. 16 The ACCC's case was that at least a significant majority of the 80 corporate Franklins stores would be acquired by "a third party or third parties". The connection between this suggestion and a cogent argument about likely future competition in the "market" chosen by the ACCC proved difficult to establish. If the suggested purchaser self-supplied, supply of goods (using the "wholesale assets") to such stores could not be part of the "market" defined by the ACCC. The other possibilities were that the suggested purchaser would use the "wholesale assets" to establish a wholesale business to supply the stores (i.e. presumably in the hands of some other ultimate purchaser(s)) or would sponsor someone else to establish a wholesale business to supply the stores, if it retained ownership of them. None of these possibilities provides much of a foundation for the proposition that in the separate wholesale market postulated by the ACCC a new entrant would appear and either do what Franklins had not done, or operate in the same way as Franklins and succeed where Franklins had not. The basic scenario was, therefore, itself full of difficulties and conjecture. 17 In my view, there was no adequate foundation for the supposition that a new wholesale business, of a kind not earlier existing, would come into existence, based somehow on the acquisition of the "wholesale assets". If it did, it would not reflect what had been done by Franklins. Franklins had failed. Any competition it represented was at an end, independently of anything Metcash did.

The hypothetical purchaser 18 Apart from the difficulties just referred to, there were substantial difficulties in satisfactorily identifying a suggested purchaser. The possibilities for the hypothetical purchaser were identified, in the ACCC's pleadings, as follows: (a) the SPAR Group; (b) Bidco [this was the KKKL consortium]; (c) other private equity interests, presently unknown to the Commission or which have not yet expressed interest, looking to make a return on investing in the Wholesale Assets (as the Spar Group, the Bidco consortium and Metcash are each seeking to do) in the event that the Proposed Transaction does not proceed; and (d) other vertically integrated overseas supermarket chains, presently unknown to the Commission or which have not yet expressed interest, looking to expand into Australia (as Pick n Pay did when it first invested in the Franklins business) in the event that the Proposed Transaction does not proceed; or (e) a joint consortium of purchases by the above in combination with one another or other financiers presently unknown to the Commission. 19 The possibilities in (a), (c), (d) and (e) may be immediately dismissed. The last three were entirely speculative and were never developed. During the course of its case any reliance by the ACCC on SPAR was also abandoned. Attention was then concentrated on "Bidco", which was the suggested KKKL consortium. The consortium consisted of proprietors and ultimate owners of retail businesses. The trial judge decisively rejected this hypothesis and the idea that the consortium was likely (in any sense) to establish a wholesale business of the kind postulated. Nevertheless, on the appeal, the ACCC pursued the idea that it was entitled to argue for a "real chance" of that hypothesis coming to fruition, and entitled to succeed on that basis. 20 The contention cannot survive the factual findings of the trial judge but it leads me to the further discussion which follows. The consequence revealed by the present case, of the application of the "real chance" test upon which the ACCC insisted, seems potentially so unsatisfactory that I cannot refrain from questioning, in the light of the facts of the present case at least, the foundation for the test upon which the ACCC relied.

The standard of proof 21 The argument by the ACCC about the "real chance" test is necessarily an argument about the elements which must be satisfied in a case under s 50 of the Competition Act, and the burden and standard of proof associated with establishing those elements. 22 Section 50(1), (3) and (6) provide: 50(1) A corporation must not directly or indirectly: (a) acquire shares in the capital of a body corporate; or (b) acquire any assets of a person; if the acquisition would have the effect, or be likely to have the effect, of substantially lessening competition in a market. … (3) Without limiting the matters that may be taken into account for the purposes of subsections (1) and (2) in determining whether the acquisition would have the effect, or be likely to have the effect, of substantially lessening competition in a market, the following matters must be taken into account: (a) the actual and potential level of import competition in the market; (b) the height of barriers to entry to the market; (c) the level of concentration in the market; (d) the degree of countervailing power in the market; (e) the likelihood that the acquisition would result in the acquirer being able to significantly and sustainably increase prices or profit margins; (f) the extent to which substitutes are available in the market or are likely to be available in the market; (g) the dynamic characteristics of the market, including growth, innovation and product differentiation; (h) the likelihood that the acquisition would result in the removal from the market of a vigorous and effective competitor; (i) the nature and extent of vertical integration in the market. … (6) In this section: market means a substantial market for goods or services in: (a) Australia; or (b) a State; or (c) a Territory; or (d) a region of Australia. 23 Section 50 erects a statutory prohibition. That prohibition may, in an appropriate case, be enforced by the Court by way of injunction or declaration. When such relief is sought, in aid of the statutory scheme, it is necessary for the party seeking relief to make out a proper case for that relief. 24 The trial judge, applying authorities to which I shall refer, accepted that whether a suggested effect on competition was "likely" was to be tested by asking whether there was a "real chance" of it occurring. However, his Honour took the view that the factual foundation upon which such a question should be answered would first need to be established on the balance of probabilities. Applying that approach to the issues in the present case, his Honour applied a "balance of probabilities" test to the ACCC's hypothesis about likely competitors to Metcash if the acquisition did not proceed, and a "real chance" test to the question whether, if the hypothesis was accepted, it was likely that substantially lessened competition would be the result. 25 On the appeal the ACCC argued that the "real chance" test should be applied to all aspects of the question arising under s 50, including whether it had made good the hypothetical factual foundation upon which to then ask whether it was likely that substantial lessening of competition would occur. In my view, this argument should not be accepted. Indeed, I do not agree that the "real chance" test should be applied at all to the application of s 50. 26 First, a general observation might be made that there are generally only two standards of proof contemplated in proceedings involving the exercise of federal judicial power - proof to the criminal standard and proof to the civil standard. It is worth noting that the Evidence Act 1995 (Cth) provides in s 140: 140(1) In a civil proceeding, the court must find the case of a party proved if it is satisfied that the case has been proved on the balance of probabilities. (2) Without limiting the matters that the court may take into account in deciding whether it is so satisfied, it is to take into account: (a) the nature of the cause of action or defence; and (b) the nature of the subject-matter of the proceeding; and (c) the gravity of the matters alleged. 27 The Evidence Act applies to all proceedings in a federal court (s 4). It does not, with limited exceptions, affect the operation of the provisions of any other Act (s 8). However, there is nothing in the Competition Act or in s 50 of the Competition Act which, in my view, alters the general rule stated in s 140 of the Evidence Act. It seems to me to follow that a case for relief in this Court, relying on s 50, must be established on the balance of probabilities. That said, deference to authority requires a more detailed analysis. 28 The establishment of any case requires conclusions about facts, to which a relevant legal standard may be applied. In the present context, the relevant legal standard is stated by s 50. I shall return in a short while to discuss the nature of the test upon which the operation of the prohibition in s 50 depends. First, it is necessary to say something about the way in which the facts, to which the legal standard must be applied, are to be established. 29 On the appeal the ACCC argued that the factual foundation to which the test in s 50 should be applied in the present case, need not be established to the standard expressed in s 140 of the Evidence Act. The ACCC argued that, not only need it only prove as a "real chance" that a substantial lessening of competition would result from the hypothesis it advanced, but also that it need only show a "real chance" that the hypothesis might come about. It follows from this approach that the ACCC argued that the Court would enforce the statutory prohibition in s 50 even if satisfied that it was more likely than not that the hypothesis advanced by the ACCC would not come to pass. That seems a very unsure foundation for the enforcement of a statutory prohibition having serious commercial consequences. It amounts to a suggestion that the authority of the Court would be used to prevent an acquisition when the Court assessed that the hypothetical facts used as a foundation to suggest a likely substantial lessening of competition would, in all likelihood, not be in place. That, indeed, was the effect of the factual findings in the present case. That would impose an almost inescapable obstacle to a very wide range of potential acquisitions. The application of a "real chance" test to the assessment required by s 50 concerning a likely effect on competition has similar consequences. I shall discuss that shortly. 30 Facts may be established directly or by inferring them from other facts which have themselves been proved. Necessarily, where prediction about the future is involved, the whole process of predicting future facts involves inference from otherwise demonstrated circumstances. Inferential reasoning, in a legal context, must take place by reference to the standard of proof which is involved. 31 In a criminal case, proof beyond reasonable doubt necessitates proof of every element of an offence, also beyond reasonable doubt, to maintain the overall integrity of the criminal process (see Momcilovic v R (2011) 85 ALJR 957; 280 ALR 221 per French CJ at [53] citing Howe v R (1980) 55 ALJR 5 at 7; 32 ALR 478 at 483). Inference does not mean conjecture, even in a civil case. In civil proceedings the inferential process "may fall short of certainty, [but] must be more than an inference of equal degree of probability with other inferences, so as to avoid guess or conjecture" (see Lithgow City Council v Jackson (2011) 85 ALJR 1130; 281 ALR 223 per Crennan J at [94], Bradshaw v McEwans Pty Ltd (1951) 217 ALR 1 at 6, Luxton v Vines (1952) 85 CLR 352 at 358, Jones v Dunkel (1959) 101 CLR 298 at 305). A court is not authorised to choose between guesses, even on the ground that one guess seems more likely than another or others. 32 These strictures applied, in my view, to the establishment of the factual foundation upon which the ACCC wished to contend that the acquisition by Metcash of the Franklins' assets would be likely to have the effect of substantially lessening competition in a market. Even if assessment of that contention could (let me accept it is so for this purpose) proceed by reference to whether there was a "real chance" that those factual circumstances would have the effect of substantially lessening competition in a market, that would not, in my view, authorise a process of fact finding which departed from the necessity to find or infer facts to the ordinary civil standard. The "real chance" test, if it applies, does not reach backwards to affect or reduce the discipline or level of proof required in establishing a proper factual foundation from which to argue for ultimate conclusions to sustain a cause of action. It follows that I agree with the approach taken by the trial judge to this issue. Any other approach amounts to saying that the civil standard of proof of factual matters has been abandoned with respect to s 50 and many other provisions of the Competition Act so that the serious (and sometimes commercially damaging) restraints imposed by the Competition Act may be activated at the election of a regulator upon the basis of hypotheses and suppositions which reach only a level of respectability - i.e. respectable "guesses" as opposed to the application of the ordinary judicial method. 33 In many cases, a proposed acquisition will present for consideration a well defined factual consequence. If, for example, there are two (or a small number) of well established competitors and one proposed simply to take over, merge with, or absorb another the future possibilities may be very straightforward. They may be as simple as comparing a market with two (or three or four) well established competitors with a market involving one less, and one with enhanced capacity. If one starts with two competitors only, the "likely" (whatever that word means) result may be monopoly. If one starts with three, the "likely" result may be market dominance. The alternative may be status quo. This is straightforward. It is not the present case. 34 Independently of any acquisition by Metcash, Franklins was leaving the market, whatever it was. It was leaving the market because its business was unprofitable and was failing. That was the commercial reality. The owners of Franklins, and its associated assets, had commenced the process of realising what they could for those assets. Upon Franklins' exit from the market any competition which it offered would cease. It was not the possibility of acquisition by Metcash which would have any initial effect on competition. This was not a takeover case. It was not a case where the status quo might be retained if the acquisition did not proceed. 35 As Franklins was not going to remain in the market, whatever happened, there was no status quo available for comparison. All comparisons were with future possibilities. All future possibilities required predictions. To propose that the ACCC could meet its obligation of providing a factual foundation for the comparison it proposed about likely competition by saying there was a "real chance" of something happening, unmistakeably invites speculation. 36 In my view, it was necessary to establish, on the balance of probabilities, what would happen if the acquisition proceeded and, importantly for the present case, if it did not proceed. Only then could the test in s 50 be applied. The application of a "real chance" test, even at this (second) point, also has the consequence, so far as s 50 is concerned, that the Court may be required to find the statutory prohibition operative when, in all likelihood, the suggested possible effect on competition will not occur. That also seems a strange, and unsatisfactory, result.

The meaning of "likely" 37 Miller's Australian Competition and Consumer Law Annotated (33rd ed) 2011 says this (at [1.50.30], 717): The word "likely" has various shades of meaning. It may mean "probable" in the sense of "more probable than not", "more than a 50 per cent chance". It may mean "material risk" as seen by a reasonable man "such as might happen". It may mean "some possibility" more than a remote or bare chance or it may mean that the conduct engaged in is inherently of such a character that it would ordinarily cause the effect specified … 38 Miller also suggests (at [1.45.45]): "It would appear that, in ss 45, 46 and 47 [of the Competition Act], the term 'likely' means that there is a real chance or possibility, rather than 'more likely than not' …". Three cases are cited in support of that conclusion: Monroe Topple & Associates Pty Ltd v Institute of Chartered Accountants in Australia (2002) 122 FCR 110 ("Monroe Topple"); Universal Music Australia Pty Ltd v ACCC (2003) 131 FCR 529 ("Universal Music"); and Seven Network Ltd v News Ltd (2009) 182 FCR 160 ("Seven Network"). I shall refer to each of them in due course. By contradistinction, Miller suggests that the meaning of "likely" in s 50 of the Competition Act has not been settled. 39 The same, or a similar, drafting technique of juxtaposing reference to a factual circumstance ("has the effect" (s 45 and s 47(10)), "would have the effect" (s 45D and s 50), "is misleading or deceptive" (Sched 2, s 18)) with a less determinative conclusion ("is likely to have the effect" (s 45 and s 47(10)), "would be likely to have the effect" (s 45D and s 50), "is likely to mislead or deceive" (Sched 2, s 18)) occurs in a number of places in the Competition Act. In some cases it is relatively easy to see that the drafter has drawn a distinction between what has happened and what might happen (ss 45 and 47(10)). Similarly, a distinction between something that is misleading or deceptive and something that is likely to mislead or deceive (Sched 2, s 18) is not conceptually too difficult. Necessarily, in the situations so far mentioned, the concept of "likely" arises in connection with a prediction about something not then manifest. In other cases (e.g. involving s 50) the distinction is less clear. Postulating that something would have a particular effect may be a firmer prediction than saying it would be likely to have that effect, but neither involves identification of an effect already manifest. Both require prediction. 40 In some (but not all) of the cases to which I shall turn shortly the perceived (but in my view non-existent) tension between the two ways of expressing the test has led to the conclusion that the first must be established on the balance of probabilities and (as a result) the second by reference to some lesser standard, often referred to as the "real chance" test. In my respectful view, that is an error of analysis and an incorrect approach to the issue of construction. The first limb of the test allows a prediction of probable (therefore likely) consequence, without account being taken of, or allowance needed for, other contingencies. The second limb concentrates on the quality of the circumstances, and the probable consequence, without permitting falsification of that probability by proof of the actual occurrence of some inherently less probable result. In other words, establishing a probable consequence will suffice, even if in fact it did not occur. In the leading case on the issue, that is what happened. Both limbs of the test, as found for example in s 50, in my view require the same standard of proof (balance of probabilities) and they should probably be regarded as constituent elements of a compound conception. As a matter of ordinary language I see no tension or inconsistency between them. As a matter of ordinary language I see no adequate foundation for concluding that the second limb of such a test imports and applies, at any stage of the process, a departure from the ordinary civil standard of proof on the balance of probabilities. 41 Some of the authorities support this view; some do not. At present, the authorities against it are the more influential. The test ("real chance") to which they give support was followed by the trial judge. 42 Australian Telecommunications Commission v Krieg Enterprises Pty Ltd (1976) 27 FLR 400 ("Krieg") dealt, amongst other things, with a statutory liability for interference with or damage to property under the control of the Australian Postmaster-General where there was "reasonable cause to believe that the doing of the work [was] likely" to result in such interference or damage. Bray CJ observed that the section (s 139B of the Post and Telegraph Act 1901-1973) had never been the subject of judicial interpretation. Bray CJ posed the question for his decision as follows (at 406): The real point of the case turns on the word "likely" in sub-s. (1) (a). Is that synonymous with "probable" or, in other words, with there being more than a fifty per cent chance of the event happening? Or is it enough if Mr. Field had reasonable cause to believe that there was some possibility more than remote or "bare", to use the phrase preferred by Mr. von Doussa, of damage or interference? 43 Bray CJ thought that the word "likely" bore its ordinary meaning, as a synonym of the word "probable", or more likely to happen than not. His Honour went on, as follows (at 407): … After all, the converse of "likely" is "unlikely" and the converse of "probable" is "improbable" and if an event can properly be described as likely though not probable, or probable though not likely, then it can also be described as unlikely though probable or improbable though likely. This seems contrary to common usage and, I am inclined to say, to commonsense. It may be sufficient to say that the natural and ordinary meaning of the word "likely" is the one I have indicated and prima facie that is its meaning in the statute and that there is nothing in the context of this statute that I can see to indicate any other meaning. But, in deference to Mr. von Doussa's able and subtle argument based on the concepts of likelihood and probability as contained in traditional formulations of the common law of negligence, I will say something more. 44 There followed an extended discussion of the common law position. Then Bray CJ said (at 410): I do not, however, regard these difficult semantic questions about the proper meaning of familiar adjectives in the various formulations of common law doctrine as decisive of the point in issue, or even very relevant. The common law depends on concepts, not on words. It is not bound to the particular phraseology adopted by a particular court. It can be refined on and expounded in more or less detail according to the circumstances of the particular case. Here we are concerned with the word "likely" in a statute. As I have said, the ordinary and natural meaning of the word is synonymous with the ordinary and natural meaning of the word "probable" and both words mean, to adopt the expression of Lord Hodson in the passage previously quoted, that there is an odds-on chance of the thing happening. That is the way in which statutes containing the words have usually been construed: see, for example, Re Bayer Products' Ltd.'s Application, per Lord Greene M.R. and per Asquith L.J.; Dowling v. South Canterbury Electric Power Board; Transport Ministry v. Simmonds. Particularly is this so when the statute is a penal statute (see Transport Ministry v. Simmonds or, I think, where, as here, an additional liability in tort beyond the common law liability is being imposed. I think that is the meaning which should be attached to the word "likely" in sub-s. (1) (a). It is the natural and ordinary meaning and there is nothing to show that another meaning was intended. Like the learned special magistrate, therefore, I think that "likely" in the subsection means "probable" and I think that that means that there is a more than fifty per cent chance of the thing happening. 45 Two years later this Court dealt with a case which directly involved s 50 of the Trade Practices Act 1974 (Cth) ("TP Act") (now s 50 of the Competition Act) although, as will be seen, some differences in the statutory test must be noted. In Trade Practices Commission v Ansett Transport Industries (Operations) Pty Ltd (1978) 32 FLR 305 ("Ansett Industries") Ansett Transport Industries Ltd offered to purchase all the issued shares in the capital of Avis Rent-A-Car System Pty Ltd. The offer was accepted. The Trade Practices Commission commenced proceedings seeking an injunction alleging that the acquisition of shares contravened s 50. 46 Section 50 was then in different terms. It did not refer to competition but referred to a corporation being (or being likely to be) "in a position to control or dominate a market for goods or services". A previous version of s 50, however, had been in a form which is closer to the present formulation. Northrop J referred to those matters as follows (at 316 and 317): The relevant market having been identified, it is necessary to determine whether, as a result of the acquisition of the shares in the capital of Avis, Ansett Operations "would be or be likely to be, in a position to control or dominate" that market. Before turning to the evidence, it is necessary to discuss the interpretation of s. 50 of the Act. … Section 50 as contained in the 1974 Act was markedly different from s. 50 as introduced by the 1977 Act. The former s. 50(1) was in the following form: "A corporation shall not acquire, directly or indirectly, any shares in the capital, or any assets, of a body corporate where the acquisition is likely to have the effect of substantially lessening competition in a market for goods or services". That section proscribed the acquisition of shares where the acquisition was "likely to have the effect of substantially lessening competition in a market". 47 It will be noted that, in the original formulation, the only test was whether an acquisition would be "likely to" have a particular effect. In the altered formulation introduced in 1977 the formulation referred to both actual and likely circumstances. Northrop J found, for reasons which were very extensively expressed, that the result of the share acquisition was not to immediately place Avis in a position to dominate the car rental market in Australia. His Honour then turned to another aspect, saying (at 339-340): It now becomes necessary to consider the secondary case put by the commission. This depends upon additional factors being provided by Ansett Operations and by which the position of Avis is enhanced to such an extent that Ansett Operations will be in a position to dominate the car rental market in Australia. … I have held that Avis is not in a position to dominate the market, and it follows that the mere acquisition of all the shares in the capital of Avis by Ansett Operations does not of itself place Ansett Operations in a position to dominate the market. For present purposes, the crucial words are "would be likely to be". This expression connotes a consideration of what is likely to happen in the future. 48 His Honour addressed that question by applying the conventional standard of proof on the balance of probabilities. His Honour said, for example (at 344): The mere existence of an economically strong company in a corporate relationship with an operator in a particular market in Australia does not lead by itself to the conclusion, on the balance of probabilities, that the economically strong company would be or be likely to be in a position to dominate that particular market. and, more directly (at 346): Although it is sufficient for the commission to establish that Ansett Operations would be likely to be in a position to dominate the car rental market as a result of the acquisition of the shares in Avis, the issue of whether Ansett Operations is in that position must be judged in the light of commercial probabilities. His Honour made a further reference to satisfaction on the balance of probabilities at 347. 49 With respect, this was a firm, straightforward and conventional approach to an issue which is at the heart of the operation of the civil justice system - namely, the standard of proof to be met by a party initiating civil proceedings against another. 50 The point of departure from the approach taken by Bray CJ in Krieg and Northrop J in Ansett Industries is the judgment of Deane J in Tillmanns Butcheries Pty Ltd v Australasian Meat Industry Employees' Union (1979) 42 FLR 331 ("Tillmanns Butcheries"), although it should be noted that no reference at all was made in Tillmanns Butcheries to the judgment of Northrop J in Ansett Industries. 51 Tillmanns Butcheries concerned s 45D of the TP Act (also s 45D of the Competition Act). Section 45D required proof of a number of elements: acting in concert; hindering or preventing the supply of goods or services; for the purpose of causing substantial loss or damage; and "the conduct … would have or be likely to have the effect" of causing substantial loss or damage. The case at first instance was heard by St John J (Tillmanns Butcheries Pty Ltd v Australasian Meat Industry Employees' Union (1978) 34 FLR 494). St John J took the view, on the facts of that case, that loss or damage within the meaning of s 45D had not been proved. As it would have been open to the plaintiff to prove that substantial loss or damage had, in fact, been caused by the impugned conduct St John J took the view that (at 498): … the plaintiff company is not entitled to argue that the conduct would be likely to cause substantial loss where the means of proving whether in fact it did cause substantial loss is in its power. 52 That approach was reversed on appeal, because the view was taken that the conduct of the respondent union was of a character, and undertaken with such a purpose, that it was likely to cause substantial loss or damage (at whatever standard proof was necessary) whether or not it actually did so in that case. It was also relevant that the effect of the conduct had been limited by an injunction which prevented it continuing. This was a case, therefore, where a probable result could be said to be contradicted by what happened. Once that was removed as an obstacle of construction the case for relief was a clear one. 53 It was not necessary, in order to determine the appeal and uphold it, to reach or express a firm or final view about the standard of proof to be employed. Bowen CJ, with whom Evatt J agreed, explicitly declined to do so. 54 Bowen CJ said (at 339): The word "likely" is one which has various shades of meaning. It may mean "probable" in the sense of "more probable than not" - "more than a fifty per cent chance". It may mean "material risk" as seen by a reasonable man "such as might happen". It may mean "some possibility" - more than a remote or bare chance. Or, it may mean that the conduct engaged in is inherently of such a character that it would ordinarily cause the effect specified. 55 After referring to some cases (both Australian and American) in the general area of discourse Bowen CJ said (at 340): The circumstances to which s. 45D may apply are so various that I hesitate to place a gloss on the section by preferring one meaning of "likely" rather than another for the determination of this particular case. It is unnecessary to do so, because I have formed the view that whichever meaning is adopted the evidence leads me to the conclusion that the likelihood of substantial loss or damage has been established. 56 It was urged upon us by counsel for the second respondent to the present appeal that we should exercise similar restraint. That is a suggestion which has a good deal to recommend it but the circumstances are now quite different. Substantial confusion has in my view arisen about this issue and it would be better to address it. In addition, the issue has arisen directly in the present case. It was one with which was necessary for the trial judge to grapple and it is one in respect of which the appellant has argued, in substance as I see it, for a further relaxation of the standard of proof. 57 Deane J in Tillmanns Butcheries dealt with the matter at greater length. His Honour said (at 346): The word "likely" can, in some context, mean "probably" in the sense in which that word is commonly used by lawyers and laymen, that is to say, more likely than not or more than a fifty per cent chance … It can also, in an appropriate context, refer to a real or not remote chance or possibility regardless of whether it is less or more than fifty per cent. When used with the latter meaning in a phrase which is descriptive of conduct, the word is equivalent to "prone", "with a propensity" or "liable". ... Thus, if I fire a rifle through drawn curtains into a quiet lane in a country village, it is not likely, in the sense of more likely than not or an odds-on chance, that I will injure anyone. It would, however, be difficult to deny that there was a real chance or possibility (or likelihood in that sense) that an occasional passer-by would be wounded by the bullet. Plainly, the act of firing a rifle through drawn curtains into a lane used by pedestrians would be an act which was, in the circumstances, prone or liable (likely in that sense) to cause injury to a passing pedestrian. (References omitted) 58 Deane J expressed his conclusion about this issue in the following way (at 347-348): The conclusion which I have reached is that, in the context of s. 45D (1), the preferable view is that the word "likely" is not synonymous with "more likely than not" and that if relevant conduct is engaged in for the purpose of causing loss or damage to the business of the relevant corporation, it will suffice, for the purposes of the subsection, if that conduct is, in the circumstances, such that there is a real chance or possibility that it will, if pursued, cause such loss or damage. Whether or not such conduct is likely (in that sense) to have that effect is a question to be determined by reference to well-established standards of what could reasonably be expected to be the consequence of the relevant conduct in the circumstances. In determining the answer to that question, it will be relevant that the persons engaging in the conduct did so with the purpose of causing such loss or damage. (Emphasis added) 59 It will be apparent from what I said earlier, that Deane J's analysis of the statutory test (and his rejection elsewhere in the judgment of the approach taken by Bray CJ in Krieg) are not matters with which I agree. However, even accepting the analysis for present purposes it does not, in my view, apply to s 50 of the Competition Act. 60 The phrases I have emphasised in the last extract seem to me to be of real significance to his Honour's reasoning in the statutory context in which the discussion occurred. Purpose and likelihood were (and are) conjoined elements in the overall satisfaction of the statutory test under s 45D of the Competition Act. That may not be said about s 50. 61 In my respectful view, it is necessary to approach the reasoning employed by Deane J in Tillmanns Butcheries with the realisation firmly in mind that the notion of purpose was directly linked by his Honour, through the prism of foreseeability, with anticipated results - "what could reasonably be expected to be the consequence of the relevant conduct in the circumstances" - i.e. having regard to the purpose of causing loss or damage. That set of circumstances finds no counterpart in s 50 and, in my respectful view, the reasoning employed by Deane J in Tillmanns Butcheries is not appropriate to the construction of s 50. 62 Bowen CJ referred to the purpose for the conduct in the case in the following terms (at 339): In the case before us the ban affected the raw materials for Tillmanns' business and even though it was a large and diversified butchery and smallgoods producer it nevertheless depended on meat. There is evidence for an inference that the ban was intended to be a total one. There is evidence too that Mr. Tillmann was not prepared to relent and that the union was quite prepared for the ban to last indefinitely. In my view, having regard to the number of beasts involved in the ban in the present case and the circumstances proved, the proper conclusion is that the purpose was to cause substantial loss or damage. 63 It was a natural inference to draw that conduct involving an outright ban of supplies which might last indefinitely until capitulation, and which was done for the purpose of causing the target business substantial loss and damage, was likely to cause that loss or damage. Bowen CJ and Evatt J were satisfied about that element whatever standard of proof was involved. There is no reason to think that Deane J was satisfied to some lesser standard of proof. The contrary is clearly the case. Deane J said (at 350): There was no end in sight of the ban on processing of livestock for the appellant. The quantities of livestock and carcasses involved make it clear that, at the time, the appellant used the facilities of the particular abattoir to a considerable extent. It can be assumed that the appellant did not have livestock slaughtered at the abattoir in the pursuit of some macabre pleasure and that the carcasses of livestock assigned to be slaughtered were required for the purposes of the appellant's business. In these circumstances, it appears to me that, at the time the proceedings were instituted and the injunction first granted, the relevant conduct was plainly likely to cause substantial damage to the appellant's business as a wholesale and retail butcher regardless of which of the alternative meanings be given to the word "substantial" [about which Deane J expressed no view]. 64 In the circumstances of Tillmanns Butcheries, therefore, there was no necessity for the application of some lesser standard of proof than the ordinary civil standard. It is also clear that Deane J's analysis was connected with the idea that a particular purpose of causing loss or damage was the origin of the conduct in question. An assumption of the necessary quality is not available in a case involving s 50 of the Competition Act. An assumption that a business acquiring shares in the capital of another does so to assist its own commercial position does not go nearly the distance which is evident from the discussion in Tillmanns Butcheries. 65 Even the illustration provided by Deane J, of firing a rifle careless of any passer by, is accompanied by pejorative overtones which are simply not available in a case such as the present. 66 Finally, as I said earlier, no doubt was expressed about the soundness of the approach taken by Northrop J in Ansett Industries, so far as it concerned s 50, which had been decided in only the previous year. 67 The next important development was the apparent approval of the observations of Deane J in Tillmanns Butcheries, by the High Court in Sheen v Fields Pty Ltd (1984) 58 ALJR 93. The statutory context was quite different. The case concerned Rules made under the Factories and Shops Act 1960 (Qld). The report of the case states: Rule 1, cl. 21 of the Rules made under the Factories and Shops Act 1960 (Q.) required that protection in accordance with the Australian Standards for Industrial Eye Protection (CZ7-1967, Z7-1967 and Z45-1967) should be provided in cases where there was a "likelihood of injury" to the eyes of an employee in a factory. The Standards required that if a "hazard" existed, eye protectors should be issued, and that if the employee was exposed to flying particles, the provision of safety spectacles was the minimum acceptable method of protection. 68 The standard of care thereby identified clearly did not depend on a probability of injury before eye protection was obligatory. 69 Gibbs CJ (with whom Mason, Wilson and Dawson JJ agreed) accepted for that purpose that a "likelihood of injury" means "a real or not remote chance or possibility regardless of whether it is less or more than 50 percent" applying the observations of Deane J in Tillmanns Butcheries, but observed that this conclusion did "not decide the present case". Again, it was not actually necessary for a "real chance" test to be applied. Liability was denied by the High Court, even on that lesser test, because the injury to the employee was, in the particular circumstances of the case, not reasonably foreseeable. He was using an unauthorised work method. The authorised methods would not have involved risk of eye injury. His unauthorised method did, but he had chosen not to wear his safety goggles. 70 In my respectful view, despite the endorsement in that statutory context, of the "real chance" test, Sheen v Fields does not support its application to the operation of s 50 of the Competition Act. 71 Later that year (1984) a Full Court of this Court approved the application of the "real chance" test to the operation of s 52 of the TP Act (now Sched 2, s 18 of the Competition Act) (Global Sportsman Pty Ltd v Mirror Newspapers Pty Ltd (1984) 2 FCR 82 ("Global Sportsman")). The Court said (at 87): A contravention of s. 52(1) is established by conduct which is misleading or deceptive or which is likely to mislead or deceive. Conduct is likely to mislead or deceive if that is a "real or not remote chance or possibility regardless of whether it is less or more than fifty per cent": cf. Tillmanns Butcheries Pty Ltd v. Australasian Meat Industry Employees' Union (1979) 42 F.L.R. 331 at 346, per Deane J.; Sheen v. Fields Pty Ltd (1984) 58 A.L.J.R. 93. 72 However, no occasion arose to apply the test to actual facts. The proceedings involved a stated case. As the Court said (at 89): One of the disadvantages of the Stated Case procedure in this matter is that questions which the parties anticipate may arise must be considered in the abstract. 73 The question in the proceedings was identified in these terms (at 92): The basic dispute between the parties concerns the respondents' contention that the publication of statements including statements of opinion made in the ordinary course of the publication of news in those parts of a newspaper which are not advertising material cannot be conduct which is misleading or deceptive or likely to mislead or deceive within the meaning of s. 52(1) of the Act. For the reasons which we have stated, we are of opinion that the respondents' assertion is incorrect. (Emphasis in original) 74 In the circumstances, in my respectful view, the apparent endorsement of the "real chance" test cannot bear too much weight, as a result of its apparent approval in Global Sportsman, if a sufficient reason exists to revisit the question in a concrete setting, such as provided by the present case. 75 In Trade Practices Commission v TNT Management Pty Ltd (1985) 6 FCR 1 ("TNT") Franki J dealt, in a very substantial piece of litigation, with allegations of contravention of s 45 of the TP Act (also s 45 of the Competition Act, but now in different terms). That provision dealt with contracts, arrangements or understandings in restraint of trade or commerce. Such a contract, arrangement or understanding was not in restraint of trade or commerce unless "the restraint has or is likely to have a significant effect on competition". Franki J referred to Tillmanns Butcheries. His Honour said (at 49): … I consider that the word in s 45(2) now under consideration is to be read with due regard to the fact that it appears in a penal statute, that it is linked with the word "significant" and that this means that, whilst the meaning need not be restricted to a situation where the odds are greater rather than equally balanced or somewhat less than equally balanced, the probability must be something not very far short of "more probably than not", except in unusual circumstances as, for example, the situation mentioned by Deane J of firing a rifle through drawn curtains into a quiet lane in a country village. (Emphasis added) 76 With respect, this is not a completely satisfactory approach to the issue. Having regard to the provisions of the Evidence Act to which I earlier referred, the civil standard applies to the disposition of a cause of action unless it is possible to conclude that some other statute has effectively modified the position. I confess I do not understand exactly what Franki J was intending to convey in the passage I have emphasised, except that it seems a less than wholehearted embrace of the "real chance" test. It does seem clear that Franki J regarded the example offered by Deane J as an exceptional one. 77 Now I may turn to the cases offered by Miller as illustrations of the proposition that the "real chance" test appears to apply with respect to the operation of ss 45, 46 and 47 of the Competition Act (although perhaps not for s 50). 78 In Monroe Topple the trial judge had dismissed an application relying on ss 45, 46, 47 and 51AC of the TP Act. Section 45(2)(a)(ii) and s 45(2)(b)(ii) prohibited a corporation from making or giving effect to a provision of a contract which has the purpose or which had or is likely to have the effect of substantially lessening competition. A case for relief had not been established. An appeal failed. It is important to understand that the appeal failed when tested against the lesser standard of proof of "real chance". In the Full Court, on appeal, Heerey J (with whom Black CJ and Tamberlin J agreed) said (at [111]): 111 As to effect, it is to be noted that the section is also satisfied if the conduct is likely to have the effect of substantially lessening competition. This involves assessing the future effect of the conduct, considered as at the time it is engaged in: Trade Practices Commission v TNT Management Pty Ltd (1985) 6 FCR 1 at 50. "Likely" does not mean "more likely than not". It is sufficient that there is a real chance or possibility that a substantial lessening of competition will occur: Tillmanns Butcheries Pty Ltd v Australasian Meat Industry Employees' Union (1979) 42 FLR 331 at 346-348 per Deane J. Tillmanns was a case under s 45D of the Act, but the reasoning of Deane J is equally applicable to the concept of likely effect in s 47(10). (Emphasis in original) 79 No further discussion was necessary. The appeal failed on the lesser test expressed by Deane J in Tillmanns Butcheries. I intend no disrespect to the judges who constituted that Full Court by saying that the matter did not appear to receive (and did not need to receive) more than passing attention in the circumstances. 80 In Universal Music somewhat reserved endorsement was given to the statements of Heerey J in Monroe Topple. In Universal Music the Court said simply (at [247]): 247 … We are prepared to assume, for the purposes of argument, that likely does not mean more likely than not, but rather that there is a real chance or possibility that a substantial lessening will occur (Monroe Topple per Heerey J at [111]). 81 Seven Network is the judgment on appeal from Sackville J in the C7 litigation. The C7 litigation also concerned the operation of s 45(2)(a)(ii) and s 45(2)(b)(ii) of the TP Act. At [747] Dowsett and Lander JJ set out the passage from the judgment of Heerey J in Monroe Topple extracted above. Their Honours said (at [749]-[750]): 749 News's submissions on this question focus upon matters of policy. It is said that the Court will normally construe legislation having a penal aspect so as to favour the subject in the event of any ambiguity. It also submits that there may be considerable difficulties for a competitor in deciding whether or not there is a real chance of an anti-competitive outcome as a consequence of proposed conduct. Further, it is said that the basis upon which the cases have adopted the "real chance" test has been that the alternative approach would give the words little functional effect. This, News submits, is not a proper basis for construing the word "likely" as meaning "a real chance" in the context of s 45(2). 750 … That is not our view of the case. We … consider that we should follow the decision in applying s 45 unless we are satisfied that it is clearly wrong. Reconsideration of policy matters will not generally be an appropriate basis for such satisfaction, at least in the absence of any evidence as to wide-spread inconvenience or injustice caused by the established approach. The decision in Monroe Topple 122 FCR 110 has stood since 2002. News's submissions do not cause us to doubt its correctness. 82 Two observations may be made about the approach which their Honours took. First, it is clear that the statements were made in response to submissions which, as their Honours said, focussed "upon matters of policy". Secondly, it seems equally clear that their Honours were not invited to embark on a more detailed analysis to assess the persuasive force of the legal reasoning which, as I have endeavoured to show, is really derived from the judgment of Deane J in Tillmanns Butcheries in quite different circumstances. In the circumstances, while of course their Honours' conclusions are entitled to substantial weight, their endorsement of Monroe Topple does not add to the analysis in a way which makes it impermissible for it now to be revisited. 83 None of the cases to which I have so far referred (including Tillmanns Butcheries) provide, in my respectful view, sufficiently persuasive support for the application of the "real chance" test when assessing whether something is likely to happen. For my own part, I prefer the earlier approach by Bray CJ in Krieg and the never disapproved judgment of Northrop J in Ansett Industries. However, there are two cases which seem to command greater attention, owing to the deliberate and considered statements they contain. One concerns s 50 of the Competition Act and one does not. 84 In News Limited v Australian Rugby Football League Limited (1996) 64 FCR 410 (at 564, 565, 571) a Full Court explicitly endorsed Tillmanns Butcheries and referred to Global Sportsman, when considering the definition of the term "exclusionary provision" in s 4D of the TP Act and, in particular, when a person was deemed to be competitive with another within the meaning of that definition. The Court said (at 565): As Deane J pointed out in Tillmanns at 487, the phrase "would be likely to have" conveys a lower degree of likelihood than the phrase "would have". Similarly, the phrase "would be likely to be" in s 4D(2) conveys a lower degree of likelihood than the phrase "would be". Given that "likely" has the meaning attributed to it by Deane J when used within the expression "would be likely to be", it must also bear the same meaning when used in the expression "is likely to be". 85 In that case, the Court went on to actually apply the test it had stated (at 571). The ultimate result was that certain Commitment and Loyalty Agreements were found to be in breach of s 45(2)(a)(i) of the TP Act. The case involved the same statutory provision, therefore, as Seven Network. 86 So far as s 50 itself is concerned the strongest statement in support of the application of the approach expressed by Deane J in Tillmanns Butcheries is to be found in the judgment of French J (when his Honour was a judge of this Court) in Australian Gas Light Company v Australian Competition and Consumer Commission (2003) 137 FCR 317 ("AGL"). This was the authority applied by the trial judge. French J acknowledged (at [343]) that there had been "some divergence in the construction of 'likely' in various provisions of the Act". He also said (at [344]) "[d]ifferent views have been expressed from time to time in connection with s 50" in that respect. His Honour referred to some of those matters including a number of the cases to which I have referred. 87 Consideration of all those matters led to conclusions stated as follows (at [347]-[348]): 347 The collocation "would have the effect, or be likely to have the effect, of substantially lessening competition" appears in similar and identical versions in other provisions of Pt IV. It appears in ss 45, 45A, 45B, 45C, 47(10) and 50A. In my opinion that formulation is intended to have the same construction throughout Pt IV. Neither language nor policy mandates a variation in its construction from section to section. In any event as a matter of construction if "likely" simply meant more probable than not, it would be difficult to distinguish the application of that limb of the formula from the application of the first limb which, having regard to the onus of proof applicable in proceedings under Pt IV, could be established on the balance of probabilities. 348 The meaning of "likely" reflecting a "real chance or possibility" does not encompass a mere possibility. The word can offer no quantitative guidance but requires a qualitative judgment about the effects of an acquisition or proposed acquisition. The judgment it requires must not set the bar so high as effectively to expose acquiring corporations to a finding of contravention simply on the basis of possibilities, however plausible they may seem, generated by economic theory alone. On the other hand it must not set the bar so low as effectively to allow all acquisitions to proceed save those with the most obvious, direct and dramatic effects upon competition. By the language it adopts and the function thereby cast upon the Court and the regulator in their consideration of acquisitions s 50 gives effect to a kind of competition risk management policy. The application of that policy, reflected in judgments about the application of the section, must operate in the real world. The assessment of the risk or real chance of a substantial lessening of competition cannot rest upon speculation or theory. To borrow the words of the Tribunal in the Howard Smith case, the Court is concerned with "commercial likelihoods relevant to the proposed merger". The word "likely" has to be applied at a level which is commercially relevant or meaningful as must be the assessment of the substantial lessening of competition under consideration - Rural Press Ltd v Australian Competition and Consumer Commission (2003) 216 CLR 53 at [41]. 88 With respect, the conclusion that the "assessment of the risk or real chance of a substantial lessening of competition cannot rest upon speculation or theory" must be accepted as compelling. However it does not seem to me, with respect, to be accurate to postulate that the ordinary requirement to prove a case (including this element of a case) to the usual civil standard would "set the bar so low as effectively to allow all acquisitions to proceed save those with the most obvious, direct and dramatic effects upon competition". I also, speaking for myself, entertain considerable doubt about the notion that the Court is intended to have a function of applying "a kind of competition risk management policy". That may be a legitimate function for a regulator considering acquisitions under s 50, and making decisions whether proceedings are to be commenced to restrain them, but once those decisions are made and proceedings commenced I see no real alternative to the Court attempting to deal with them on the basis of identifiable legal standards which are sufficiently certain and well established to yield predictable outcomes. Asking whether there is a "real chance" of something occurring seems to me, with respect, to invite and endorse speculation and conjecture. The range of possibilities (greater than "mere possibility") existing below a more than equal chance or possibility seems potentially so extensive as to create great uncertainty. The problem cannot be overcome by employing the technique favoured by Franki J in TNT of suggesting that the probability "must be something not very far short of 'more probably than not', except in unusual circumstances". That approach betrays the potential uncertainty of departing from the usual civil standard. 89 I accept the persuasive force, and the authoritative source, of the judicial pronouncements (particularly those of Deane J in Tillmanns Butcheries and French J in AGL where the matter has received closest attention and analysis) in support of the "real chance" approach. If I were sitting at first instance I would feel obliged, as did the trial judge in the present case, to give effect to that approach. As a member of a Full Court, however, I feel at liberty to express reservations about the matter, which I have endeavoured to do in the preceding discussion. In my view, the "real chance" test should not be applied to s 50 of the Act. 90 If it is appropriate to apply that test at all, I agree with the approach taken by the trial judge. Moreover, in the circumstances of the present case, even adopting the full scope of the approach urged by the ACCC would not have changed my view about the outcome of the appeal. Even if I had been persuaded that the trial judge had erred in applying a requirement of proof on the balance of probabilities to the establishment of the facts constituting the so-called "counterfactual" (which I am not), testing the matter on the basis of whether it should have been concluded that there was a "real chance" of the outcome suggested by the ACCC would have made no difference. The trial judge decisively rejected that suggested outcome as involving any "real chance". He said, plainly, that he had come to that view applying the "real chance" test to that issue, in case it was necessary to do so. Contrary to the submissions advanced for the ACCC the analysis which preceded that statement is not confined by the terms in which it is expressed to the application of a test on the balance of probabilities. The trial judge said repeatedly that it was unlikely that various elements of the scenario suggested by the ACCC would occur. It follows necessarily that he did not accept that it was likely that they would occur, whether as a "real chance" or otherwise, nor that the overall hypothesis had a "real chance" of recurring.

Conclusion 91 For the reasons I have given I agree that the appeal should be dismissed. I certify that the preceding ninety (90) numbered paragraphs are a true copy of the Reasons for Judgment herein of the Honourable Justice Buchanan.

Associate: Dated: 30 November 2011 IN THE FEDERAL COURT OF AUSTRALIA NEW SOUTH WALES DISTRICT REGISTRY GENERAL DIVISION NSD 1533 of 2011

BETWEEN: AUSTRALIAN COMPETITION AND CONSUMER COMMISSION

JUDGES: FINN, Buchanan, Yates JJ DATE: 30 NOVEMBER 2011 PLACE: SYDNEY

YATES J Introduction [92] Background [106] Overall distribution arrangements [108] Independent retailers [111] Metcash and the IGA banner group [118] Franklins [130] SPAR [143] The Commission's counterfactual case [148] The future with the share acquisition by Metcash [155] The future without the share acquisition by Metcash [158] The Pick n Pay store sale process [158] The TMT Consortium offer [160] The KKKL consortium [177] The Commission's grounds of appeal [211] The parties' submissions on appeal [215] Consideration [226] The commission's case on market definition [238] The process and purpose of market definition [244] The primary judge's findings on constraints operating on Metcash [269] The perception of competitive constraints [272] The major supermarket chains as a constraint [278] Threat of sale as a constraint [284] The threat of constraint by Woolworths as a wholesaler [286] Franklins as a constraint [288] The primary judge's conclusions on market definition [298] The Commission's grounds of appeal [304] Consideration [310] The hypothetical monopolist test [310] The product dimension of the market [331] The functional dimension of the market [336] Failure to consider functionally separate markets [339] The pricing discretion of independent retailers [344] Metcash's pricing discretion - Project Energise - Project Ling Chi [362] Store sales as a constraint [376] Conclusion on functional dimension [378] Conclusion [379] Effect on competition [386] Disposition [392]

introduction 92 The Australian Competition and Consumer Commission (the Commission) appeals from a judgment dismissing its application for injunctive relief in relation to the acquisition by Metcash Trading Limited (Metcash) of shares in Interfrank Group Holdings Pty Limited (Franklins) from Pick n Pay Retailers (Pty) Limited (Pick n Pay). 93 On 1 July 2010 Metcash entered into an agreement with Pick n Pay to acquire all the issued shares in the capital of Franklins for $215 million. On 8 December 2010 the Commission commenced its proceeding in this Court seeking, in effect, to restrain the share acquisition from proceeding to completion. 94 The Commission's case was that the effect or likely effect of the share acquisition would be to substantially lessen competition in what it defined as the Independent Wholesale Grocery Market (the defined market), in contravention of s 50(1) of the Competition and Consumer Act 2010 (Cth) (the Act). 95 This market was defined as a market in New South Wales and the Australian Capital Territory (the defined territory) in which: (a) the suppliers were Metcash, Franklins and SPAR Australia Limited (SPAR); (b) the acquirers were Independent Supermarket Retailers, as defined by the Commission; and (c) the products were a range of Wholesale Packaged Groceries, as defined by the Commission. 96 I will return to describe aspects of the definition of this market. For present purposes it is sufficient to note that this market was one that was functionally delineated as a wholesale market. 97 The primary judge was not persuaded that there was a separate market for the wholesale supply of packaged groceries to independent supermarket retailers, as defined by the Commission. It followed, on his Honour's reasoning, that the Commission's application must fail on its pleaded case. 98 Nevertheless, the primary judge proceeded to consider the Commission's so-called "counterfactual" case as to why Metcash's share acquisition would have the effect, or would be likely to have the effect, of substantially lessening competition in the defined market. Ultimately, however, his Honour rejected that case. His Honour was not persuaded that there was a real chance, let alone a case on the balance of probabilities, that the counterfactuals for which the Commission contended would come to pass. 99 The primary judge regarded the Commission's counterfactual case as one based on possibilities that were no more than speculation. It followed that his Honour was not persuaded that, if the acquisition proceeded to completion, there would be, or would be a real chance of, a substantial lessening of competition in the defined market. Indeed, his Honour considered that it was quite likely that the share acquisition would strengthen the capacity of independent retailers operating under Metcash's IGA banner to compete more vigorously with the major supermarket chains. 100 In the end result, the primary judge concluded that the share acquisition would not contravene s 50(1) of the Act. 101 The Commission's case on appeal was that the primary judge erred in his application of economic and legal principles in considering the question of market definition: his Honour should have found that the relevant market to assess the potential competitive harm identified by the Commission was a wholesale market for the supply of packaged groceries to independent retailers in the defined territory. 102 Moreover, the Commission contended that the primary judge erred in his consideration of the Commission's counterfactual case by adopting and applying a standard that was too high for the purposes of assessing whether the share acquisition was likely to lead to a substantial lessening of competition. 103 The Commission also contended that the primary judge failed to apply the factors listed in s 50(3) of the Act to determine whether the acquisition would be likely to lead to a substantial lessening of competition in the defined market. 104 After the primary judge gave judgment, the share acquisition proceeded to completion on 30 September 2011. 105 On appeal, the Commission sought a declaration of contravention and stated that, if successful, it intended to commence divestiture proceedings.

Background 106 In order to understand the Commission's case and the primary judge's rejection of it, it is necessary to have some appreciation of the chains of distribution of products that come to be supplied, in the defined territory, as grocery items to the consuming public. 107 The primary judge gave a detailed explanation of these matters based on evidence that, for the greater part, was not controversial. It is not necessary to repeat the details of that explanation. It is sufficient to note the following matters, which I have taken from the primary judge's more detailed description.

Overall distribution arrangements 108 Manufacturers and primary suppliers (who, generally speaking, are either nationally-based or State-based) supply products to self-supplying supermarket chains (relevantly, Woolworths, Coles, Aldi and Franklins), or wholesalers who in turn supply independent retailers. Metcash is one of those wholesalers. It is Australia's largest grocery wholesale distribution and marketing company. In circumstances which I will explain, Franklins also operated as a wholesaler and supplied independent retailers, notwithstanding that it also operated a self-supplying supermarket chain. 109 The self-supplying supermarket chains and independent retailers supply grocery items to consumers. The self-supplying supermarket chains undertake several stages in that process. They negotiate with and acquire products from manufacturers and primary suppliers, including fresh produce from farms. They deliver the products to their retail stores, price those products, and sell them to consumers. On the other hand, independent retailers acquire their products from wholesalers or, less often, directly from manufacturers and primary suppliers. They are, generally speaking, responsible for pricing the products and selling them to consumers. 110 Wholesalers negotiate with and acquire products from manufacturers and primary suppliers, including fresh produce from farms, and supply these products to independent retailers. Sometimes the wholesalers organise the delivery of these goods to the independent retailer; sometimes the independent retailer makes its own arrangements for delivery.

Independent retailers 111 Independent retailers operate under different formats. Stores with a floor space greater than 1,200 square metres are commonly described in the trade as "supermarkets", although this description can apply to small stores. At the other end of the spectrum, stores with a floor space of less than 350 square metres are commonly described as "convenience stores". Stores that lie in between and are neither supermarkets nor convenience stores are often described as "top-up stores". As a general observation, the smaller the store, the more limited is the product range on offer. 112 Stores of a similar size and character, and which are supplied by the same wholesaler, usually operate under a common "banner", and thus are part of a "banner group". Members of banner groups adopt a common public brand, reflected in a store's trading name, get-up and appearance. Such stores are generally offered a range of retail services by their wholesaler. They are encouraged to follow common stocking and discount policies. They have access to shared funds for promotional programs. 113 The primary judge noted that this approach to retailing simulates, to a degree, the pattern of business employed successfully by self-supplying supermarket chains, while retaining independence of action for the individual retailer, particularly on pricing. 114 Nevertheless, operating under a banner does require the individual retailer to accept a degree of subordination to the larger commercial strategy of the banner group, so as to present a common face to the consuming public, reflected in similar stocking policies, similar pricing and common promotions. 115 The primary judge noted that the existence of banner groups, where the wholesaler controls the banner name and group members have an obligation to conform to certain styles and standards, is central to a wholesaler's commercial performance and prospects: if "banner discipline" is poor, there will be a loss of competitiveness. 116 The primary judge noted that competition amongst all retailers of grocery items is reflected in both price and non-price factors. The non-price factors include location, range of merchandise, store layout and presentation, check-out facilities, hours of trading, and personal service. 117 The primary judge found (at [12]) as follows: The grocery industry in Australia is a highly competitive industry characterised by high volumes and low margins. The operators of self-supplying supermarket chains are extremely disciplined and endeavour to standardise their offerings at any one time. Nevertheless, those offerings are not static, but shift as the chain operators explore market opportunities and develop new strategies. On the other hand, the products offered by independent retailers exhibit greater diversity than those of the chains in areas such as, for instance, size and location. Independent retailers may choose, or be forced, to rely upon factors other than price to attract customers.

Metcash and the IGA banner group 118 Metcash operates, Australia-wide, several business units or divisions, including the IGA-Distribution division and the IGA-Fresh division. IGA-Distribution is the largest division. It has seven distribution centres in New South Wales. Metcash supplies various grocery products to independent retailers of various sizes, ranging from convenience stores to large format supermarket stores. These retailers are independent of the self-supplying supermarket chains. 119 The IGA banner is Metcash's "public face". Metcash provides branding and other support services, as well as grocery products, to independent retailers comprising the IGA banner group. 120 The IGA banner group has three tiers: Supa IGA, IGA and IGA X-press. 121 Supa IGA stores are supermarket stores with floor sizes from about 1,200 square metres. These stores account for about 40% of sales by the IGA-Distribution division. 122 IGA stores are top-up stores, with floor sizes between about 350 square metres and 1,200 square metres. These stores account for about 30% of sales by the IGA-Distribution division. 123 IGA X-press stores are convenience stores, with floor sizes up to about 350 square metres. These stores account for about 2 - 3% of sales by the IGA-Distribution division. 124 Not all independent retailers supplied by Metcash are members of the IGA banner group. Moreover, some independent retailers supplied by Metcash do not obtain all their grocery products from Metcash. 125 Metcash has no control over the prices that independent retailers charge consumers, save in relation to promotions where a "ceiling price" may be fixed. Thus, in general terms, the retailers it supplies are free to set their own retail prices. 126 Nevertheless, given that independent retailers must compete with major supermarket chains in their vicinities, particularly Woolworths and Coles stores, Metcash encourages the independent retailers it supplies to "benchmark" their standard shelf prices according to the standard shelf prices charged by Woolworths or Coles, particularly in relation to "key value items" that represent particular products whose prices are known and used by consumers to assess the value on offer from a particular store. 127 The primary judge found (at [35]) that competing with Woolworths and Coles at the retail level is essential for the success of the independent grocery network as a whole. This led Metcash to establish a retail pricing service for independent retailers, called the Mix & Match system, by which Metcash carries out the task of gathering and monitoring standard shelf prices for between 1,200 and 1,500 grocery items every week (and for other items less frequently) in Woolworths and Coles stores. The derived data is allocated to categories of products, such as breakfast cereals, canned vegetables and baby products, and then used to generate different pricing zones. As Woolworths has been the traditional price leader, its prices provide the benchmark for independent grocery pricing, although Metcash checks against Coles' prices as well. The information Metcash provides in its Mix & Match system then enables the retailer to select how far above or below Woolworths' standard shelf prices the retailer wishes to price its products. 128 The primary judge described the choices available to retailers using the Mix & Match system as follows: 37 Of the pricing zones generated by Metcash, the zone corresponding with standard shelf prices at the relevant Coles or Woolworths store is regarded as a base zone. Independent retailers select different pricing zones for different categories of goods, and communicate their selections to Metcash. In the Mix & Match System, zone 60 is generally equivalent to Woolworths' standard shelf prices, subject to minimum gross profit requirements. Therefore, if an independent retailer wants to charge the same as Woolworths for a particular category of products, it selects zone 60 for that category. If an independent retailer wants to set its prices above or below Woolworths for a particular category of products, it selects a zone other than zone 60. For example, zone 76 is equal to zone 60 plus one per cent, so a retailer using zone 76 pricing will achieve an additional one per cent gross profit compared to a retailer using zone 60. The retailer can choose pricing zones across its entire range of grocery products, or for a particular category of grocery products. Once the retailer's price selections are communicated, Metcash generates a unique price file for the retailer, according to the retailer's selection, and provides that price file to the retailer electronically each week. The price file can be entered directly into the retailer's pricing system. 38 Of the various available pricing zones, there are five zones with gross profit percentage below zone 60 and seventeen zones with gross profit percentage above zone 60. Thus, there are altogether 23 zones from which to choose. Each individual retailer decides which zone or zones it will operate on, whether that be zone 60, a zone above or below zone 60, or some combination of zones. The zone chosen might depend upon the competition faced by the store, the banner under which the store operates, where the store sits in the marketplace, and the gross profit that the retailer seeks to achieve. 129 The discretion available to independent retailers to price above "zone 60" prices was an important aspect of the Commission's case as it related to the question of market definition. It is a matter to which I will return.

Franklins 130 The Franklins grocery business, under Pick n Pay's economic ownership, was carried on in New South Wales. It involved both wholesale and retail activities. 131 Franklins operated 80 retail stores in its own right (the Franklins corporate stores) and franchised a further 10 stores that also operated under the Franklins banner (the Franklins franchise stores). It operated as a self-supplying supermarket chain in respect of the corporate stores and as a wholesaler in relation to the franchise stores. In each case the products it supplied included packaged groceries, health, beauty and cosmetic products, general merchandise and fresh produce. 132 Pick n Pay is a South African corporation. It is a member of the Pick n Pay group of companies. At the time of the trial, this group consisted of a network of 888 retail stores in South Africa and other countries (including Australia). The ultimate holding company, Pick n Pay Stores Limited, is listed on the Johannesburg Stock Exchange. 133 Pick n Pay acquired its interest in Franklins in 2001. Its initial investment was approximately $133.7 million. Over the years its investment had grown to approximately $289.4 million. However, Franklins had made losses in seven of the 10 years that it had operated under Pick n Pay's ownership. As at 31 August 2010, Pick n Pay had accumulated losses of $105.1 million from its original investment. There is evidence to the effect that, as at the end of October 2010, Franklins was losing $2 million a month and that, were it not for Pick n Pay's support, Franklins would have been in voluntary administration. There is also evidence to the effect that Franklins' distribution centre was operating at only 60% capacity and that, in order to achieve scale efficiency, it required "another 20 good stores". 134 Franklins had not always carried on self-supplying or wholesaling activities. Those activities were of significantly more recent origin. 135 On 14 September 2001 Franklins entered into an agreement with Metcash for the wholesale supply of groceries to Franklins supermarkets. Under this agreement Metcash provided warehouse services to Franklins for a price and agreed to maintain specified levels of service. It also agreed to pass on to Franklins all discounts, rebates and allowances associated with the products that Franklins purchased from it. 136 However, in late 2002, Franklins became aware that Metcash was not passing on certain discounts, allowances and rebates, in accordance with the agreement. This experience led to an assessment by Franklins that wholesale supply by Metcash did not leave it with sufficient margin. 137 In April 2003 Franklins began exploring alternatives to supply by Metcash, including self-supply. 138 In early 2004 Franklins decided to invest in wholesale operations for the establishment of a self-supply function for Franklins. It thereafter terminated its agreement with Metcash and commenced self-supply operations in early 2005. 139 Once Franklins had put in place its self-supply arrangements, it turned to consider whether it would expand and improve the performance of its business through franchising. 140 The evidence shows that Metcash reacted to this development. In this connection the Commission relied on the efforts of Metcash to protect its business by activities under the name Project Energise. The Commission also relied on Project Ling Chi (also dubbed Project Death By A Thousand Cuts), which was a response by Metcash to an announcement by Franklins that it would be expanding its wholesale operations. These activities by Metcash were also an important aspect of the Commission's case as it related to the question of market definition. Once again, they are matters to which I will return. 141 The evidence at trial was that, despite all of Franklins' activities over more than five years to promote and attract independent supermarket operators to join its franchise system, it had only attracted five franchisees operating 10 franchise stores. Two of those stores were formerly Franklins corporate stores that had been purchased by Franklins managers and converted into Franklins franchise stores. Furthermore, in order to induce one of the five franchisees to convert from the IGA banner group, Franklins had to sell two of its better performing corporate stores. 142 Apart from its "Franklins" and "No Frills" brands, and assets that were specific to its retail activities (leaseholds, and management and franchise agreements) Franklins' business assets included agreements with various information technology licensors, logistics providers, manufacturers and other suppliers (under which Franklins acquired groceries and supplied them to the 90 corporate and franchise stores) and various intangible assets, including pricing systems and expertise in promotions and advice, which assisted Franklins in acquiring groceries and supplying them to retail stores.

SPAR 143 SPAR's principal activities are the procurement, storage and distribution of fast moving consumer goods, liquor warehousing, and the provision of marketing and retail support services for SPAR bannered franchisee stores, retailer training programs, and retail property development. 144 As at December 2010, SPAR supplied 153 stores in Queensland, 65 stores in New South Wales, 14 stores in the Australian Capital Territory, and six stores in the Northern Territory. 145 SPAR has a distribution centre in Acacia Ridge, Brisbane. It supplies around 13,000 lines of fast moving consumer goods. It does not have the infrastructure or the facilities necessary to operate a distribution centre for perishable products such as fruit and vegetables. 146 Some of the stores that SPAR supplies in New South Wales and the Australian Capital Territory only buy products every two or three weeks on an ad hoc basis. Stores north of Port Macquarie are supplied by SPAR directly from the Acacia Ridge distribution centre. The retailers operating those stores organise for the pick-up and delivery of goods from the distribution centre to the stores. They pay the cost of freight to those stores. 147 SPAR supplies stores located south of Port Macquarie from cross-docking depots in Sydney and Canberra. These depots allow for a large delivery truck to unload goods transported from Brisbane. Retailers obtaining supply through the cross-docking depots are required to organise the delivery of the products they purchase and bear the costs of freight between the cross-docking depot and their stores.

The Commission's counterfactual case 148 The Commission's case at trial proceeded on the basis that, in order to determine whether it could be said that the share acquisition would be likely to have the effect of substantially lessening competition in the defined market, it was necessary to consider the future state of that market "with" and "without" the acquisition by Metcash. This approach was in accordance with authority: Dandy Power Equipment Pty Ltd v Mercury Marine Pty Ltd (1982) 64 FLR 238 at 259-260; Outboard Marine Australia Pty Ltd v Hecar Investments (No 6) Pty Ltd (1982) 66 FLR 120 at 124; Australian Gas Light Company v Australian Competition and Consumer Commission (2003) 137 FCR 317 at [352]. In the present case, the future "without" the acquisition was described as the "counterfactual" case. So understood, the counterfactual case was an essential element of the calculus employed to detect and evaluate future change in the process of competition. The correct identification of the relevant market was clearly central to that determination. 149 Although the correctness of the primary judge's conclusions with respect to the defined market were strenuously challenged on this appeal, it is convenient, nevertheless, to deal, firstly, with the Commission's counterfactual case before dealing with the issue of market definition. This can be done because the primary judge dealt with the Commission's case in this respect on the assumption that the Commission had correctly defined the relevant market, and because, although pleaded in various ways, the Commission's counterfactual case came to be distilled in the proposition that it was likely, in the sense of a real chance, that, absent the acquisition by Metcash, a particular consortium of buyers, described in the appeal as the KKKL consortium, would be an alternative acquirer of Franklins' assets, or at least a substantial majority of those assets, and would be able to establish a wholesaling operation to independent retailers, in competition with Metcash. 150 The acceptance of that proposition, as a matter of likely fact, was critical to the acceptance of the Commission's case that the share acquisition by Metcash would be likely to have the effect of substantially lessening competition. Without establishing that likelihood, the Commission's case on contravention could not succeed. This is because, given the loss-making state of the Franklins business, Pick n Pay had determined that it would quit its Australian grocery retailing operations on the best possible basis, even if Metcash's share acquisition did not proceed to completion. The primary judged noted (at [344]) that it was common ground that Pick n Pay would sell the Franklins business and assets to the acquirer or acquirers who made the best offers and had the capacity to complete. It did not follow, however, that, absent Metcash's share acquisition, Franklins' wholesaling assets would be taken up by another, such that wholesaling activities deploying those assets would continue to be carried on, either as before or at all. 151 The key to carrying on a grocery wholesaling business is the ability to achieve scale efficiency in those operations by servicing a sufficient number of retailers whose commercial support would make those operations viable. The parties were in dispute as to the dollar amount of wholesale sales that would be necessary to provide such scale efficiency in the defined market. I have, however, already referred to the fact that there was evidence that, in supplying the 90 corporate and franchise stores, the Franklins distribution centre was only operating at 60% capacity and needed "another 20 good stores" to achieve scale efficiency. I have also referred to the losses that Franklins had suffered. 152 It was a key element in the Commission's counterfactual case that a significant majority of the 80 Franklins corporate stores, by sales revenue, would be acquired by a third party or third parties who would, in those circumstances, be likely to acquire and continue to operate the Franklins wholesale assets. No other case was pleaded. 153 In the absence of a real chance of someone other than Metcash taking up the Franklins wholesale assets and continuing Franklins wholesaling activities, the field of grocery wholesaling in the defined territory would be left, substantially, and by default, to Metcash simply by reason of Pick n Pay's decision to cease its Australian operations. In that counterfactual world, it could not be said that the share acquisition by Metcash would, itself, be likely to substantially lessen competition in the defined market and thus engage the prohibition of s 50(1) of the Act. 154 It is important to note here that, during the course of the trial, the Commission abandoned its original contention that SPAR was a viable alternative acquirer of the Franklins wholesale assets. The Commission was thus left with a case that rested on an acceptance that the KKKL consortium was a credible alternative purchaser. No other market entrant, and no other potential acquirer of the Franklins wholesale assets, was propounded.

The future with the share acquisition by Metcash 155 In his reasons for judgment the primary judge noted (at [346]) the Commission's contentions that, if Metcash's share acquisition proceeded to completion, Metcash would: (a) close-down Franklins' wholesale supply arrangements; (b) sell most or all of the Franklins corporate stores to independent retailers, and (c) require those retailers to enter into supply arrangements with it that would prevent the retailers, for a significant period, from obtaining supply from an alternative wholesaler. 156 The Commission contended that, if those matters came to pass, there would not be enough wholesale volume to support the establishment of a competing wholesaler to offer packaged groceries to independent retailers in the defined territory. 157 These contentions do not appear to have been contested, ultimately, by the respondents, as reflecting the likely state of the market should the Metcash share acquisition proceed to completion. These contentions were certainly not disputed by the respondents during the course of the appeal.

The future without the share acquisition by Metcash

The Pick n Pay store sale process 158 In October 2010 Pick n Pay implemented a process for the sale of the Franklins corporate stores. Despite some criticisms made by the Commission about the credibility of that process, the primary judge found (at [351]) that this process was implemented by Pick n Pay as an alternative means of disposing of the Franklins assets should the share acquisition by Metcash not proceed to completion. There was no appeal from that finding. 159 The primary judge found (at [351]) that the results of that process provided meaningful information relevant to the assessment of what would be likely to happen should Metcash's acquisition be prohibited. In the course of that process, prospective purchasers were required to make a non-binding indicative offer, nominating a price for each store intended to be purchased. As a result of that process, 43 parties made non-binding indicative offers for one or more of the Franklins corporate stores. These included offers made by Woolworths, Coles, Metcash, Franklins staff and independent retailers, including those who owned stores operating under the IGA banner. There were a number of overlapping offers. For this reason, it was not possible to determine the total financial return that Pick n Pay might expect under such a process. The primary judge noted, however, that the process attracted considerable interest. I will return to some aspects of the evidence dealing with the offers that were made. It was common ground that, if Metcash was prohibited from proceeding with its share acquisition, and if Pick n Pay decided to sell the Franklins assets by a store sale process, a new sale process would have to be undertaken.

The TMT Consortium offer 160 Quite apart from the Pick n Pay store sale process, on 4 November 2010 TMT Partners Consortium Pty Limited (TMT Consortium) made a non-binding indicative offer (the TMT Consortium offer) to acquire the Franklins corporate stores, the "Franklins" brand, the "No Frills" brand, and the rights and obligations relating to the Franklins distribution centre, for the sum of $110 million. It is necessary to record some of the primary judge's salient findings dealing with the background to the making of this offer. 161 The moving party behind the making of this offer was Mr Theo Koundouris, a director of Supabarn Supermarkets Pty Limited (Supabarn). This company is part of a family business, called the Koundouris Group, that specialises in property development, construction and property management in Canberra and Sydney. The primary judge described the Koundouris Group as having a wealth of experience in retailing and property. Mr Koundouris had worked in the retail supermarket industry since 1991 when the first Supabarn supermarket was acquired in Canberra. Supabarn operates a number of supermarkets under the Supabarn banner in New South Wales and the Australian Capital Territory. Supabarn obtains many of its packaged groceries by wholesale from Metcash. It also obtains supply from other sources, where it is possible and cost effective to do so. 162 In August 2010 Mr Koundouris sought advice in connection with the possible acquisition of some or all of the Franklins assets by a proposed consortium of independent supermarket operators led by Supabarn. In this appeal the consortium was called the KKKL consortium. This led to the preparation of a business plan which proposed, as its primary objective, that the consortium become the leading group of independent supermarket operators in New South Wales under a strategic relationship with a dedicated warehouse. Consortium members would be expected to enter into mutually binding agreements relating to wholesale supply, marketing, branding, and trading terms. The project was called Project Vertigo. 163 Following the advertising by Pick n Pay of its store sale process, contact was made by Mr Koundouris' representative, TMT Partners Pty Limited (TMT), with Pick n Pay's representatives handling that process. TMT, whose main representative in these events was Mr Gary Lowrey, made clear that it was acting on behalf of a group of investors who were interested in the store sale process and the Franklins distribution assets. 164 In his reasons for judgment, the primary judge detailed the various discussions and meetings in the period up to 4 November 2010 (when the TMT Consortium offer was made) that were held between TMT (principally, Mr Lowrey) and those acting on behalf of Pick n Pay, including its solicitors (principally, Mr Mark Stanbridge of Blake Dawson). 165 The primary judge also detailed the evidence of a limited number of discussions involving Mr Koundouris and Messrs Andrew and Vasilli Karellas in relation to the proposal. The Messrs Karellas are brothers and, with their parents, directors of Karellas Investments Pty Limited, which owns and operates supermarkets at Cremorne, Blaxland and Pyrmont in New South Wales as Supa IGA supermarkets. 166 The primary judge noted that, apart from these discussions, TMT and Mr Lowrey did not have dealings with anybody on behalf of the proposed consortium other than Mr Koundouris, from whom all instructions in relation to Project Vertigo came. 167 The primary judge dealt with the course of events following the making of the TMT Consortium offer on 4 November 2010, and made findings to the following effect. 168 By letter to TMT dated 12 November 2010, Blake Dawson referred to an earlier meeting at which it had been made clear that Pick n Pay required certain further information in order to assess the merits of the TMT Consortium offer. The letter noted, however, that there had been a failure to provide any of that information. 169 When TMT Consortium responded on 17 November 2010, it did so in terms which indicated that it had a strong interest in the assets and business activities detailed in its offer, but that the form of any such acquisition (that is, whether by asset acquisition or share acquisition) could only be determined with the benefit of due diligence and an understanding of what warranties and indemnities might be provided by Pick n Pay. TMT Consortium suggested that any decision by Pick n Pay in respect of its offer could not be made until after the Commission's attitude to Metcash's acquisition was known. 170 Mr Stanbridge told Mr Lowrey on 18 November 2010 that it was imperative that Pick n Pay understand (a) who the consortium members were; (b) how the consortium proposed to deal with landlords; and (c) how the consortium proposed to fund the offer and associated financing requirements. Mr Lowrey responded that the consortium could not take the risk of identifying its members while Metcash remained a likely purchaser. 171 Following the conversation between Mr Stanbridge and Mr Lowrey, Mr Lowrey suggested to Mr Koundouris that financial information should be provided to Pick n Pay to demonstrate the consortium's bona fides. Mr Koundouris then made arrangements to have a meeting with his bank the following week. Mr Lowrey informed Mr Stanbridge that arrangements were being made with the consortium's bankers and that he hoped to have something available in the following week. 172 Blake Dawson, through Mr Stanbridge, wrote to Mr Lowrey on 19 November 2010 indicating that Pick n Pay needed to be persuaded that the TMT Consortium offer would provide a more certain outcome in a shorter time frame than was likely under the store sale process. The letter stated that the information that Pick n Pay had requested, but TMT Consortium had refused to supply, was directly relevant to timing and certainty of execution. The letter said that it was essential the consortium provide details of (a) its ability to complete the acquisition; (b) its capacity to fund the purchase and the replacement of bank guarantees in the order of $30 million; (c) the sources of its funding; (d) the identity of the members of the consortium; and (e) the consortium's position in relation to each of the key terms set out in the store sale process documents. The letter also sought other information and concluded by saying that, if TMT Consortium wished to be considered by Pick n Pay as a potential purchaser, it would need to supply the requested information by no later than close of business on 20 November 2010. 173 Mr Lowrey responded to this letter on 20 November 2010. He said that the anonymity of the key consortium members was a requirement of the TMT Consortium offer. He said that, while a significant amount of preparatory work had been completed, the consortium had been reluctant to incur large costs when Pick n Pay was focused on completing with Metcash. He said that if that position changed, the consortium was ready to move quickly. He said that the responses required in relation to financial capacity could be delivered on Monday 29 November 2010. 174 Blake Dawson replied by letter on 23 November 2010. The letter noted that TMT Consortium had not afforded Pick n Pay the opportunity to consider the merits of its offer, and that the timely supply of the additional information was critical to Pick n Pay's deliberations. The letter ended by saying that, because of the lack of information regarding the TMT Consortium's offer, including the identity of the members of the consortium, funding sources, and the approach to completion risks and landlords' consents, Pick n Pay was unable to place any meaningful weight on that offer. 175 Following receipt of that letter, Mr Lowrey and Mr Koundouris discussed the position and reached a consensus that, at that time, there was nothing else that they could do. 176 None of this evidence was challenged on appeal. It represents the state and extent of the TMT Consortium offer.

The KKKL consortium 177 I have referred to the fact that, on this appeal, the consortium was referred to as the KKKL consortium. This is because, on the Commission's case, the consortium was comprised of interests representing the Koundouris, Krnc, Karellas and Lionis families. 178 I have briefly described the business interests of the Koundouris and Karellas families. It is necessary to say something briefly about the business interests of the Krnc and Lionis families. 179 Mr John Krnc has been involved in the supermarket business in the Australian Capital Territory for about 27 years. During that time he has owned and operated several supermarkets with his brothers Steven, Tony and Robert. The four Krnc brothers also operate stand-alone liquor stores in the Australian Capital Territory, through a company called Supergrocer Pty Limited. At the relevant time Mr John Krnc was on the State board of IGA-Distribution for New South Wales and the Australian Capital Territory. He was also a director of Independent Liquor Retailers Pty Limited, which operates stores under the Local Liquor banner. He was also a director of Direct Fruit Distribution Pty Limited, which operates a warehouse at Fyshwick in the Australian Capital Territory, that supplies fresh produce to local retailers. 180 The Lionis family operates four Franklins supermarkets located at Springwood, Newtown, South Hurstville and Cronulla in New South Wales, through a company called Benzat Holdings Pty Limited (Benzat). Mr Peter Lionis and his father, Michael, are directors of that company. Michael Lionis has been involved in the grocery industry for over 30 years, originally operating a supermarket in Canberra before moving to New South Wales. Peter Lionis has been involved in grocery retailing for approximately 19 years, dealing with suppliers, wholesalers and other retailers. Peter and Michael Lionis manage the four Franklins supermarkets. 181 Benzat operated the Springwood supermarket under the Supa IGA banner and the Newtown supermarket under the IGA banner until 2007, when the stores converted to become Franklins franchise stores. As part of these arrangements, Benzat acquired two existing Franklins supermarkets that served South Hurstville and Cronulla. 182 The primary judge (at [385]) summarised the Commission's contentions with respect to the KKKL consortium, as follows: The Commission contends that the Koundouris, Krnc, Karellas and Lionis interests are operated by persons with considerable experience and expertise in the retail grocery business, and who have formulated and continue to formulate plans for the acquisition of Franklins. It says that the Koundouris, Krnc, Karellas and Lionis interests are operated by astute business people who are ready, willing and able to acquire the Franklins business and assets, subject to acquiring the information that would be made available as part of an ordinary due diligence process. The Commission says that there is persuasive evidence that the proposed consortium consisting of those interests intends to acquire the Franklins assets and operate a business supplying independent supermarket retailers in NSW and the ACT in competition with Metcash. It says that the consortium members remain interested, despite the paucity of information provided to them and the apparent attempts by Pick n Pay to dissuade them from showing interest, which attempts include, the Commission says, the misrepresentation of the actual losses being incurred by Franklins. The Commission contends, therefore, that, if an offer were made by the consortium, there is a real chance that Pick n Pay would accept it, because an offer for the whole business would avoid substantial shutdown costs and would offer a faster and cleaner exit than a store by store sale process. 183 At trial the Commission advanced a number of supporting contentions which the primary judge described in some detail at [386] to [395]. It is sufficient to note, for present purposes, the following contentions: (a) Pick n Pay would most likely sell Franklins as a going concern because a store by store sale process would face numerous significant hurdles that render it unlikely as the means by which Pick n Pay would ultimately recover its capital investment in Franklins (including the prospect that the store sale process would face similar competition issues to those that affected the Metcash share acquisition). (b) In this connection, there is every reason to believe that an acquisition by Woolworths or Coles of an individual store that prevents competition in the defined market would itself be prohibited by s 50(1) of the Act. (c) This contention was extended to Ritchies Stores Pty Limited (Ritchies), an independent retailer in which Metcash had acquired a 26% interest. Under the store sale process, Ritchies had made a bid of $90.1 million, exclusive of stock, for 21 stores. (d) Thus a prediction about who would be the highest offeror for the Franklins assets must be premised on the particular offeror making the highest lawful offer. (e) In any event, the outcome of an individual store sale process is not so certain that the Court would find that an offer by the KKKL consortium did not have a real chance of success. (f) Even if Pick n Pay were to pursue a store by store by store sale process, each of the members of the KKKL consortium would be a credible offeror for stores and other assets under that process. (g) By taking control of the store sale process, Pick n Pay should not be seen as the gatekeeper of any merger counterfactual. (h) Evidence given by Mr Dennis Cope, the chief financial officer of the Pick n Pay Group, that an offer from the proposed consortium would not be entertained, is not persuasive because Pick n Pay, as a profit-maximising firm, would accept the highest lawful offer. (i) An offer for the whole of the Franklins business would provide a faster and cleaner exit than a fresh store sale process and would avoid substantial shutdown costs for Pick n Pay. (j) A wholesale operation conducted by the prospective KKKL consortium would be viable. Such a possibility should not be dismissed out of hand as being based on nothing but speculation or theory. By pursuing an offer for the Franklins assets, the KKKL consortium members considered the possibility of such a venture to be economically feasible and not merely speculative. (k) The current volumes of retail sales of the members of the prospective KKKL consortium could be added to a sufficient proportion of the volume of Franklins sales to achieve a greater scale than Franklins was able to achieve, and the KKKL consortium would sell stores to attract franchisees who would bring extra volume to a proposed wholesale operation. 184 In the end result, however, the primary judge concluded (at [396]) that the KKKL consortium could not succeed in an offer along the lines foreshadowed by the TMT Consortium offer. The primary judge concluded that it was unlikely that Pick n Pay would accept such an offer. His Honour made a number of important findings in support of those conclusions. It is necessary to record those findings in some detail. 185 First, the primary judge found that there was no credible evidence to support the proposition that the KKKL consortium was likely to make a serious binding offer for all, or a significant part, of the Franklins assets. 186 In this connection the primary judge found that many fundamental aspects of the consortium's intentions remained unresolved. A great many matters were required to be resolved, and a great deal of work was required to be done, in order to create a new wholesaling business in competition with Metcash. 187 The primary judge (at [398]) described the position thus: The proposed consortium is far from having undertaken the detailed steps that will be required. Importantly, no binding offer could be submitted until the following matters were resolved: the identity of the members of the consortium and the share that each member in the consortium would take; the corporate structure that would be used to acquire the assets for the consortium; the terms of any consortium agreement, including the rights and obligations of members of the consortium, the equity structure, voting rights and decision-making process, board representation, transfer of equity interests and provision for exit; the precise assets that the consortium would wish to acquire; the price to be paid for the assets to be acquired; the amount of working capital necessary to operate the business, including expenditure for the improvement of stores; the funding of the purchase price, including the relative proportions of debt and equity funding; the numbers of Franklins Corporate Stores that would be retained by the consortium, be sold or be closed down; how the proposed consortium would obtain landlords' consents and assignments of leases of the Franklins Corporate Stores; the brand that would be used by the consortium; the relationship between the consortium members in relation to the operation of a warehouse and distribution assets, and the terms, including pricing strategies, on which any warehouse or wholesale business would supply groceries, both to consortium members and to non-members; ownership of the warehouse assets; and the employment of personnel who would operate the warehouse and carry out buying functions. 188 In this connection the primary judge noted evidence from Mr Vasilli Karellas that, before putting in a binding offer, the proposed consortium would need to resolve most of the matters listed above. His Honour also noted evidence from Mr Koundouris that there were many matters to finalise. 189 Secondly, the primary judge found that there was no evidence that the KKKL consortium had adequate funding to acquire all or a significant part of the Franklins assets. Indeed, his Honour found that there was no credible evidence of the capacity of the prospective consortium members to obtain such funding. 190 In this connection the primary judge noted a letter sent by the National Australia Bank to the directors of the Koundouris Group on 10 October 2010. His Honour found, however, that this letter fell well short of providing any assurance of funding; there was no information provided as to the number of stores, or the amount of funding required. His Honour found that this letter indicated no more than a willingness on the part of the bank to consider any detailed request for funding. 191 The primary judge also noted in this connection that Mr Vasilli Karellas had told the Commission in September 2010 that it would be difficult for the consortium to get financing should it decide to make an offer. The primary judge noted, on the other hand, that Mr Koundouris had said that he was confident of obtaining funding because of what his bank manager had told him when he was contemplating a possible offer to acquire Franklins in 2007. The primary judge concluded, however, that it was unlikely that Pick n Pay would share that confidence. 192 Thirdly, the primary judge found that there was doubt about the final membership of the proposed consortium. 193 In this connection the primary judge noted evidence from Mr Lionis that he considered the TMT Consortium offer to be more in the nature of an exercise in getting information about the Franklins business, with a view to the possibility of putting forward a credible offer. He said that he needed a good deal more information before he was prepared to put in a binding offer. The primary judge reasoned that that evidence supported the conclusion that the lodging of the TMT Consortium offer was merely the first step in the process of engaging with Pick n Pay. Indeed, Mr Lionis said there was no consortium at the time the TMT Consortium offer was made. 194 The primary judge noted that Mr Vasilli Karellas accepted that, before the consortium could make a binding offer, he and his family members would have to resolve whether they wished to participate. The primary judge said that, although Mr Karellas gave evidence in general terms of several meetings with Mr Koundouris, and later with Mr Koundouris, Mr Krnc and Mr Lionis, there was no certainty in the terms of the discussions about which Mr Karellas gave evidence. 195 The primary judge noted that Mr John Krnc accepted that he did not have a single document relating to his involvement in the consortium, and said nothing in his affidavits about the acquisition of Franklins. 196 The primary judge noted that Mr Koundouris gave generalised evidence about the making of the TMT Consortium offer and, although speaking of Supabarn's interest in purchasing the whole of the Franklins business, he did so without mentioning any consortium. 197 In the end result, the primary judge concluded that, notwithstanding the evidence of some of the consortium members that they were prepared to follow through with the proposal if the Metcash acquisition did not proceed, that evidence did not provide an adequate foundation upon which to conclude that the proposed consortium's interest was credible and likely to produce a realistic binding offer. 198 Fourthly, the primary judge accepted that Pick n Pay intended to sell the Franklins business on a store by store basis in the event that the share acquisition by Metcash did not proceed to completion. 199 In this connection the primary judge noted Pick n Pay's evidence that any binding offer submitted by the KKKL consortium would not be accepted unless it were for an amount that exceeded Pick n Pay's assessment of the likely financial return to it from a disposition on a store by store basis. In short, the non-binding indicative offer of $110 million would need to be substantially increased for it to have any prospect of acceptance. The offer of $110 million was markedly inferior to the net financial return expected through a competitive store tender process. Importantly, the primary judge noted evidence by Mr Cope that he had serious doubts about the ability of the KKKL consortium to complete an acquisition of the Franklins business. The primary judge noted Mr Cope's concerns about the ability of the consortium to fund the purchase and replace bank guarantees concerning leases and workers' compensation. The primary judge also noted Mr Cope's concern about the absence of any material indicating that the proposed consortium could provide landlords with sufficient comfort to persuade them to consent to assignments of leases. 200 The primary judge also found that there was no credible basis for concluding that Pick n Pay would wait to see whether the proposed consortium could make a binding offer. Moreover, the primary judge concluded that even if a binding offer could be made, the high execution risk would make that offer unattractive to Pick n Pay. 201 The primary judge then turned to consider what would happen under a store sale process. His Honour accepted that, in the circumstances before him, it would be necessary for Pick n Pay to undertake a renewed store sale process should Metcash's share acquisition be prohibited. The results of the initial store sale process suggested, however, that it was unlikely that the proposed consortium would be the highest offeror for more than a handful of the Franklins corporate stores. The primary judge noted that, between them, the respective consortium interests submitted bids for a total of 18 stores, one of which was not in fact owned by Franklins. A consortium member was the top or equal top offer in respect of only two of the stores. The primary judge concluded that, on that basis, the KKKL consortium was unlikely to obtain any material number of stores through a fresh store sale process. 202 The primary judge noted that, in order for the KKKL consortium to succeed to the extent contemplated by the Commission's counterfactual, it would be necessary for the consortium to make an offer for substantially more stores, in a store by store sale process, than the sum of the individual offers previously made by the prospective consortium participants. Even then, the prospects of success would rest on either (a) the consortium offering more than Woolworths, Coles, Metcash, Ritchies or any other independent retailer who intended to obtain supply from Metcash, or (b) the Commission successfully preventing any of those entities from obtaining any number of stores that would prevent the consortium acquiring a significant majority of them. 203 As to the latter matter, the primary judge noted evidence from Mr Grimwade, the most senior officer of the Commission giving evidence, that the Commission had not yet considered what might happen under a renewed store sale process. He could not give any assurance that the Commission would oppose every sale of a store to any of Woolworths, Coles, or Ritchies, and could not say that any particular offeror, including Woolworths, Coles and Ritchies, would be precluded from acquiring stores. 204 Fifthly, the primary judge did not accept that, if the KKKL consortium acquired the Franklins assets, it would establish a viable and sustainable wholesale operation in competition with Metcash. 205 In this connection, the primary judge made the following important findings: (a) The members of the consortium were retailers, not wholesalers, and it would be necessary to employ people with considerable skill to conduct the wholesale operations. (b) The preliminary work conducted by Mr Koundouris and Mr Lowrey indicated that a majority of the Franklins corporate stores would need to be sold or franchised to unidentified third parties; but there was no evidence as to who the third parties might be or how any franchising business would work. (c) Mr John Krnc gave evidence, which the primary judge found to be logical and compelling, that, in any event, the last thing a retailer would want to do would be to take supply from someone who was a competitor in the marketplace (such as the individual consortium members). (d) Mr Koundouris assumed that the KKKL consortium could simply take over any supply contract held by Franklins, and get the benefit of any terms negotiated by Franklins, but no reason was advanced why any manufacturer or primary supplier would agree to novate their supply contracts to a new entity with no track record in wholesaling. It was more likely that the proposed consortium would need to negotiate new terms which could well be materially less favourable than the terms under which Franklins operated. (e) The Franklins business, as a loss-making business that had been unable to compete with Coles and Woolworths, provided a most unpromising foundation for a new and competitive wholesaling business. The primary judge noted that Mr Koundouris had asserted that the proposed consortium could operate the warehouse as a cost centre with a 2.5% margin. The primary judge concluded, however, that there was no evidence to support the fact that the warehouse could be operated on that margin. The primary judge noted that the figure of 2.5% had not been derived from any calculation, and appeared to be no more than a speculative guess by Mr Koundouris. (f) In any event, operating the warehouse at the suggested margin said nothing about the attractiveness of prices to independent retailers. Independent retailers would not choose between wholesalers on the basis of the wholesalers' respective margins. Rather, they would choose between wholesalers on the basis of price and the range of associated services offered to support the retailer. (g) There was no evidence as to the net price at which the proposed consortium could or would acquire groceries from manufacturers and other primary suppliers. But it was clear that the proposed consortium would be at a significant scale disadvantage to Metcash, who would be able to buy on better terms. (h) Even if it were possible for the proposed consortium to step into the shoes of Franklins, Franklins had not been able to compete effectively or sustainably with Coles, Woolworths or Metcash. (i) There is no evidence that the proposed consortium could provide any of the various support services provided by Metcash to independent retailers to assist them to compete with Coles and Woolworths. 206 Sixthly, the primary judge found that there was no evidence as to who would buy, or take a franchise of, unwanted stores acquired by the KKKL consortium. The primary judge noted that Mr Koundouris had prepared a two page document concerning Franklins, in which Mr Koundouris had planned to keep 23 stores, to franchise 45 stores, and to close 12 stores. The primary judge made a number of criticisms about the analysis in this document. Significant amongst these criticisms was what his Honour described as the "undeveloped plan" to franchise 45 stores, given that Franklins had only been able to franchise 10 stores to five franchisees in over five years of effort. The primary judge therefore gave little weight to the planned franchise of 45 stores, which was central to Mr Koundouris' calculations of the amount ultimately required to fund the acquisition of the Franklins business. 207 After noting the obvious difficulties with the plan to franchise unwanted stores, the primary judge turned to consider what would happen if the consortium sought, instead, to sell those stores. The primary judge found that there was no evidence to suggest that the consortium would adopt anything but a profit-maximising strategy in its own store sale process. His Honour therefore concluded that it was likely that the buyers would be Coles, Woolworths, Metcash, or Ritchies and other independent supermarket owners who may choose to be supplied by Metcash. Pursuing such a strategy would lead to a result that was quite inconsistent with the Commission's counterfactual case. In other words, selling the stores would erode the very base that was required to undertake wholesaling operations. 208 With all these considerations in mind, the primary judge concluded: 419 Finally, the proposed consortium's interest is, at best, speculative. That is to say, its interest is substantially undeveloped. Its primary goal to date has been, clearly enough, to block the Metcash acquisition without spending the money and resources that would be required to demonstrate that it is itself in a position to make a serious, binding offer. The interest of the consortium is driven by strategic considerations aimed at creating an opportunity for it to commence to develop some sort of proposal, the details of which are as yet unclear. 420 In the light of the matters set out above, including the numerous and detailed basic matters still needing to be resolved, such as membership and participation, the lack of any proper business plan, the lack of information as to available funding, the absence of identification of purchasers for the Franklins Corporate Stores to be sold or franchised, and the lack of evidence of experience of consortium members in wholesaling operations, the establishment of a viable wholesale business by the consortium must be regarded as entirely speculative. Further, based on Mr Cope's assessment and analysis of the Store Sale Process, even if a serious proposal could be developed, any such proposal would be likely to produce a lower financial return for Pick n Pay than that expected from a store by store sale process. 209 The primary judge also concluded: 423 If the Metcash transaction does not proceed, the likelihood is that the Franklins stores will be sold in groups or individually to purchasers including Ritchies, Coles, Woolworths and independent retailers. In relation to the stores that are sold to independent retailers, it is likely that those retailers would obtain supply of packaged groceries from Metcash. The Commission has identified no other likely source of supply. Further, the intention of the seller in these circumstances is critical. I am satisfied that it is unlikely that Pick n Pay would enter into any transaction with the proposed consortium. 210 Ultimately the primary judge concluded that, on the basis of the evidence before the Court, he was not persuaded that it was more likely than not that the KKKL consortium would make an offer to acquire the whole, or a significant majority, of the Franklins assets, that would be accepted by Pick n Pay. Indeed, the primary judge was not persuaded that there was a real chance that that would happen.

The Commission's grounds of appeal 211 In its notice of appeal the Commission contended that the primary judge erred in finding that it was a matter of pure speculation whether a binding offer might ever be made by the KKKL consortium for the Franklins assets or that Pick n Pay would ever accept any binding offer from the KKKL consortium for those assets. It contended that the primary judge also erred in finding that the establishment by the KKKL consortium of a viable business for the wholesale supply of packaged groceries to independent retailers in New South Wales and the Australian Capital Territory must be regarded as entirely speculative. It contended that the primary judge erred in finding that the acquisition by Metcash of a material parcel of the Franklins stores was a more likely scenario than any acquisition by the KKKL consortium. 212 It is important to note that the findings which the Commission contends were made in error are in fact conclusions to which the primary judge came. These conclusions were based on findings of primary fact that are not challenged in this appeal. 213 The Commission contended that the primary judge should have found that, if Metcash's share acquisition did not proceed, there was a real but not remote chance that the KKKL consortium would make a binding offer for the whole, or binding offers for a substantial proportion of, the Franklins assets that would be accepted by Pick n Pay and that, in those circumstances, the consortium would commence wholesale supply of packaged groceries to independent retailers in New South Wales and the Australian Capital Territory. In short, the Commission contended that the primary judge should have reached conclusions contrary to those to which his Honour had come. 214 In this connection the Commission contended that the primary judge erred in finding that, in order for the Commission to succeed, it needed to establish that it was more probable than not that the Commission's proffered counterfactual involving the KKKL consortium would occur. It contended that the primary judge should have found that, in order to succeed, it was only necessary for the Commission to show that that counterfactual was likely to occur as a real but not remote chance.

The parties' submissions on appeal 215 The Commission's principal submission was that, in reaching the conclusions to which he came, the primary judge applied an incorrect standard. In this connection, s 50(1) of the Act provides that: A corporation must not directly or indirectly: (a) acquire shares in the capital of a body corporate; or (b) acquire any assets of a person; if the acquisition would have the effect, or be likely to have the effect, of substantially lessening competition in a market. [Emphasis added] 216 The primary judge observed that the phrase "be likely to have", as there used, is capable of having the meaning "more probable than not" or a meaning reflecting the lesser standard of "a real chance". 217 This conclusion was based on the analysis of the case authorities undertaken by French J (as his Honour was) in AGL at [342]-[348]. In that case his Honour observed that, although there has been some divergence in the construction of the word "likely" as used throughout the Act, the weight of authority favoured the meaning of "a real chance" rather than "more likely than not": Tillmanns Butcheries Pty Ltd v Australasian Meat Industry Employees' Union (1979) 42 FLR 331 at 347; Global Sportsman Pty Ltd v Mirror Newspapers Pty Ltd (1984) 2 FCR 82 at 87; News Limited v Australian Rugby Football League Limited (1996) 64 FCR 410 at 564-565; Monroe Topple & Associates Pty Ltd v Institute of Chartered Accountants in Australia (2002) 122 FCR 110 at [111]. His Honour observed (at [347]) that neither language nor policy would mandate a variation of construction in the meaning of "likely" from section to section in the Act. Importantly, his Honour also observed that, if "likely" meant "more probable than not" it would be difficult to distinguish between the application of the two limbs of s 50(1) relating to competitive effect. This was because a finding that an "acquisition would have the effect", as opposed to would "be likely to have the effect", would necessarily be based on facts established on the balance of probabilities. As a matter of reasoning, a finding based on the alternative or second limb - that an acquisition would "be likely to have the effect" - must be one established by a different and lesser standard of persuasion. 218 Although in their written submissions on appeal Metcash and Pick n Pay formally reserved their respective positions, none of the parties, on appeal, sought to articulate an argument to the effect that, in its particular statutory context, the word "likely" had a meaning other than "a real chance". The arguments advanced on appeal all proceeded on the basis that "likely" did have that meaning. 219 The issue that divided the parties was the primary judge's conclusion (at [145]) that the Commission must establish, on the balance of probabilities, what the future state of the market will be, both with and without the proposed acquisition. In other words, the Commission had to establish that its counterfactual case - the future without the share acquisition by Metcash - was more probable than any competing hypothesis. Specifically, the Commission had to establish that it was more likely than not that, if the Metcash share acquisition did not proceed to completion, the KKKL consortium would make an offer to acquire the whole, or a significant majority of, the Franklins assets that would be accepted by Pick n Pay, and that the consortium as a wholesaler would commence to supply packaged groceries, as defined by the Commission, to independent retailers in the defined territory. 220 Once that scenario was established, the future with and without the Metcash share acquisition could then be evaluated to determine whether the acquisition would be likely to substantially lessen competition in the defined market. If there was a real chance that competition would be lessened, then the acquisition would contravene s 50(1) of the Act. 221 The Commission submitted that a requirement to establish a counterfactual case on the balance of probabilities posits an erroneous standard that is inconsistent with, and which undermines, the proper application of the "real chance" limb of s 50(1) of the Act. 222 In aid of its submission the Commission pointed to the fact that "elevating" a counterfactual to proof on the balance of probabilities is inherently problematic where there are multiple counterfactuals, each of which has a real chance of occurring. In those circumstances it asks: Does an applicant fail if it can demonstrate a real chance of its counterfactual occurring but a respondent can point to two or more alternative counterfactuals that also have a real chance of occurring? Similarly, does an applicant fail because it cannot show that its counterfactual is more likely than a respondent's counterfactual, even though there is a real chance of the applicant's counterfactual occurring? 223 For their part, Metcash and Pick n Pay each submitted that the primary judge was correct to analyse this aspect of the case in the way that his Honour did. 224 Metcash submitted that the "real chance test" does not apply as the standard of proof in relation to the identification of counterfactuals. When considering what would occur with and without the acquisition, the ordinary civil standard of proof applies. Metcash submitted that this follows from the dicta of Smithers J in Dandy Power at 259-260 ("… it is necessary to assess … the probable nature and extent of competition …"), which is taken as establishing the conceptual framework for determining whether there will be a substantial lessening of competition in a given case. Metcash also submitted that this approach is supported by the approach taken in AGL by French J (at [356]), at the behest of the Commission. It submitted that the Commission's rhetorical questions are answered by simply determining which counterfactual, out of a range of counterfactuals, is more likely. 225 Pick n Pay's submissions were, in this respect, substantially to the same effect. It submitted that the necessary starting point is the fact that an applicant alleging contravention of s 50(1) bears the onus of establishing a contravention on the civil standard of proof. It submitted that the burden extends to establishing something about the conditions of the market absent the impugned transaction, and that a distinction must be drawn between this burden and the fact ultimately to be proved, namely whether the transaction would be likely to substantially lessen competition. It submitted that a distinction is to be drawn between the "ultimate competition analysis" and the underlying facts to be proved as the basis for the comparison to be made. In short, the counterfactual world must be established on the balance of probabilities, including choosing which of several counterfactuals is the most likely.

Consideration 226 The competing submissions of the parties throw up difficult, and in some respects perplexing, questions about the standard of proof that is to be applied, not only in cases falling under s 50 but also under other provisions of the Act, particularly those directed to determining the likely state of competition in a given market. 227 If one accepts that the starting point is to draw a distinction between circumstances where an acquisition would have the effect of substantially lessening competition, established on the balance of probabilities, and circumstances where an acquisition would be likely to have that effect, established on the basis that there is a real chance that that would be so, it can be seen that s 50(1) itself imposes its own differential standards of proof, at least so far as the determination of competitive effect is concerned. The utility of imposing differential standards, as a matter of legislative policy, is not at all clear, given that contravention will always be established on the lowest threshold being satisfied. Nevertheless, if one is to proceed on that basis in the present case, then one question involving one evaluative judgment emerges: would the acquisition be likely to substantially lessen competition in the relevant market? 228 The answer to that question points to, and depends on, the interrelationship of all the facts, matters and circumstances which, in combination, define the future state of affairs that is characterised as being "likely". If, for the purpose of satisfying the requisite legal standard, "likely" is taken to have the meaning of "a real chance", then it is difficult to see why that standard should not apply to determine the existence and interrelationship of all those facts, matters and circumstances. If not, the possibility exists that different legal standards will intrude into inseparable elements of the calculus employed to detect change to the state of competition. As I have noted, a counterfactual is no more than an element of that calculus. Conceptually, it has no separate existence or purpose in the present context, other than as an aid to detect the existence and extent of change in the process of competition. 229 Moreover, in the continuum of fact-finding, there may not be a bright line between those facts that determine the future state of a market and those facts that determine the future state of competition in that market. Indeed, one can envision examples where the facts that show the likely future state of the market will be the very facts that are determinative of a finding about the likely future state of competition in that market. In those cases, can fact-finding be regulated by two different standards of proof? To require, in those cases, the adoption, if that be conceptually possible, of a higher standard for one purpose (to determine the state of the market) would be to obliterate the threshold to which the second limb of s 50(1) has subjected the impugned conduct. 230 It is not necessary, however, to come to a final view on these matters because the primary judge concluded that, if the Commission was only required to establish a real chance that its counterfactual case would come to pass in the event that the Metcash acquisition did not proceed to completion, the Commission had not satisfied that standard. 231 The Commission challenged this finding by the primary judge. It said that this finding was inherently improbable on the facts and was not attended by reasons. In my view those submissions cannot be sustained. 232 It is clear that the primary judge gave detailed consideration to the evidence before him and, if I may say so, set out his findings in a methodical and comprehensive way. As I have already noted, those findings are not challenged. The Commission's challenge is to the weight to be given to, and the conclusions to be drawn from, those findings. 233 The primary judge's findings are compelling and fully support his Honour's conclusion (at [425]) that it was a matter of pure speculation as to whether a binding offer would ever be made by the KKKL consortium, and as to whether such an offer would be accepted by Pick n Pay. It follows from this finding that there could not be a real chance that a binding offer would be made by the KKKL consortium which would be accepted by Pick n Pay. 234 In AGL, French J (at [348]) observed: The meaning of "likely" reflecting a "real chance or possibility" does not encompass a mere possibility. The word can offer no quantitative guidance but requires a qualitative judgment about the effects of an acquisition or proposed acquisition. The judgment it requires must not set the bar so high as effectively to expose acquiring corporations to a finding of contravention simply on the basis of possibilities, however plausible they may seem, generated by economic theory alone. On the other hand it must not set the bar so low as effectively to allow all acquisitions to proceed save those with the most obvious, direct and dramatic effects upon competition. By the language it adopts and the function thereby cast upon the Court and the regulator in their consideration of acquisitions s 50 gives effect to a kind of competition risk management policy. The application of that policy, reflected in judgments about the application of the section, must operate in the real world. The assessment of the risk or real chance of a substantial lessening of competition cannot rest upon speculation or theory. To borrow the words of the Tribunal in the Howard Smith case, the Court is concerned with "commercial likelihoods relevant to the proposed merger". The word "likely" has to be applied at a level which is commercially relevant or meaningful as must be the assessment of the substantial lessening of competition under consideration - Rural Press Ltd v Australian Competition and Consumer Commission (2003) 216 CLR 53 at [41]. 235 Those observations are apposite to the present case. The primary judge was called upon to make a "real world" assessment based on matters that were commercially relevant and meaningful. The primary judge did so. His Honour was not required to move on mere possibilities, let alone speculative possibilities. Indeed, to have done so would have involved error. 236 Moreover, having set out his findings of fact, and the detailed reasons for them, it was not then necessary for the primary judge to embark upon a further analysis to justify the conclusion, which was plainly open to him, that there was not a real chance that the Commission's counterfactual case would come to pass. I would add that, contrary to the Commission's submission, I do not accept that, in coming to that conclusion, the primary judge failed to make an overall assessment, but instead made a piecemeal assessment that, somehow, did not truly evaluate the likelihood of the prospect that was advanced for his Honour's consideration. 237 In my view the Commission has not demonstrated appealable error in the primary judge's evaluation of, and conclusions on, the Commission's counterfactual case based on the application of a "real chance" test. In the absence of the Commission demonstrating appealable error, its case for contravention of s 50(1) of the Act must fail, regardless of the question of market definition. It follows that, for this reason alone, its appeal must fail.

The commission's case on market definition 238 As I have noted, the primary judge concluded (at [342]) that the Commission had failed to discharge its onus of establishing that there was a separate market for the wholesale supply of packaged groceries to independent supermarket retailers, as defined by the Commission. 239 The primary judge accepted (at [180]), in accordance with the Commission's case, that the geographic dimension of any market for the wholesale supply of packaged groceries would be New South Wales and the Australian Capital Territory. However the primary judge (at [203]) did not accept that the product dimension of the market should be delineated by reference to packaged groceries, as the Commission would have it. 240 In this connection, the Commission, in its amended statement of claim, had defined the product dimension of the market as follows: Wholesale packaged groceries comprise a significant proportion of the goods sold in supermarkets. Wholesale packaged groceries include branded and generic items such as breakfast cereal, canned food, biscuits, flour, tea, coffee, soft drinks, nappies, cleaning products, personal hygiene products and frozen goods. Wholesale packaged groceries do not include fresh items such as fresh fruit and vegetables, meat, deli or bakery items … 241 More importantly for the purposes of this appeal, the primary judge reasoned (at [334] and [336]) that, if Franklins is in the relevant market with Metcash (as the Commission contended), the major supermarket chains, particularly Woolworths and Coles, must be included in that market. This was because, on the primary judge's findings, the major supermarket chains imposed a closer (in the sense of greater) competitive constraint than Franklins on Metcash. The inevitable consequence of that reasoning for the present case was that the definition of the market must accommodate multiple functional levels given that the major supermarket chains were and remain vertically integrated and the constraint that the primary judge found was one that was ultimately imposed on Metcash because of fierce competition among grocery retailers. 242 It is important to note that, based on his findings as to the degree of constraint exercised by the major supermarket chains on Metcash, the primary judge reasoned (at [430]) that it did not much matter, in the circumstances of the present case, whether the major supermarket chains were taken into account when defining the market or when considering the effect that the proposed acquisition would have on competition in the market defined by the Commission. Either way, the primary judge reasoned, the major supermarket chains constrained Metcash closely and would continue to do so. Nevertheless, the primary judge saw the market as most appropriately defined by taking into account the major supermarket chains. 243 The primary judge's conclusion in that regard was the focus of much debate on appeal. Even though my own views on the fate of the Commission's counterfactual case lead me to conclude that the appeal should be dismissed in any event, I wish to express my views about the correctness of the primary judge's conclusions on market definition. Before doing so, however, it is convenient to note some aspects of the process of market definition and the purpose for which that process is undertaken.

The process and purpose of market definition 244 In Australian Competition and Consumer Commission v Liquorland (Australia) Pty Ltd (2006) ATPR ¶42-123 Allsop J said (at [429]): Market definition is not an exact physical exercise to identify a physical feature of the world; nor is it the enquiry after the nature of some form of essential existence. Rather, it is the recognition and use of an economic tool or instrumental concept related to market power, constraints on power and the competitive process which is best adapted to analyse the asserted anti-competitive conduct. 245 The seminal expression of "the market" for the purposes of Australian competition law is to be found in Re Queensland Co-operative Milling Association Ltd; Re Defiance Holdings Ltd (1976) 25 FLR 169 at 190, where the Trade Practices Tribunal said: We take the concept of a market to be basically a very simple idea. A market is the area of close competition between firms or, putting it a little differently, the field of rivalry between them. (If there is no close competition there is of course a monopolistic market.) Within the bounds of a market there is substitution-substitution between one product and another, and between one source of supply and another, in response to changing prices. So a market is the field of actual and potential transactions between buyers and sellers amongst whom there can be strong substitution, at least in the long run, if given a sufficient price incentive. Let us suppose that the price of one supplier goes up. Then on the demand side buyers may switch their patronage from this firm's product to another, or from this geographic source of supply to another. As well, on the supply side, sellers can adjust their production plans, substituting one product for another in their output mix, or substituting one geographic source of supply for another. Whether such substitution is feasible or likely depends ultimately on customer attitudes, technology, distance, and cost and price incentives. It is the possibilities of such substitution which set the limits upon a firm's ability to "give less and charge more". Accordingly, in determining the outer boundaries of the market we ask a quite simple but fundamental question: If the firm were to "give less and charge more" would there be, to put the matter colloquially, much of a reaction? And if so, from whom? In the language of economics the question is this: From which products and which activities could we expect a relatively high demand or supply response to price change, i.e. a relatively high cross-elasticity of demand or cross-elasticity of supply? 246 The need to consider the possibilities of substitution in defining a market for competition law purposes is enshrined in the definition of "market" for the purposes of the Act. Section 4E provides: For the purposes of this Act, unless the contrary intention appears, market means a market in Australia and, when used in relation to any goods or services, includes a market for those goods or services and other goods or services that are substitutable for, or otherwise competitive with, the first-mentioned goods or services. 247 One tool that is used to provide an analytical framework to identify and evaluate substitution possibilities is the "SSNIP test", also referred to in the present case as "the hypothetical monopolist test" (a description which I will use in these reasons). This test involves determining whether a hypothetical monopolist supplier could profitably impose a small but significant non-transitory increase in price (most commonly, but not necessarily, between 5 and 10%) for the supply of a relevant product. Starting with the firm and product in issue, the market boundaries are expanded to include all sources of close substitutes that would defeat the increase. The smallest area, generally in terms of product identification and geographic space, over which the hypothetical monopolist can profitably impose the increase, shows the boundaries of the market. 248 The hypothetical monopolist test is predicated on the availability of data on variables, such as costs, prices, revenue and sales, over a sufficiently long period of time to enable a mathematical determination to be made about how changes by a firm to its prices affect its own demand. In competition law, however, the test is not always applied in that way. Sometimes it is applied without data as a "thought experiment" to make a qualitative assessment about the product and geographic dimensions of the market: Seven Network Limited v News Limited [2007] FCA 1062 at [1786]. Indeed, paragraph 4.22 of the Commission's own Merger Guidelines promotes and justifies such an approach: While the [hypothetical monopolist test] is a useful tool for analysis, it is rarely strictly applied to factual circumstances in a merger review because of its onerous data requirement. Consequently, [the Commission] will generally take a qualitative approach to market definition, using the [hypothetical monopolies test] as an "intellectual aid to focus the exercise". 249 It is apparent that, when the hypothetical monopolist test is applied in this fashion, conclusions can be reached about the boundaries of a market on which reasonable minds might differ. It follows from this realisation that a difference in opinion in the identification of market boundaries does not necessarily signify the presence of error in the evaluative process. 250 It is also apparent that, when the test is applied in this fashion, considerations other than price (the ability to "give less") can be accommodated in the evaluative process. In Seven Network Ltd v News Ltd (2009) 182 FCR 160 at [670] Dowsett and Lander JJ remarked: We do not treat the SSNIP test as being irrelevant to the question of market identification. However a qualitative application of the test requires identification of its purpose. As we understand it, the test looks to the actual or likely effect of competitive conduct, or potential competitive conduct, upon price and other conditions of supply, including quality of the product. However competitive conduct may not have an immediate and obvious effect upon those matters. Particularly in a relatively new industry, competitors may be looking for longer term, rather than shorter term, advantages. The "richness" of the concept of competition referred to in Re Queensland Co-operative Milling Association Ltd 25 FLR 169 means that competition may take many forms. Its effects may be immediate or delayed. The SSNIP test addresses the effects of competition, but it does not define the way in which it occurs. 251 The hypothetical monopolist test says little about the functional dimensions of the market. While substitution provides a basis for defining product and geographic markets, it does not provide a meaningful basis for determining the functional dimensions of a market. As Smith and Walker explain in Smith R and Walker J, "Part IIIA Efficiency and Functional Markets" (1998) 5 Aust Jnl of Corp Law 1 at 9: What point is there in asking how "substitutable" is one stage of production or distribution for another, when each is a complementary part of the process, like a link in a chain? We can meaningfully ask if one garment is substitutable for another garment, in the minds and intention of the provider or the buyer; but there is no point in asking whether the materials from which either garment is made are substitutable for the garment itself! 252 And yet, the behaviour of participants at one functional level may have a substantial constraining effect on the behaviour of participants at another functional level that is sufficient to warrant the inclusion of all those activities in the market the subject of attention. This can arise, for example, where the chains of distribution exhibit a sufficient degree of vertical integration. As Smith and Walker explain, the relevant issue for the delineation of functional markets is not whether there is substitution between the functional levels, but whether substitution between products or geographic sources of supply at another functional level constrains the behaviour of market participants at the initial functional stage: Smith and Walker at 9. The authors go on to explain (at 11): This would seem to be consistent with the application of the SSNIP test: if the initial functional market delineation does not incorporate all the relevant competitive constraints, it would not be possible for a hypothetical monopolist at that functional stage to impose a SSNIP. It would be necessary to expand the functional market and incorporate those constraints. 253 In Davids Holdings the relevant market for the purpose of the application of s 50 of the Act was found to be the market for the supply of grocery products by independent wholesalers to independent retailers in Queensland and northern New South Wales. Davids Holdings Pty Ltd (Davids) was the holding company of a group of companies whose principal activities were the wholesaling and distribution of a range of grocery products and liquor in Victoria, New South Wales and Queensland. QIW Retailers Limited (QIW) was the holding company of a group of companies whose principal activities were the wholesaling and distribution of a similar range of grocery products in Queensland and northern New South Wales. Davids and QIW competed fiercely for the business of independent retailers in Queensland and northern New South Wales. Although the major activity of QIW was this wholesaling activity, it also had not insignificant retailing activities. Davids proposed to take over the shares in QIW. In that case the trial judge found that, based on the market definition to which I have referred, a merged Davids-QIW entity would, or would be likely to, dominate the market, in contravention of s 50 of the Act (in its then form). 254 On appeal, Davids challenged the identification of the relevant market. It contended that the definition was too narrow and propounded a market for the supply and distribution of grocery products to consumers in Australia, or alternatively in the geographic area consisting of the eastern mainland States of Australia. This market would have included both the wholesale and retail functional levels of activity. On appeal it was submitted that the trial judge had made a fundamental error in failing to find that independent retailers were in competition in the same market with the chain stores, providing an upstream constraint on wholesalers. In considering that matter, von Doussa J (with whom O'Loughlin J agreed) noted (at 227-228) that the market identified by the trial judge followed essentially from findings of fact that had been made. 255 At 229C-F, von Doussa J said: In my opinion the trial judge did not make the fundamental error alleged. His Honour referred to the concepts of close competition and strong substitution which are central to the definition of a market, and on which this general submission of the appellant is premised. When doing so his Honour emphasised that the application of these concepts in a particular case involves questions of degree to be judged having regard to the practical realities of the commercial activities under consideration. This is in accordance with established authority: Queensland Wire (supra) at 187, 195-196 and 199 and Arnotts Ltd v Trade Practices Commission (supra) at 331-332. Recognising that this was so his Honour concentrated his attention on the facts established by the evidence before the Court. The factual conclusions reached deny the validity of the argument advanced by the appellants. The trial judge recognised that it is important to the identification of the relevant market for the purpose of the proceedings to consider the extent to which the competition and restraining influences of the activities of the participants at the retail level have an upstream effect upon competition at the wholesaling level. As a matter of fact he concluded that the degree of upstream influence does not provide significant competitive pressure so as to bring the chain stores and the independent wholesalers into close competition. If the conclusion that the independent wholesalers and the chain stores are not in close competition at the wholesale level is to be upheld as an inference fairly open on the primary facts, then it is not necessary to decide if there is a single retail market. In my opinion that conclusion is not only open, but is strongly supported by the evidence. 256 It is apparent from that passage that the facts of the case were, as one would readily accept, determinative of the evaluation that had been undertaken to define the relevant market. 257 In the same case, Drummond J (at 245) said: The merged firm would undoubtedly be constrained, as his Honour recognised, by the existence of the national chains, but only in the sense that there is a limit beyond which the merged firm could not raise its prices to the independent retailers without bringing those retailers into price competition with the retail chains to such an extent that the merged firm would not be able to maintain its own raised prices to the retailers. However, the fact that there are identified constraints upon the pricing activities of the merged firm and of the firms whose merger is under investigation does not mean that the organisations who have that constraining influence must of necessity be included in the market of which those firms themselves form part. As von Doussa J observes, the learned trial judge appreciated that the task of market identification for the purposes of s 50 of the Trade Practices Act 1974 (Cth) involves matters of degree. The point is clearly made by Dawson J in Queensland Wire Industries Pty Ltd v Broken Hill Pty Co Ltd (1989) 167 CLR 177 at 199: Important as they are, elasticities and the notion of substitution provide no complete solution to the definition of a market. A question of degree is involved - at what point do different goods become closely enough linked in supply or demand to be included in the one market - which precludes any dogmatic answer ... The process is an inexact one ... 258 Subsequently, in Re QIW Ltd (1995) 132 ALR 225 the Trade Practices Tribunal came to consider a proposed acquisition by Davids of the issued shares and other securities of Composite Buyers Ltd (Composite Buyers). The principal activities of both Davids and Composite Buyers were the wholesale distribution of groceries, refrigerated foods, general merchandise and liquor. The Commission had granted Davids authorisation to undertake the proposed acquisition. QIW sought a review of the Commission's determination. 259 In its decision, the Tribunal noted the judgments in Davids Holdings, at first instance and on appeal, remarking that the case had turned on its own facts and issues which were different from those in the review being conducted by the Tribunal. In considering the functional dimension of the relevant market, the Tribunal noted QIW's contention that the market was a separate wholesale market in New South Wales and Victoria. It expressed the view (at 265) that wherever there are market transactions of significance there is a need to distinguish a separate functional level. However, it also noted a difficulty in that it was only the independent sector that was not vertically integrated; some 70% of turnover at wholesale in Australia passed through the hands of vertically integrated supermarket chains. This led the Tribunal to conclude (at 266) as follows: In summary, there is an Australia-wide or national market for the distribution of grocery products to the consuming public via integrated retail chains and independent wholesalers supplying independent retailers. We distinguish two wholesale sub-markets of relevance to this application, namely, transactions between independent wholesalers and independent retailers in (1) the New South Wales region and (2) the Victorian region. We use the term "sub-market" to refer to a field of rivalry that is "especially close or especially immediate" reflecting "some discontinuity in substitution possibilities": Tooth & Tooheys at 18,197; QCMA at FLR 190 ; ATPR 17,247. In specifying these wholesale sub-markets we have, in effect, made two "cuts" in the pattern of substitution within the market as a whole: a functional cut to separate certain wholesale transactions between independent wholesalers and independent retailers, and a geographic cut to separate certain transactions within the independent sector that centre upon the geographic pattern of physical distribution. There are also retail submarkets whose functioning has significance for this application. We do not regard the various sub-markets as separable markets, however, for the activities concerned are subject to the ultimate discipline of pervasive competition with the national integrated chains. 260 The earlier Full Court decision in Singapore Airlines Limited v Taprobane Tours WA Pty Ltd (1991) 33 FCR 158 provides another example (albeit in the context of proceedings for contravention of s 46 of the Act) where the functional dimension of the market was defined by reference to both wholesale and retail activities. 261 In that case French J (with whom Spender J and O'Loughlin J each agreed) referred to the role of "the market" in competition law. His Honour said (at 174): In competition law it has a descriptive and a purposive role. It involves fact-finding together with evaluative and purposive selection. In any given application it describes a range of economic activities defined by reference to particular economic functions (eg manufacturing, wholesale or retail sales), the class or classes of products, be they goods or services, which are the subject of those activities and the geographic area within which those activities occur. In its statutory setting the market designation imposes, on the activities which it encompasses, limits set by the law for the protection of competition. It involves a choice of the relevant range of activity by reference to economic and commercial realities and the policy of the statute. To the extent that it must serve statutory policy, the identification will be evaluative and purposive as well as descriptive. 262 That case concerned an alleged misuse of market power, contrary to s 46 of the Act, in relation to the provision of airline services from Australia to destinations off shore. The critical question in the case was how the product dimension of the market was to be defined. However, when dealing with the functional dimension of the market, French J said (at 182): At the functional level it would be possible to limit the market to the supply of services by airlines to wholesalers. Such a limitation, however, seems unduly restrictive. The integration of wholesale and retail activities within Singapore Airlines and other agencies suggests that the appropriate functional level comprehends the supply of airline services to wholesalers, packaged tours by wholesalers to retailers and packaged tours by retailers to consumers. The exercise of market power at either of the two upstream functional levels is closely connected with and capable of affecting competition downstream in the chain identified. 263 In AGL French J (at [378]) said: The concept of market describes, in a metaphorical way, an area or space of economic activity whose dimensions are function, product and geography. A market may be defined functionally by reference to wholesale or retail activities or a combination of both … 264 In Re Fortescue Metals Group Ltd (2010) 242 FLR 136 the Australian Competition Tribunal, in dealing with proceedings for access under Part IIIA of the Act, came to consider the functional dimension of a market where vertically integrated producers were involved. The Tribunal said: 1040 Another question that arises is: Should two stages of the supply chain be "collapsed" into the same functional market? This question comes up when the constraint is indirect. For example, vertically integrated retail firms may consume only their own wholesaling services, while vertically separated retailers may consume only vertically separated wholesaling services. Although there is no direct competition between vertically integrated and vertically separated wholesalers, they indirectly constrain each other's behaviour. To explain, if a vertically separated wholesaler increased its prices and the vertically separated retailer passes this price increase on to consumers, the vertically separated retailer would lose market share to the vertically integrated retailer, who would consume more of their in-house wholesale services. 1041 It is when activity at one functional level (eg retail) constrains activity at another level (eg wholesale) that one may ask whether transactions at the two levels should be treated as being in the same market. Although the approach in merger cases has not been uniform, the answer has generally been in the affirmative. We believe, however, that it is preferable to find there to be separate markets, otherwise the market would consist of some firms that compete with each other (eg the vertically integrated retailer and the vertically separated retailers) and others that do not (eg the vertically separated wholesalers and the vertically separated retailers). This would be at odds with a commercial view of markets, as vertically separated suppliers at the two levels would not see themselves as part of a combined market. It would also be at odds with the approach taken to determining product and geographic markets, where the aim is to include substitution possibilities within the same market and exclude other products and geographic areas. In any event, as the two functions are usually complements and not substitutes, they ought not to be included in the same market. Instead, each functional market should be treated as relevant to the way in which vertically integrated firms constrain the market power of vertically separated firms, and vice versa. 1042 Why does it matter whether vertically integrated firms are included in the market or whether a functional split is created between two processes in the supply chain? It is true that in merger cases and abuse of market power cases, whichever approach is taken is unlikely to change the result. In Pt IIIA cases, however, the distinction is critically important because the promotion of a material increase in competition will only satisfy criterion (a) if it occurs in a market other than the market for the service. We have explained why, in our view, it is preferable to place only "competitors" (ie those who act on and react to a direct or indirect constraint) in the market. While the purposive approach to market definition means that it is possible for divergent tests to be applied to different sections of the Trade Practices Act, the purpose of market definition in Pt IIIA, ss 46 and 50 cases is to allow for an analysis of competition. This rather suggests that the approach to market definition should be uniform. 265 The Tribunal's observation that the result in merger cases and abuse of market power cases is unlikely to be affected whichever approach is taken, has resonance in the present case. As I have noted, in the present case, the primary judge expressed a similar view, having regard to the evidence before him and the findings based on that evidence which led his Honour to conclude that Woolworths and Coles exercised a closer constraint on Metcash than Franklins. 266 These authorities show that, as a matter of principle, there is no reason why, for competition law purposes, a market cannot be defined by reference to multiple functional levels. They illustrate that it might be appropriate to do so where downstream activities function to constrain upstream behaviour. Whether that is so depends on the facts presented for consideration and the evaluation of those facts by the relevant decision-maker. 267 Quite apart from these authorities, it should be noted that the Act itself proceeds on the basis that markets might be defined in this way. Section 50(3) lists the mandatory considerations to be taken into account when considering the competition test in s 50(1). One of those considerations is the nature and extent of vertical integration in the market: s 50(3)(i). 268 With those observations in mind, it is necessary to record the principal findings that the primary judge made which led him to reject the market definition propounded by the Commission.

The primary judge's findings on constraints operating on Metcash 269 The Commission's case was that Franklins was a close competitive constraint on Metcash, but that the major supermarket chains were not. It was for this reason that the Commission contended that the relevant market did not include the major supermarket chains as competitors. The Commission's case was that, although the major supermarket chains had an ultimate constraining effect on Metcash, it did not follow that they were participants in the market, properly defined. 270 It should be noted that, in contending that Franklins was a close competitive constraint on Metcash, the Commission's case on market definition focused on Franklins as a wholesaler, not as a self-supplying retailer. The significance of this lies in the fact that the close competitive constraint provided by Franklins was said to reside in its wholesale supply to the ten Franklins franchise stores and its threatened supply to other independent retailers, not its retailing activities through the 80 Franklins corporate stores. 271 The primary judge considered the question of constraint by reference to the following matters: The perception of competitive constraints. The major supermarket chains as a constraint. The threat of store sales as a constraint. The threat of constraint by Woolworths as a wholesaler. Franklins as a constraint.

The perception of competitive constraints 272 The primary judge noted that the views and practices of those within the industry are often instructive when considering the appropriate definition of the market: Boral Besser Masonry Limited v Australian Competition and Consumer Commission (2003) 215 CLR 374 at [257]. In this connection, the Commission had contended that Metcash and Franklins perceived each other as fierce competitors, whereas there was no suggestion in any internal Metcash document that it considered the major supermarket chains to be close or even distant constraints at the wholesale level, even though these documents referred to the competitive behaviour of the major supermarket chains at the retail level. 273 The primary judge found (at [211]-[214]) that Metcash's internal documents certainly demonstrated that it regarded the major supermarket chains as competitors at the retail level; that Metcash often acted on the basis that the major supermarket chains were major competitors; and that both Franklins and SPAR also identified the major supermarket chains as principal competitors. Correspondingly, Woolworths' board reports identified IGA (that is, the Metcash brand) as one of its three major competitors. 274 At trial, the Commission placed considerable reliance on the minutes of Metcash's executive management committee meetings and an analysis prepared by Mr Jardim who was the Chief Executive Officer of the IGA-Distribution division of Metcash from 2000 until February 2010. The primary judge summarised the minutes of meetings held in the period 16 April 2004 to 10 December 2004, and also, to a lesser extent, the minutes of meetings through to 2010. There were references in these minutes to the threat of new wholesale entry by Franklins and the activities of Franklins as a franchisor. The primary judge found (at [228]) that it was apparent that Metcash "very much had Franklins in mind" and that senior executives of Metcash "were clearly concerned with the competitive effect of Franklins as a wholesaler". However, the primary judge found (at [229]) that it was equally apparent that Metcash also had the activities of Coles and Woolworths in mind and that this material clearly demonstrated "that Metcash was very much concerned with the major supermarket chains as having a significant competitive effect on its business activities". 275 At [230] the primary judge found: The Commission asserts that Franklins, and in particular its franchise operation, provided a greater constraint on Metcash than the major supermarket chains, such that the major supermarket chains did not provide a close competitive constraint, while Franklins did. Certainly, Franklins was a standing item at the meetings of the Metcash executive management committee. However, the minutes of those meetings do not support the Commission's assertion. The minutes show that the major supermarket chains were also a standing item at the meetings. Comparison of the minutes suggests that the members of the committee spent more time discussing the major supermarket chains than they did discussing Franklins. Whereas references to Franklins decrease over time, the references to the threats imposed by the major supermarket chains increase. 276 The market activity analysis on which the Commission placed reliance was prepared by Mr Jardim in May 2009. It was a brief marketplace overview of recent activity and a comparative snapshot of Metcash's major competitors, and the pressures on them. The primary judge summarised that analysis and concluded that it disclosed a focused concern of Metcash, through Mr Jardim, about the competitive effect of the major supermarket chains on the IGA network. Significantly, the primary judge found (at [238]) that it did not disclose any concern about Franklins' wholesale activities, insofar as those activities had an impact on the activities of Metcash. 277 The Commission did not challenge these findings on appeal.

The major supermarket chains as a constraint 278 The primary judge found that Woolworths and Coles have approximately 80% of the national grocery market share, and that independent retailers regard the major supermarket chains as their competitors. The primary judge noted the Commission's acceptance that there was vigorous competition between independent retailers and the major supermarket chains. This competition involved both price and non-price factors (such as range, home brand or private label product range, store location, store appearance, standards of service, quality of fresh produce, and trading hours). The Commission contended, however, that the constraint from the major supermarket chains on Metcash's wholesale pricing was far from close and was limited to a minority of the stores supplied by Metcash. It followed, on the Commission's case, that the major supermarket chains placed no more than a limited constraint on Metcash. 279 The primary judge did not accept these contentions. 280 The primary judge found (at [257]-[260]) as follows: (a) Independent retailers are unable to increase their prices in response to increases in wholesale prices without losing significant volume to the major supermarket chains. Thus, if independent supermarkets lose business, Metcash loses business. (b) Metcash has adopted a strategy of assisting independent retailers supplied by it to compete against the major supermarket chains in terms of both price and non-price factors. (c) The scale and intensity of retail competition is extreme and increasing. (d) Volume is the key to success for major supermarket chains, independent retailers and Metcash alike. Both Metcash and the major supermarket chains are constantly looking for ways to improve volume. (e) There is competition between Metcash and the major supermarket chains for volume. 281 The primary judge found (at [269]) that while the constraints placed on Metcash by the major supermarket chains were, at least in part, indirect, they were nevertheless powerful and imposed a much closer and more effective constraint on Metcash than the constraint that Franklins imposed (a matter to which I will return). 282 The primary judge reasoned (at [270]) that, because there were tight margins in the industry (a fact on which the Commission itself relied), wholesale pricing directly affected retail competition and retail competition, likewise, affected wholesale pricing. This was one reason why the vertically integrated supermarket chains would not supply into a wholesale market for independent retailers, because to do so would confront them with the difficult task of juggling wholesale and retail prices. Another reason was that the major supermarket chains wanted to drive independent retailers out of business and appropriate their market share. The primary judge reasoned (at [295]) that it would be strange for the major supermarket chains to endeavour to enter a market for wholesale supply to competing independent retailers because to do so would only assist those retailers to compete better with the major supermarket chains' own retailing activities. 283 The primary judge concluded (at [271]) that, by reason of the constraint placed upon Metcash by the major supermarket chains, there was no scope for Metcash to raise its wholesale prices to derive monopoly profits.

Threat of sale as a constraint 284 The primary judge found (at [272]-[274]) as follows: (a) The major supermarket chains had expanded beyond large scale format stores into medium and smaller stores across Australia, with the consequence that almost every independent store faced local area competition from one of the chains. (b) The threat of the major supermarket chains to an independent retailer's stores was a significant and constant source of concern to Metcash. (c) In response to this activity, Metcash had established a strategy (called the Barons Strategy) which involved it acquiring a minority equity interest in independent retailers' stores. It had also resorted to lending funds to independent retailers to enable them to purchase stores. In some cases, Metcash had become a head lessee in respect of a store site in return for a right of first refusal. (d) There was competition between Metcash and the major supermarket chains for the retention and acquisition of retail stores and sites. (e) Metcash wanted all of the available Franklins corporate stores. Woolworths, among others, was interested in acquiring the more profitable of those stores. 285 The primary judge thus concluded (at [279]) that the threat of sale by independent retailers to the major supermarket chains constrained Metcash's wholesale pricing decisions.

The threat of constraint by Woolworths as a wholesaler 286 The primary judge found (at [294]) that it was unlikely that the major supermarket chains would, in the foreseeable future, offer wholesale supply to independent retailers in the defined territory. Although Woolworths had at other times supplied grocery products to independent retailers (and continues to do so in Tasmania through an equity participation in a company called Statewide Independent Wholesalers), and in late 2010 had mooted the possibility of investing with SPAR in the Franklins corporate stores, the primary judge found (at [293]) that it appeared unlikely that it would be profitable for Woolworths to supply its competitors at the expense of its own sales. The primary judge found it to be equally unlikely that independent retailers would rely on wholesale supply from their largest retail competitor. 287 The primary judge concluded that such interest as Woolworths had shown in acting as a prospective wholesaler to independent retailers did not support an inference that Metcash was constrained in that regard by Woolworths, or any other major supermarket chain: [296].

Franklins as a constraint 288 The Commission's case was that Franklins was Metcash's closest competitive constraint in the defined territory. It contended that this was evidenced by Project Energise and Project Ling Chi, to which I have briefly referred. 289 Under Project Energise, Metcash offered independent retailers additional rebates if those retailers achieved certain targets, referable in part to purchases made from Metcash. Retailers had to enter into a five year supply agreement with Metcash and submit detailed business plans, including refurbishment strategies, for each of their stores. Retailers also had to provide Metcash with security over their businesses, including a right of first refusal in the event that the retailer's business was to be sold. 290 The Commission contended that Project Energise was implemented by Metcash as a competitive response to Franklins' decision to self-supply and to its threat to operate as a wholesaler to independent retailers. 291 Metcash accepted that Project Energise was prompted in part by those concerns. It contended, however, that, more generally, Project Energise was designed to increase IGA retail sales volume and thus increase the revenue of the IGA-Distribution division. It contended that the project was prompted, more specifically, by the desire to encourage IGA retailers to invest in refurbishing their stores, and by a concern that Woolworths and Coles might acquire IGA stores. 292 Quite apart from this initiative, Metcash had offered Supabarn supermarkets (the Koundouris family interests) terms that were superior to the Project Energise rebate. These terms were back dated by more than eight months and also relieved the operators from undertaking a store refurbishment program. Metcash had also offered the Project Energise terms to some retailers without the requirement to achieve teamwork scores for the first six months of the arrangement. The Commission contended that these matters signified that Metcash's primary intention was to "lock in" independent retailers to prevent them from shifting to Franklins. 293 The Commission further contended that Project Energise led to significant increases in the percentage of rebates paid by Metcash to independent retailers in the defined territory, with no corresponding increase in rebates paid to independent retailers in other states. It contended that an inference should be drawn that the rebates reflected a price reduction that was considered by Metcash to be a necessary change to its profit-maximising pricing, in response to the competitive threat posed by Franklins. 294 The Commission contended that if Franklins were to be removed as a competitive restraint there would be an incentive for Metcash to revert to higher pricing. 295 The primary judge was not persuaded that Project Energise gave rise to an inference as to what Metcash would be likely to do if it acquired Franklins. His Honour was not persuaded that Project Energise demonstrated that Metcash had surrendered monopolistic profits in the face of a threat from Franklins. The primary judge found (at [324]) that Project Energise was driven by a desire to increase the volume of sales in New South Wales and that this was as much a response to Woolworths and Coles as it was to Franklins. 296 Under Project Ling Chi Metcash aimed to attack Franklins' wholesale operations in various ways. The Commission pointed to it as being evidence of the likelihood of retaliatory action by Metcash against any new competitor. However, the primary judge concluded (at [328]) that Project Ling Chi did not bear significantly on the degree of competitive constraint imposed by Franklins on Metcash. 297 The primary judge subsequently made a number of important findings relating to the relative constraint imposed by Franklins on Metcash. Those findings (at [330]-[334]) can be summarised as follows: (a) The competitive constraint imposed by Franklins upon Metcash's wholesaling activities was far less significant than the constraint imposed by the major supermarket chains. (b) The constraint imposed by Franklins had diminished over time. On the other hand the constraint imposed by the major supermarket chains was strong and increasing. (c) Franklins did not regard itself as a wholesaler, and its franchise operations were not set up or conducted on the basis that it was a wholesaler. (d) Direct demand-side substitution by consumers between supermarkets operated by independent retailers and supermarkets operated by the major supermarket chains imposed a substantial competitive constraint upon Metcash as a supplier of packaged groceries at the wholesale level. (e) Direct demand-side substitution by consumers between supermarkets operated by independent retailers and supermarkets operated by Franklins did not impose any substantial constraint on Metcash. (f) Franklins had weaker bargaining power with manufacturers and primary suppliers than Metcash and had not been able to obtain the same terms as Metcash. (g) Franklins operated only in New South Wales, whereas Metcash operated throughout Australia. This enhanced Metcash's relative bargaining power. (h) The threat of independent retailers ceasing to take supply from Metcash and selling their stores to one of the major supermarket chains imposed a substantial competitive constraint on Metcash as a supplier of packaged groceries at the wholesale level. (i) The threat of independent retailers switching to become a Franklins franchisee did not impose a meaningful constraint upon Metcash.

The primary judge's conclusions on market definition 298 In the context of considering the constraint imposed by Franklins upon Metcash, the primary judge concluded (at [334]) that, if Franklins was in the relevant market with Metcash (as the Commission clearly contended), it must be the case that the major supermarket chains, which his Honour found to be a closer competitive constraint than Franklins, must be included in that market. 299 The primary judge found (at [337]) that Metcash was not simply a wholesaler because, in the protection of its IGA banner group, it was intimately involved in the retail activities of IGA stores. Those could not be divorced from Metcash's wholesaling activities. To do so would involve a significant degree of artificiality. 300 The primary judge also found (at [338]) that Franklins was essentially a supermarket retailing business and that Metcash was not acquiring the shares in Franklins so as to acquire its wholesale assets (which were largely warehouse and distribution operations carried out through third parties under contract). Rather, the object of Metcash's attention was the acquisition of the Franklins corporate stores, with a view to expanding the volume of Metcash's own wholesale sales. 301 The primary judge further found (at [339]-[341]) that the grocery industry was characterised by a high degree of vertical integration in the distribution supply chain. The major supermarket chains placed a very significant constraint on the capacity of independent retailers to increase price or decrease other services without the likely loss of business. That constraint also constrained the capacity of a wholesaler (specifically, Metcash) to increase its prices to independent retailers. 302 The primary judge noted (at [341]) that Metcash and Pick n Pay had contended for a national market for the supply of packaged groceries, fresh products, general merchandise and health, beauty and cosmetic products to the consuming public by way of integrated retail chains and independent wholesalers supplying independent grocery retailers. I interpolate that this alternative market definition accords with that adopted by the Tribunal in QIW at 266. The primary judge noted, however, that the Commission had not suggested that Metcash's acquisition would be likely to substantially lessen competition in such a market. 303 Ultimately, the primary judge was not persuaded that there was a separate wholesale market, as defined by the Commission. That conclusion was, of itself, determinative against the Commission's case for contravention of s 50(1) of the Act. No other market was relied on.

The Commission's grounds of appeal 304 In its notice of appeal, the Commission contended that the primary judge should have found that the market it had defined was the relevant market to determine whether the acquisition by Metcash was likely to lead to a substantial lessening of competition. In this connection it contended that the primary judge failed to apply the relevant principles identified in Davids Holdings to determine the product and functional dimensions of the market for s 50(1) purposes. 305 The Commission also contended that the primary judge erred in his identification and application of "the hypothetical monopolist test". 306 Further, the Commission contended that the primary judge should have found that: (a) because of the pricing differential between a significant number of independent retailers supplied by Metcash and the major supermarket chains, Metcash had a degree of pricing freedom that meant that the chains did not impose a close constraint on its wholesale pricing decisions; and (b) because Metcash was able to engage in sustained and significant differential pricing to independent retailers, it was not closely constrained by any indirect competition at the retail level by the major supermarket chains. 307 The Commission also challenged some more specific findings made by the primary judge. In that regard, it contended that the primary judge erred in finding that the threat of sale by independent retailers of their stores to major supermarket chains relevantly constrained Metcash's wholesale pricing decisions. It also contended that the primary judge erred in finding that Project Energise was as much a competitive response to Woolworths and Coles as it was to Franklins as an alternative wholesale supplier. 308 As argued, the Commission's case on appeal on the question of market definition was based on eight broad submissions. These were that the primary judge erred: (a) by failing to correctly identify and apply the hypothetical monopolist test; (b) by failing to correctly identify the product dimension of the market; (c) by failing to consider the evidence as to the functional dimension of the market; (d) by failing to recognise that many independent retailers have the discretion to set prices significantly above the major supermarket chains; (e) by failing to take account of the extent of pricing discretion by Metcash at the wholesale level; (f) by failing to appreciate the significance of Project Energise; (g) by failing to appreciate the significance of Project Ling Chi; and (h) by finding that store sales may be a close constraint. 309 As will be apparent, some of these submissions were co-extensive with a specific ground of appeal, as I have summarised them. Some were put forward by the Commission as being common to a number of the grounds of appeal. Most (namely, subparagraphs (c)-(h), and aspects of subparagraph (a) above), were directed to the primary focus of this part of the Commission's appeal, which was the primary judge's rejection of a separate wholesale market for the purpose of analysing the competitive effect of Metcash's acquisition.

The hypothetical monopolist test 310 The Commission raised what it said were five interrelated errors made by the primary judge in identifying and applying the hypothetical monopolist test. It submitted that the cumulative effect of these errors led the primary judge to find a significantly broader market than the market he should have found, had the hypothetical monopolist test been correctly applied. 311 I should say at the outset that, in my view, the ultimate significance of these errors, if they be errors, is questionable, having regard to the totality of the evidence before the primary judge and the fact that his Honour was being invited to apply the hypothetical monopolist test only as "an intellectual aid to focus the exercise". That exercise was to arrive at a definition of the market that best suited the identification and analysis of the effect or likely effect upon competition, if any, that Metcash's acquisition of the Franklins shares would have if it proceeded to completion. In undertaking that exercise it was necessary for his Honour to evaluate the competitive process involved in the supply of what can be described broadly, in the present case, as groceries, identifying the existence of, or potential to exercise, market power in that supply, and the existence of constraints or likely constraints on the exercise of that power. 312 As stated in Heydon JD, Trade Practices Law (Lawbook Co., subscription service) at [3.245] (update 115): The dimensions of a market are real, not theoretical. To define those dimensions the best evidence will come from the people who work in the market: the marketing managers and salesmen, the market analysts and researchers, the advertising account executives, the buyers or purchasing officers, the product designers and evaluators. Their records will establish the dimensions of the market; they will show the figures being kept at competitors' and customers' behaviour and the particular products being followed. They will show the potential customers whom salesmen are visiting, the suppliers whom purchasing officers regularly contact, products against which advertising is directed, the price movements of other suppliers which give rise to intra-corporate memoranda, the process by which products are bought, what buyers must seek in terms of quantities, delivery schedules, price flexibility, why accounts are won and lost. 313 In the present case, the hypothetical monopolist test was no more than an aid to arrive at the appropriate market definition. It was not an end in itself. 314 One of the errors identified by the Commission (namely, his Honour's observation that, in applying the hypothetical monopolist test, a "smaller and smaller" market must be postulated) can be dismissed immediately as being an inconsequential and obvious slip about how the test is to be applied. As I have noted, in application, the boundaries of a candidate market are expanded (not contracted) to include all sources of close substitutes that would defeat the increase. Thus all close substitutes are captured and form part of the market definition. Proceeding in this way, it is the smallest area, in terms of product identification and geographic space, over which the hypothetical monopolist can profitably impose the increase, that gives the boundaries of the market. Importantly for present purposes, and despite the slip to which I have referred, the primary judge went on to recognise this in his discussion of the test. 315 The other asserted errors can be dealt with relatively briefly. 316 First, the Commission submitted that, in applying the hypothetical monopolist test, the primary judge erred in not commencing with an overlap product supplied by the merger parties. This submission is really directed to the Commission's contention that the primary judge failed to correctly identify the product dimension of the market or, to put the matter in terms of his Honour's actual finding, that the primary judge erred in finding that the product dimension of the market should not be limited by reference to packaged groceries, as the Commission had defined them. 317 It should be noted that the primary judge expressly acknowledged that the hypothetical monopolist test should be applied to the wholesale price of packaged groceries when considering the product dimension of the market: see at [202]. The essence of the Commission's present submission appears to be that the primary judge should not only have started with its own proffered product definition as a candidate market but, on an application of the hypothetical monopolist test, been satisfied with that definition, such that his Honour should not have contemplated that the product dimension of the market should be expanded to include other products supplied by Metcash and Franklins as wholesalers. 318 The Commission's submission does not articulate why its own definition was, necessarily, the correct starting point, beyond stating that its product definition includes "must have" or "customer-critical" items without which independent retailers "cannot operate". That might be so, but the inclusive nature of the Commission's own definition underscores its inherent imprecision and, consequently, begs the question: what, factually, is the correct starting point? In any event, and perhaps more importantly for present purposes, it is not apparent, on the face of the reasons, that, in undertaking his evaluation, the primary judge did not use the Commission's product definition as his starting point. In short, I am not satisfied that the primary judge committed the conceptual error that the Commission has alleged. The Commission's submission simply rests in its own asserted correctness. 319 The real issue for this case, however, is not whether the Commission's proffered definition is the correct starting point, but whether it is the correct end point, and thus defines the boundaries of the product market with relative but nevertheless appropriate precision. Here the primary judge was persuaded on other evidence before him that the construction of the product dimension of the market for which the Commission contended was, in his Honour's evaluation, artificial. In my respectful view, that was a finding to which the primary judge was entitled to come on the evidence before him. I am not persuaded that his Honour erred in that view. 320 Secondly, the Commission submitted that, in applying the hypothetical monopolist test, the primary judge failed to recognise that the ultimate constraints on the pricing activities of a hypothetical monopolist do not mean that the firms that have that constraining influence, in this case the major supermarket chains, must necessarily be included in the same market. 321 In my view the primary judge did not err as the Commission has submitted. The primary judge's reasons (at [430]) make perfectly clear his Honour's appreciation that firms that impose indirect constraints are not necessarily included within the market, as a matter of market definition. However, the primary judge's finding was that the major supermarket chains imposed a much closer constraint on Metcash than Franklins. Based on that finding, the primary judge concluded that, for the purposes of the present case, the market should be defined in a way that included, and thus accommodated, those constraints. 322 Thirdly, the Commission submitted that the primary judge failed, in applying the hypothetical monopolist test, to consider the price that would apply in the absence of the merger, and apply the SSNIP to that price. In this connection the Commission submitted that the primary judge assumed that the prevailing prices charged by Metcash provided the correct standard and that his Honour failed to recognise that Metcash was already exercising a degree of market power in its pricing. In short, the Commission submitted that the primary judge committed the so-called "cellophane fallacy": United States v EI du Pont de Nemours & Company (1956) 351 US 377; see also the discussion in Seven Network Ltd v News Ltd (2009) 182 FCR 160 at [355]-[356]. 323 The significance of this submission, with reference to the "cellophane fallacy" in the context of the application of the hypothetical monopolist test, is unclear. One would naturally think that, by calling in aid the "cellophane fallacy" in this context, the Commission's submission was directed, once again, to challenging the correctness of the primary judge's conclusion about the product dimension of the market. However, the Commission's submission appears to have been intended to be directed to a quite different matter, namely whether the major supermarket chains constrained Metcash to the degree found by the primary judge. In other words, the gravamen of the Commission's submission appears to be directed to the correctness of the primary judge's conclusions about the functional dimensions of the market, a matter to which I shall come. 324 Be that as it may, the Commission has not, in my view, made good its submission, as a matter of fact. It is important to note that no quantitative analysis was undertaken by the Commission of Metcash's prices to determine whether they were monopoly prices. In the absence of such evidence it is difficult to see how the primary judge proceeded in error as the Commission has alleged. Indeed, the Commission's submission seems to be somewhat contrary to the position it adopted at trial, namely that Franklins was Metcash's closest competitive constraint in the defined territory and that the threat of Franklins entering into self-supply forced Metcash to drop prices under Project Energise: see at [335]. To this must be added, on the primary judge's findings, the fact that Metcash was even more constrained in its pricing decisions by the major supermarket chains. These considerations tell against the likelihood that the primary judge proceeded in error, merely assuming, as it were, that Metcash was not already exercising a degree of market power in its pricing. 325 Finally, the Commission submitted that, in applying the hypothetical monopolist test, the primary judge applied a SSNIP at the wrong functional level, and at a magnitude that was too large. In this connection the Commission submitted that the primary judge failed to consider whether a price increase by a hypothetical monopolist wholesale supplier could be profitable at the wholesale level. It also submitted that the primary judge should have applied a small percentage, such as a 1 to 2% SSNIP, to the overall wholesale price or a 5 to 10% SSNIP to its product margin. 326 Once again, this submission is directed, conceptually, to considerations that inform the product dimension of the market. I have already referred to the fact that the primary judge's evaluation of the product dimension of the market, and his Honour's rejection of the Commission's demarcation of that dimension, was influenced by other evidence. 327 It is also important to note that the context for the Commission's particular challenge in this regard is the primary judge's rejection (at [189]-[200]) of the Commission's submission that the hypothetical monopolist test should be applied in respect of wholesale profit margins or mark-ups as distinct from wholesale price. Although the Commission urged a similar position on appeal, it did not identify why the primary judge was in error in finding that the hypothetical monopolist test should be applied to the wholesale price of packaged groceries. For my part, I can discern no error in his Honour's conclusion in that regard. 328 The particular statement which the Commission criticised was his Honour's observation (at [202]) that, although the hypothetical monopolist test should be applied to the wholesale price of packaged groceries, the magnitude of the SSNIP, in order to satisfy the test, would need to cause an increase in the price of packaged groceries at the retail level by a real or noticeable amount. The primary judge went on to observe that there was no evidence that a hypothetical monopolist wholesaler of packaged groceries could increase prices profitably, such that the price of groceries at the retail level would increase by a significant amount. 329 The primary judge's challenged statement recognises that the SSNIP should be applied at the wholesale level. Thus, no error is indicated in this regard. However, what the primary judge's statement also raised, in a qualitative way, was the magnitude of the SSNIP. I have already referred to the fact that, at a theoretical level, a SSNIP of 5 to 10 % is usually considered. This is the level that the primary judge applied. The Commission's submission was that, if a SSNIP of 5 to 10% were imposed on the wholesale price in the present case, this would significantly raise the bar to show that the increase would be profitable. It also submitted that, at this level, the SSNIP would be likely to lead to a failure to detect the ability of a monopolist wholesaler to significantly increase the price it charges for its stage in the supply chain. Ultimately, however, these submissions were not developed on appeal in any way that would demonstrate their correctness or, more importantly, that the primary judge's conclusions on the question of market definition were wrong. In so far as the Commission's submissions strayed into the area of the quality and magnitude of the constraint on Metcash exercised by the major supermarket chains, this matter is best considered below in the context of the Commission's submissions that were more specifically directed to challenging the primary judge's conclusions with respect to the functional dimensions of the market. 330 It is sufficient for present purposes for me to say that the Commission's challenge to the way in which the primary judge qualitatively deployed the hypothetical monopolist test does not persuade me that, on that basis, his Honour's conclusions about the appropriate product and functional dimensions of the relevant market were wrong.

The product dimension of the market 331 I have already discussed the Commission's definition of the product dimension of the market and noted that the primary judge was persuaded on other evidence that the constriction of the product dimension of the market for which the Commission contended was, in his Honour's evaluation, artificial. 332 The primary judge found (at [184]) that Metcash supplied a broad range of products to retailers that included dry groceries, frozen goods, dairy products, fresh fruit and vegetables, delicatessen products and general merchandise. The primary judge found (at [14]) that Franklins also supplied a broad range of products to its corporate and franchise stores that included packaged groceries, health, beauty and cosmetic products, general merchandise and fresh produce. The effect of the Commission's case was that there was a separate market for the goods supplied by Metcash and Franklins at wholesale (including fresh fruit and vegetables and delicatessen products) that did not fall within the Commission's definition of packaged groceries. 333 On appeal, as it did at trial, the Commission submitted that fresh fruit and vegetables and delicatessen products are not substitutes for a range of packaged groceries. It submitted that there was no evidence that independent retailers could substitute these products for packaged groceries if the wholesale prices of packaged groceries increased, or that the hypothetical monopolist would be constrained from increasing its prices in respect of packaged groceries by the threat of its customers (independent retailers) switching to these products. 334 As I have noted above, the real issue here is the outer boundary of the product dimension of the market. The primary judge found (at [184]) that when retailers come to sell the broad range of products supplied at wholesale by both Metcash and Franklins, they seek to make a profit margin across the bundle of goods, not just those goods that fall within the Commission's product definition. The primary judge also found (at [186]) that there was evidence that when retailers obtain products from sources other than Metcash, the process of substitution is not confined to fresh produce, but extends to packaged groceries as well. 335 The primary judge was called upon to make an evaluative judgment about the product dimension of the market, taking into account the totality of the evidence before him. As I have said, his Honour's finding was one to which he was entitled to come on that evidence. In my view the Commission has not demonstrated that the primary judge's rejection of its market definition in this respect was an error, as opposed to simply a different conclusion on the evidence, on which reasonable minds could differ.

The functional dimension of the market 336 The Commission submitted that the primary judge erred in a number of respects when dealing with the functional dimension of the relevant market. 337 These submissions reflected the substance of the Commission's submissions at trial, which were to the following effect. The major supermarket chains operated as an uneven constraint on the pricing behaviour of independent retailers, many of whom exhibited a degree of relative pricing discretion. This meant that the major supermarket chains were not a close competitive constraint on the upstream activities of Metcash as a wholesaler. This was said to be reflected in the fact that Metcash itself exhibited a relative degree of pricing discretion. On the other hand, Metcash responded directly to the threat of Franklins as a wholesale supplier. This was illustrated by Metcash effectively cutting prices and locking in supply under Project Energise, and by taking retaliatory action under Project Ling Chi. These features showed that Franklins provided a greater constraint than the major supermarket chains on Metcash as a wholesaler. Accordingly, the correct framework to analyse the competitive effect of Metcash's acquisition was a wholesale market, and the primary judge was in error in concluding otherwise. 338 I turn now to consider the elements of the Commission's case on appeal in this regard.

Failure to consider functionally separate markets 339 The Commission submitted that the primary judge failed to consider that functionally separate wholesale operations indicated a functionally separate wholesale market. It referred to the fact that Metcash was not vertically integrated (despite its contractual relationship with IGA suppliers) and supplied only independent retailers, many of whom did not operate under the IGA banner and therefore did not receive the range of ancillary services offered by Metcash to IGA retailers. 340 In my view this submission cannot be sustained. It is plain that the primary judge did consider and take into account the existence and extent of transactions occurring at the wholesale level when dealing with this aspect of the Commission's case. There was, however, a significant body of other evidence before the primary judge which served to show, on his Honour's findings, that the true market behavioural constraints on Metcash came, to a great extent, from the major supermarket chains. 341 In this connection his Honour was strongly influenced by an appreciation that a firm supplying goods at the wholesale level may be constrained not only by other firms operating at that level but also by firms operating downstream, particularly where supply chains show significant vertical integration, principally, in this case, through the major supermarket chains. The primary judge was mindful of the captive production involved in that supply and of the strong competition at the retail level exerted by the major supermarket chains against independent retailers. 342 The primary judge (at [268]) observed: The demand of independent retailers for groceries at the wholesale level depends on the demands of their retail customers. Accordingly, where there is an increase in price at the wholesale level, vertically integrated firms that do not face that increase are able to take market share from non-integrated firms. Where there is an integration of wholesale and retail activities, the exercise of market power at either of the two upstream functional levels is closely connected with, and capable of affecting, competition downstream in the relevant chain. 343 Thus, far from ignoring the existence of transactions at the wholesale level, the primary judge gauged the significance of those transactions in the light of the captive production of the major supermarket chains.

The pricing discretion of independent retailers 344 The Commission submitted that the primary judge erred in finding (at [340]) that the major supermarket chains exerted a "very significant" constraint on independent retailers. It submitted that this finding was based on an earlier finding by the primary judge (at [240]) that independent retailers must match or be close to the major supermarket chains on price. It submitted that this earlier finding was, itself, based on no more than speculation and ignored evidence that many independent retailers supplied by Metcash did not price match or price closely to the major supermarket chains. The Commission submitted that the pricing evidence showed that the degree of constraint by the major supermarket chains on independent retailers was limited and uneven. It submitted that the primary judge incorrectly assumed very strong substitution by retail customers between independent retailers and major supermarket chains in response to differences in retail prices. 345 These submissions were made in the context of a broader submission that the primary judge failed to recognise that many independent retailers had the discretion to set prices significantly above the major supermarket chains. 346 To illustrate its submissions in this regard, the Commission pointed to the fact that, within the IGA banner group, 66% of the stores supplied be Metcash were IGA or IGA X-press stores; that the prices charged by IGA X-press stores were significantly higher than the prices charged by IGA stores; and that the prices charged by IGA stores were materially higher than the prices charged by Supa IGA stores. It submitted that the primary judge did not consider the evidence of operators of IGA X-press and IGA stores about the extent to which they priced above the major supermarket chains. 347 The Commission also sought to illustrate its submissions by pointing to an analysis it had performed in respect of "zone 60 pricing". It said that this analysis demonstrated that there were significant pricing disparities (greater than 2%) between the major supermarket chains (Woolworths in particular) and independent retailers supplied by Metcash. 348 There are a number of things to be said about these submissions. 349 First, the primary judge's finding about the degree of constraint exercised by the major supermarket chains on independent retailers related to the capacity of the independent retailers to increase price or decrease other services without the likely loss of business. In other words, his Honour's finding related to the constraint placed on independent retailers to compete on price and non-price factors; it was not simply a finding about price competition at the retail level. 350 In this connection the primary judge found (at [256]) that retailers unable to meet the major supermarket chains on price looked for other ways to remain competitive with them by emphasising other advantages of their offerings, including location, convenience, access to parking, breadth and depth of product range, community involvement and local connections, levels of customer service, trading hours, quality of fresh produce, and store layout and appearance. The primary judge saw these factors as being very important elements in the process of competition between the major supermarket chains and independent retailers. 351 The primary judge also found (at [259]) that Metcash provided non-price services to independent retailers to assist them in competing with the major supermarket chains. These services included providing a private label range (the Black & Gold product range), negotiating schemes whereby suppliers fund donations by retailers to local communities, providing loans to retailers to finance store refurbishments, and funding marketing activity to promote the IGA brand with a view to attracting custom away from the major supermarket chains. 352 In light of the primary judge's findings about non-price competition, the fact that pricing differences existed between retailers does not demonstrate error in the primary judge's conclusion regarding the degree of constraint exercised by the major supermarket chains on those retailers. 353 Secondly, contrary to the Commission's submission, the primary judge did consider pricing differences between stores operating under the IGA banner in different formats. In that connection, however, the primary judge found (at [254]) that, nationally, sales to IGA X-press stores represented only 2-3% of the total volume of Metcash's sales. His Honour reasoned that information about those stores was, therefore, minimally significant to the question of constraint exercised by the major supermarket chains. In my view that reasoning does not reflect error. 354 Thirdly, the primary judge (at [250]-[254]) did consider the Commission's zone 60 pricing analysis, but identified a number of flaws in it. In that connection his Honour observed that the analysis did not show pricing requested by retailers who were not in the IGA banner group. Even then the analysis did not cover the pricing requested by approximately one third of Super IGA and IGA retailers in the defined territory. Further, the analysis did not take into account promotional pricing, which the primary judge found (at [253]) to be an important element of price competition. In that connection his Honour found that 40 to 50% of the volume of the IGA-Distribution division was sold "on promotion". Moreover, the Commission's analysis failed to take into account sales volume more generally. I have already referred, in that regard, to his Honour's finding that the IGA X-press stores represented only 2-3% percent of the volume of Metcash's sales. 355 In my view these were matters that the primary judge was entitled to take into account when considering the usefulness of the Commission's zone 60 pricing analysis and the weight to be given to it in all the circumstances. 356 One particular criticism challenged by the Commission was the primary judge's observation that the zone 60 pricing analysis did not take account of promotional pricing. The Commission submitted that there was no evidence to support a finding that the inclusion of promotional pricing would have affected its analysis. This submission, however, simply deflects attention from the substance of the primary judge's criticism, which was that conducting a zone 60 analysis without detailed consideration of the implications of promotional pricing was likely to lead to erroneous conclusions. This was because promotional pricing allowed independent retailers to engage in "investment buying". By this means retailers could acquire stock at discounted prices in quantities beyond their immediate needs, thus giving them the ability to charge prices on those goods below recommended standard shelf prices for periods that extended beyond the normal period for promotions. 357 The Commission also sought to challenge the correctness of some incidental and very generalised observations made by the primary judge when considering the Commission's zone 60 pricing analysis. For example, when observing that it did not follow that retailers who sought supply from Metcash at prices 2 or more percent higher than zone 60 prices were not closely constrained by vigorous competition, the primary judge referred to the possibility that some retailers might charge prices above the major supermarket chains because of higher costs rather than as an attempt to make supra-normal profits. His Honour also said that stores with a lower turnover will have higher unit costs and are likely, therefore, to need to charge higher prices to make a profit. 358 The Commission submitted that there was no evidence before the primary judge that smaller stores supplied by Metcash had higher costs. In this connection it relied on a report prepared for Metcash by PricewaterhouseCoopers which indicated that, as a percentage of revenue, IGA stores had a lower cost of doing business than Supa IGA stores, and that IGA X-press stores had a similar cost of doing business to Supa IGA stores. It also relied on that report to show, more generally, that IGA and IGA X-press stores generated higher retail margins than Supa IGA stores and that while 73% of Supa IGA stores were in a competitive area, only 40% of IGA stores were in such an area. 359 There is an immediate difficulty that confronts the Commission's submissions in this regard. The PricewaterhouseCoopers Report makes clear that it was based on incomplete data, the accuracy of which its authors could not, in any event, vouch for. The report noted that the need to extrapolate from store survey data to derive retailer EBIT rendered the analysis, in that regard, inherently inaccurate. The incompleteness of the data used in the report is reflected in the fact that only 38% of all stores responded to the survey, with only 32% of IGA stores and only 21% of IGA X-press stores responding. Significantly there were, for whatever reason, no responses from IGA X-press stores in New South Wales. 360 Thus the report provides a poor foundation from which to challenge the primary judge's very generalised observations about costs. Even then, the report says nothing about the position of retailers who were not within the IGA banner group. Furthermore, in making these observations, the primary judge was drawing attention to the fact that, in undertaking this particular pricing analysis, the Commission did not itself attempt to address the cost to Metcash of supplying the smaller stores, or the other fixed and variable costs faced by those stores compared with larger supermarkets. 361 In my view none of the matters to which the Commission pointed shows that the primary judge's findings about the nature and degree of the constraint placed upon independent retailers by the major supermarket chains were made in error. It was open to the primary judge to come to these conclusions based on the evidence before him, which included the evidence of independent retailers who spoke of their pricing policies. Those retailers also spoke of the need to compete not just on price. As the primary judge noted, Metcash and its retailers had a common interest in ensuring that the retailers remained competitive with the major supermarket chains so that Metcash was able to protect, and seek to build, its volume of sales.

Metcash's pricing discretion - Project Energise - Project Ling Chi 362 The Commission submitted that the primary judge erred by failing to recognise that Metcash enjoyed a degree of pricing discretion and that that freedom demonstrated that it did not face close constraint from the major supermarket chains. The Commission submitted that this pricing discretion was illustrated by a number of matters. 363 First, the Commission pointed to what it says were targeted price support programs that Metcash offered to those of its retailers that faced the most direct competition from the major supermarket chains at the retail level. One of those programs was called Project Lion. The primary judge found (at [312]) that this was a targeted support program that was offered by Metcash in 2009-2010 to IGA retailers facing direct competition from the major supermarket chains. However, the fact that the price support was, in some instances, targeted, illustrates no more than that Metcash made a business decision to provide particular support where it thought that that support was most needed in order to protect its own interests. In this regard, its own interests were best protected by ensuring that it could maintain and, if possible, enhance its volume of sales to retailers by ensuring that those retailers were price competitive with the major supermarket chains. If anything, the existence of price support programmes, targeted or otherwise, demonstrates, rather than denies, the reality of the constraint which the primary judge found the major supermarket chains exercised on Metcash. 364 Secondly, the Commission pointed to Metcash's imposition, over objection, of a five cent per carton increase in the service fee it charged to retailers, as illustrating the degree of its pricing discretion. The primary judge dealt extensively with the circumstances under which this increase came to be implemented: [49]-[65]. The primary judge's findings include the fact that Metcash had originally proposed a ten cent per carton increase to the service fee. The proposed increase was discussed at an executive management committee meeting of Metcash on 25 August 2006 in the context of providing a fund for rental subsidies to enable the IGA banner group access to major shopping centres and to provide subsidies (termed "price into product" subsidies) to meet competition. It also seems that part of the increase was to recover increased costs incurred by Metcash, particularly over the preceding 18 months. There had been no increase to the service fee since 2001. 365 The primary judge's findings in respect of the imposition of this increase certainly show that there was significant and widespread resistance to it. This, of itself, is not surprising. However, Metcash did not simply impose the increase unilaterally. It consulted with the national and State boards of IGA retailers. Faced with the opposition it found, Metcash finally resolved to impose a smaller increase of five cents per carton, with effect from 1 November 2006. The primary judge found (at [264]) that this was the first increase in Metcash's service fee in a number of years, and was a relatively modest increase. The primary judge reasoned that these circumstances supported, rather than gainsaid, Metcash's contention that it was constrained by reason of the vigorous downstream competition between independent retailers and the major supermarket chains. In my view that was a conclusion to which his Honour was entitled to come on the evidence. 366 Thirdly, the Commission pointed to Project Energise as exemplifying the existence of Metcash's pricing discretion. As I have already noted, the primary judge found that Project Energise was as much a response to the major supermarket chains as it was to Franklins. The Commission submitted that the primary judge erred in that conclusion. 367 In that connection the Commission submitted that Project Energise was simply Metcash's response to the threat of entry by Franklins as a wholesaler. It represented no more than a price cut to retailers in the face of that threat. The Commission pointed to the fact that Project Energise was only introduced in New South Wales and the Australian Capital Territory and submitted that its offering was more generous than the offering under an unrelated Victorian program. It also pointed to the fact that even more favourable terms were offered to Supabarn and some other retailers. It submitted that Project Energise was a "natural experiment" demonstrating actual market behaviour in response to the threat of Franklins' emergence as an alternative supplier to independent retailers and thus provided persuasive evidence of the correct framework (namely, a wholesale market) in which to analyse Metcash's acquisition: Liquorland at [444]. 368 Project Energise was a program available to all independent retailers within the IGA banner group, under which those retailers were not only given incentives to achieve targets, in part referable to purchases made from Metcash, but also to develop five year business plans detailing refurbishment strategies. These aspects of the program reflected a concern on the part of Metcash to ensure that its sales volumes were protected, if not enhanced, by retailers being able to provide improved offerings by way of price and non-price forms of competition with other retailers not supplied by Metcash. Given the primary judge's finding that independent retailers regarded the major supermarket chains as their competitors, and the fact that Woolworths and Coles had 80% of the national grocery market share, it can be seen that the focus of these incentives was the enhancement of the ability of retailers supplied by Metcash to compete with the major supermarket chains. Indeed, one of the project's stated aims was to grow the market share of the New South Wales IGA banner group to 12%. 369 The primary judge found (at [321]) that Metcash did in fact receive tangible benefits from the project. Even though Metcash had lost the Franklins business, it gained, through this project, significant new business, reflected in enhanced profits. Furthermore, Project Energise conferred on Metcash a right of first refusal to purchase the stores of participating retailers. There was evidence that this element of the project was, in part, responsive to the threat of Woolworths and Coles purchasing the stores of independent retailers and thus foreclosing supply that would otherwise be provided by Metcash. 370 In my view the primary judge did not err in finding that Project Energise was as much a response to the major supermarket chains as it was to Franklins. This conclusion was supported by the evidence and open to be made by his Honour. 371 Fourthly, the Commission pointed to Project Ling Chi as a targeted response to Franklins' wholesaling activities. It submitted that the primary judge erred by concluding that Project Ling Chi did not bear significantly on the degree of competitive constraint imposed by Franklins on Metcash. 372 It may be accepted that Project Ling Chi was a response to Franklins as a competitor. The primary judge did not think otherwise. But the significance of this project, in all the circumstances, was really a matter of weight for his Honour. 373 Project Ling Chi was planned as a formalised ten month campaign with a support budget of $1 million. Its aim was to attack Franklins' wholesale operations by means such as placing pressure, in various ways, on Franklins' supplier of logistic services as well as the suppliers of its private label products, who were also suppliers of such services and similar products to Metcash, and by spreading false information about Franklins. 374 The primary judge found that there was no evidence about how much Metcash did in fact spend on this campaign. His Honour reasoned that even if Metcash did spend $1 million in implementing it, that amount was fairly minimal in the scheme of things. There was also evidence before the primary judge that, rather than lasting the proposed 10 months, the project lasted only six to eight weeks. 375 In my view the primary judge's assessment of the relative significance of that project, in light of all the other factors and considerations to which his Honour referred on the question of constraints, does not reflect error.

Store sales as a constraint 376 The Commission submitted that the primary judge erred in finding (at [279]) that there was competition between Metcash and the major supermarket chains for the retention and acquisition of retail stores and sites. The Commission submitted that these acquisitions occurred in a different market and, at best, could only constitute evidence of an ultimate constraint on Metcash. It submitted that the primary judge confused exit from a market with a competitive constraint within a market. It also submitted that many independent retailers' stores were likely to be too small to be of interest to the major supermarket chains. 377 In my view these submissions do not really engage with the significance of the primary judge's finding. It was clearly in Metcash's interests to ensure that the major supermarket chains did not acquire the stores of independent retailers supplied by Metcash, where they thought it desirable to do so. Plainly enough, any such acquisition would represent a loss of sales to Metcash. This was the motivation for Metcash implementing the Barons Strategy of acquiring a minority equity interest in large successful independent retailers as a means of preventing the erosion of the volume of its sales through the expansion of the major supermarket chains beyond large scale format stores into medium and smaller stores. In my view this was a relevant consideration for the primary judge to take into account when identifying the constraints placed upon Metcash. In my view the Commission has not demonstrated error in his Honour's conclusion.

Conclusion on functional dimension 378 The Commission's submissions on appeal with respect to this aspect of the question of market definition really reflect the substance of its submissions at trial. Its focus on appeal was not to challenge many of the primary judge's findings of primary fact but, rather, to challenge the correctness of his Honour's conclusions drawn from those facts. In my view the Commission has not demonstrated that the primary judge's conclusions about the existence and extent of the constraints placed upon independent retailers and Metcash by the major supermarket chains were not open to be made on the facts as the primary judge found them to be.

Conclusion 379 In order to identify the relevant market for the purpose of the case before him, the primary judge was called upon to make an evaluative judgment that was informed not only by his Honour's findings of primary fact but also by principles, elucidated in a substantial body of case law, relating to the identification of markets for competition law purposes. In the main, those principles are not contentious. 380 The Commission's overarching contention was that the primary judge failed to apply the relevant principles identified in Davids Holdings to determine the product and functional dimensions of the market for s 50(1) purposes. It is clear from the primary judge's reasons (at [204]) that, at trial, the Commission relied on Davids Holdings to advance a finding that there were clearly separate and distinct wholesale and retail transactions relating to packaged groceries that were of significance and indicated separate functional markets. 381 In its appeal the Commission, although acknowledging that much depends on the facts as found, really sought a parallel application to the present case of the ultimate findings in Davids Holdings about the relevant market. 382 In my view the Commission has not demonstrated that the primary judge proceeded on a wrong principle when considering the question of market definition in the present case. The primary judge's prefatory remarks at [150]-[165] and [174]-[175] show that his Honour was seized of the relevant principles. Furthermore, the primary judge approached the task of defining the market by taking account of commercial reality, not simply economic theory. His Honour proceeded on the basis that the economic concept of a market must be applied in a practical way to accommodate the concerns of the Act with those of business and commerce. In my respectful view, this was the correct approach to take. The parties did not suggest otherwise. 383 It is plain from the primary judge's reasons that his Honour had regard to a considerable body of evidence that was called on the question of market definition. It is equally plain that his Honour undertook a detailed and systematic analysis of that evidence in reaching his conclusions about the market definition proffered by the Commission. For the reasons I have indicated, those conclusions were, in my view, open to the primary judge. 384 Critical to the primary judge's conclusions in this regard were his findings concerning the difference in the constraint imposed on Metcash by the major supermarket chains compared with Franklins. Based on the evidence before him, the primary judge considered that the major supermarket chains imposed a much more powerful constraint than Franklins on Metcash's market behaviour. In light of that fact, and the significant degree of vertical integration in the supply of groceries, it was open to the primary judge to view the appropriate market, for the purposes of determining the s 50(1) question before him, as one incorporating the activities of the self-supplying chains. 385 Having said this, a case could be made, on the evidence, that the market of interest was a wholesale market, with the major supermarket chains providing an indirect constraint on the participants in that market. As I have noted, the primary judge was alive to this possibility but rejected it. His Honour considered that, in the present case, the separation of wholesaling and retailing functions tended to confuse the analysis and that, in light of the more powerful constraint imposed by the major supermarket chains on Metcash, compared with the lesser constraint imposed by Franklins, it was not possible to determine the competition consequences of Metcash's acquisition without taking into account the constraints from the major supermarket chains as market participants. I am not persuaded that, in the particular circumstances of the present case, his Honour erred in coming to that ultimate conclusion and, thus, in rejecting the existence of a separate wholesale market.

Effect on competition 386 Having rejected the Commission's market definition, and its counterfactual case, the primary judge concluded, correctly, that it was not necessary to consider whether, as a real chance, competition would be substantially lessened by Metcash's acquisition. The primary judge nevertheless summarised the competing contentions of the parties in respect of the factors listed in s 50(3) of the Act as mandatory factors to be taken into account in applying the competition test in s 50(1). His Honour concluded (at [460]) that it was quite likely that the acquisition of Franklins by Metcash would strengthen the capacity of independent retailers operating under the IGA banner to compete more vigorously with the major supermarket chains. Moreover, in light of the speculative possibility that the KKKL consortium would ever be able to make an offer that would be accepted by Pick n Pay, the primary judge was not persuaded that the Metcash acquisition of Franklins would substantially lessen competition in the market propounded by the Commission. 387 The Commission challenged the primary judge's ultimate assessment in that regard. 388 First, it contended that the primary judge failed to consider the potential for close competition in the pleaded market if Metcash's acquisition did not proceed. It also submitted that the primary judge erred in finding that, given the very small level of involvement of Franklins as a wholesaler in the pleaded market, market concentration would not markedly increase, irrespective of the acquisition. The Commission submitted that, in each case, the primary judge failed to consider the extent to which the Franklins wholesale assets would be utilised as potential competition to Metcash in a scenario where the acquisition did not proceed. 389 In my view those submissions cannot be sustained. The primary judge was called upon to consider the case on the basis on which it was put to the Court, not on some hypothetical basis that was not advanced by any of the parties, and in particular by the Commission as the moving party. The only possible acquirer of the Franklins wholesale assets advanced by the Commission was the KKKL consortium. The primary judge rejected that possibility as speculative. Having done so, it was not incumbent on the primary judge to conjure other possibilities not advanced by the parties which, one can assume, could only stand as even more remote possibilities. The Commission's recourse to the potential for close competition, in those circumstances, is illusory. 390 Secondly, the Commission contended that the primary judge erred in not finding that barriers to entry in the pleaded market were high. In fact, the primary judge made no findings about the height of barriers to entry. In my view it is far from clear that his Honour's failure to make the finding for which the Commission contends could be, in the circumstances, an appealable error: Sydneywide Distributors Pty Ltd v Red Bull Australia Pty Ltd (2002) 55 IPR 354 at [4]. It was no more than a failure to make a specific finding that would have been, at best, a subsidiary finding on a question that was, in light of his Honour's other conclusions, merely hypothetical. In any event, his Honour's failure (if it be such) to make that specific finding was completely without consequence in light of his rejection of the market propounded by the Commission as well as its counterfactual case. 391 Thirdly, the Commission contended that, in considering the effect of the Metcash acquisition on competition, the primary judge erred in failing to take into account the potential for Metcash to increase profit margins or prices significantly and sustainably. Once again, this submission cannot be sustained. It is simply a repetition of the Commission's case that the degree of constraint imposed by the major supermarket chains was limited and uneven. The primary judge found (at [271]) that, because of the constraint imposed on Metcash by the major supermarket chains, there was no scope for it to raise its wholesale prices and obtain monopoly profits.

Disposition 392 In my view the appeal should be dismissed, with costs. I certify that the preceding three hundred and one (301) numbered paragraphs are a true copy of the Reasons for Judgment herein of the Honourable Justice Yates.