FEDERAL COURT OF AUSTRALIA

Australian Competition and Consumer Commission v P T Garuda Indonesia Ltd [2016] FCAFC 42

ORDERS

THE COURT ORDERS THAT:

1.    By 24 March 2016, the parties file a minute of proposed orders to give effect to these reasons.

Note:    Entry of orders is dealt with in Rule 39.32 of the Federal Court Rules 2011.

ORDERS

THE COURT ORDERS THAT:

1.    By 24 March 2016, the parties file a minute of proposed orders to give effect to these reasons.

Note:    Entry of orders is dealt with in Rule 39.32 of the Federal Court Rules 2011.

REASONS FOR JUDGMENT

DOWSETT AND EDELMAN JJ:

INTRODUCTION

1    This is an appeal from a trial which ran for fifty-seven days. The issues were not complex but the parties raised every imaginable point and tendered huge volumes of material. The central point raised by the Australian Competition and Consumer Commission (the “Commission”) on the appeal was a very short one: what is the meaning of a market “in Australia”? But the respondent airlines to the appeal continued a scorched earth policy. The notice of contention filed by P T Garuda Indonesia Ltd (“Garuda”) raised 94 issues. And almost the entire record of the trial proceedings was electronically reproduced on appeal. A handful of issues in the notices of contention were abandoned, but there were others which were not abandoned but about which no submissions were made. Some of the other issues raised in the notices of contention depended upon factual findings by the primary Judge about which we were referred to only a snapshot of the evidence. This introduction focusses upon the main issue on the appeal. As we explain, we would allow the appeal on that issue.

2    One matter in Garuda’s notice of contention was conceded, although it is unclear whether it will have any substantial effect, and another (concerning limitation) may require further submissions in light of these reasons. Otherwise, we reject the issues raised in contention by the respondent airlines.

3    At first instance in these proceedings, the central issue arose in the context that the Commission sought to establish that Air New Zealand Ltd (“Air NZ”) and Garuda had engaged in collusive behaviour. The Commission asserted that Air NZ and Garuda’s collusive behaviour consisted of fixing surcharges and fees on the carriage of air cargo from Hong Kong, Singapore and Indonesia into Australia. That conduct was said to be unlawful because it was contrary to s 45 (read with s 45A) of the Trade Practices Act 1974 (Cth) (the “Trade Practices Act”) (as it was named before being rebadged as the Competition and Consumer Act 2010 (Cth)). The primary Judge rejected that case. The Commission now appeals against that decision.

4    The primary Judge found that the markets for airborne cargo out of Hong Kong, Singapore and Indonesia were not markets “in Australia” within s 4E of the Trade Practices Act. At some stage a party or freight forwarder wishing to send cargo to Australia by air would have to choose the airline which would carry that cargo. The market in question was that in which air cargo services were supplied and acquired. His Honour concluded that the market was located at the point at which the choice of airline (a “switching decision”) took effect. That point was, in his Honour’s view, the point at which the cargo was delivered to the airline for carriage.

5    For the reasons below, and with genuine respect to his Honour’s meticulous judgment, we would allow the appeal on this point. The question to be asked is not whether a switching decision is given effect in Australia but whether a market is in Australia. All aspects of the market are relevant in determining whether it is in Australia. For instance, the presence of importers (customers) in Australia is not, as the primary Judge concluded at [263]-[264], irrelevant to the determination of whether the market is in Australia.

6    Both the text and purpose of s 4E require identification of whether a market is in Australia by (i) identification of the market; and (ii) location of that market. It has been customary to identify the market by reference to its “product dimension”, “geographic dimension”, and “functional dimension”. These dimensions are not independent of one another, nor are they unrelated to the location of the market. This is particularly so because the “market” concept does not refer to only a physical marketplace or market square where buyers come to purchase from sellers.

7    Ultimately, the determination of whether a market is “in Australia” is an evaluative exercise, which should not exclude any aspect of the market from consideration. In this case, Air NZ and Garuda supplied a suite of air cargo services to each port in Australia, commencing the provision of those services outside Australia. But (i) the suite of services they provided included important components which were provided in Australia; (ii) the services were marketed and ultimately supplied to customers, including significant customers in Australia; and (iii) there were Australian barriers to entry into the market. Wherever else the market might also have been located, the market was “in Australia”. This conclusion is based on the legislative text of s 4E of the Trade Practices Act when read with ss 45 and 45A. It is a conclusion which is consistent with the purpose of s 4E and the overarching purpose of the Trade Practices Act, being “to enhance the welfare of Australians through the promotion of competition and fair trading and provision for consumer protection”. It is also consistent with Australian authorities to which we refer later in these reasons. Those authorities emphasise matters other than the physical location of a supplier, or where any substitution is given effect, as relevant factors in the identification of the market. It is also consistent with the conclusions which have been reached upon similar fact patterns in New Zealand and in Europe.

THE COMMISSION’S APPEAL

The “market in Australia” issue

The Commission’s pleaded case

8    The Commission’s pleaded case against Air NZ and Garuda was that each airline contravened ss 45(2)(a)(ii) and 45(2)(b)(ii) of the Trade Practices Act in various ways. Alternatively, the Commission said that each contravention occurred, “only insofar as the provisions of the pleaded arrangements and understandings applied to the supply of air freight services on direct and indirect services to and from Australia, and on all substitutes for those services or any sector of those services”. Although the air freight services supplied by Air NZ and Garuda involved different points of origin (Hong Kong, Singapore and Indonesia), it was common ground in this appeal that the application of legal principle would lead to the same result in both cases (app ts 125).

The relevant provisions of the Trade Practices Act

9    At the relevant times, s 45(2) of the Trade Practices Act provided as follows:

45.    Contracts, arrangements or understandings that restrict dealings or affect competition

(2)    A corporation shall not:

(a)    make a contract or arrangement, or arrive at an understanding, if:

(i)    the proposed contract, arrangement or understanding contains an exclusionary provision; or

(ii)    a provision of the proposed contract, arrangement or understanding has the purpose, or would have or be likely to have the effect, of substantially lessening competition; or

(b)    give effect to a provision of a contract, arrangement or understanding, whether the contract or arrangement was made, or the understanding was arrived at, before or after the commencement of this section, if that provision:

(i)    is an exclusionary provision; or

(ii)    has the purpose, or has or is likely to have the effect, of substantially lessening competition.

10    In its case against Air NZ and Garuda concerning ss 45(2)(a)(ii) and 45(2)(b)(ii), the Commission relied upon the deeming provision in s 45A(1):

45A.    Contracts, arrangements or understandings in relation to prices

(1) Without limiting the generality of section 45, a provision of a contract, arrangement or understanding, or of a proposed contract, arrangement or understanding, shall be deemed for the purposes of that section to have the purpose, or to have or to be likely to have the effect, of substantially lessening competition if the provision has the purpose, or has or is likely to have the effect, as the case may be, of fixing, controlling or maintaining, or providing for the fixing, controlling or maintaining of, the price for, or a discount, allowance, rebate or credit in relation to, goods or services supplied or acquired or to be supplied or acquired by the parties to the contract, arrangement or understanding or the proposed parties to the proposed contract, arrangement or understanding, or by any of them, or by any bodies corporate that are related to any of them, in competition with each other.

11    The effect of s 45A is to deem a provision of a contract, arrangement or understanding (or one that is proposed) to satisfy the s 45 requirement (purpose, effect, likely effect etc) in relation to substantially lessening competition. Relevantly to this case, the deeming will occur where the provision has the purpose, effect, or is likely to have the effect of, “fixing, controlling or maintaining … the price for … goods or services supplied … by the parties to the contract, arrangement or understanding … in competition with each other”.

12    The Commission fastened upon the words of s 45A “in competition with each other” and then directed attention to the definition of “competition” in s 45(3):

For the purposes of this section and section 45A, competition, in relation to a provision of a contract, arrangement or understanding or of a proposed contract, arrangement or understanding, means competition in any market in which a corporation that is a party to the contract, arrangement or understanding or would be a party to the proposed contract, arrangement or understanding, or any body corporate related to such a corporation, supplies or acquires, or is likely to supply or acquire, goods or services or would, but for the provision, supply or acquire, or be likely to supply or acquire, goods or services.

13    The reference to “competition in any market” then invites attention to the definition of “market” in s 4E which we consider in detail below. It provides that the term “market” means a “market in Australia”.

14    Air NZ and Garuda submitted that the Commission was required to prove that they had engaged in the relevant conduct including “fixing, controlling or maintaining” the prices of their air cargo services. The Commission did not dispute that the definition in s 4E was engaged. In particular, the Commission did not submit that the words “in competition with each other” in s 45A(1) did not engage s 4E because those words concern the competition between the parties to the contract, arrangement or understanding rather than competition in relation to a provision of a contract, arrangement or understanding. Such a submission would have required attention to be given to whether the deeming provision could have been intended to travel beyond the geographical restriction upon the concept of market in s 45, or whether ss 45 or 45A had provided for a “contrary intention” within s 4E. It suffices to proceed on the basis, which was common ground, that the Commission was required to prove that the relevant conduct occurred in a “market in Australia”.

The primary Judge’s reasons, grounds of appeal and notices of contention

15    In Pt 4 of his judgment, the primary Judge considered the identification of the market, under the heading, “Was there a market in Australia?”. His Honour’s detailed reasons for concluding that there was no relevant market in Australia focus on cargo services out of Hong Kong, upon the basis that the relevant circumstances are the same as those in Singapore and Indonesia. His Honour noted that the respondent airlines argued for a narrow construction of s 4E, asserting that such an approach would observe the requirements of international comity. As we consider that the meaning of s 4E is clear, that consideration has no real weight with respect to the construction question. However we do not see that our approach to s 4E offends against any aspect of international comity.

16    His Honour also noted that the operation of the requirement that a market be in Australia had not been authoritatively settled, although some cases had suggested that the whole of a market need not be in Australia. His Honour observed that these cases may not authoritatively state the law in this regard. In Emirates v Australian Competition and Consumer Commission [2009] FCA 312; (2009) 255 ALR 35 at 50-53 [55]-[74] Middleton J discussed the question, having regard to earlier decisions of Hill and Lindgren JJ. That discussion, and his Honour’s conclusions closely resemble our own approach to s 4E.

17    At [212]-[219], the primary Judge considered the content of the relevant market or “field of rivalry”. At [212], his Honour’s conclusion was that, “In this field or area substitution in response to competitive pressures occurs”. His Honour also observed that:

Movements in price or the standard of goods or services which take place outside the field or area will induce no significant competitive consequences inside it.

This proposition tacitly assumes that relevant movements in price or the standard of goods and services supplied are located within the relevant market. We agree with that proposition. However it seems to be inconsistent with his Honour’s conclusions concerning the location of a market.

18    His Honour then discussed substitution, observing that it was basic to the process of market definition, and that it was necessary to consider both demand side and supply side substitution. He next asked, “How strong must the substitution effect be?”. At [214] his Honour considered that it must be “strong” and that:

Underlying these notions is the idea that the existence of substitutes operates to restrain the market power of those who are in the market.

19    At [217] the primary Judge said:

The product dimension is the ‘what’ question; the geographical dimension is the ‘where’ question; and the functional dimension is concerned with the location of market participants along the supply chain.

20    Under the heading, “The product dimension”, his Honour considered the following matters:

(1)    the relevant routes;

(2)    whether mail was included in the relevant markets;

(3)    whether chartered flights were included in those markets; and

(4)    whether the use of integrators was included in those relevant markets.

We need not consider any of those matters.

21    At [252] his Honour said:

I conclude that the product dimension for each market consisted of the services of flying cargo from Hong Kong to individual ports in Australia. None of these markets included mail carriers, charter flight operators or integrators.

22    At [253]-[265] the primary Judge considered the geographical dimension, by first examining the services provided by the airlines, namely:

(1)    transport services, including special handling requirements, timetabling and also whether or not delivery was by direct or indirect flight;

(2)    ground handling services at points of origin and destination; and

(3)    enquiry services at airports, noting that the airlines might supply the services themselves, or contract them out to third parties, possibly using an airline based at the airport in question.

His Honour observed that each of these services had a geographical element. Payment was generally made at the point of origin, that is, Hong Kong. Payment at the destination was rare and therefore not included in the analysis.

23    Concerning the identification of market participants, his Honour said that the primary issue was whether or not the market included as participants, only the airlines and freight forwarders, or also included some shippers. This issue involved two aspects, namely:

(1)    how persons might choose to switch patronage in the face of a small but significant and non-transitory increase in price (“SSNIP”) in a relevant market; and

(2)    who made the switching decisions, and where they were made.

24    His Honour explained the SSNIP at [220]. His explanation of the concept was not in dispute. In 1977, Professor (now Justice) Breyer summarised an early version of what is now the SSNIP test as “the notion of drawing a circle around a set of sales. Within the circle one should find sales with the following characteristic: were all those sales in the hands of a single seller, that seller would have the power to raise price significantly above the competitive level [and maximise profits]”: Breyer S, “Five Questions About Australian Antitrust Law” (1977) 51 Australian Law Journal 28, 34. The SSNIP approach has been traced to the United States Justice Department 1892 Merger Guidelines, based on the work of Mr Werden. It has not been without controversy: Stigler GJ and Sherwin RA, “The Extent of the Market” (1985) 28(3) Journal of Law and Economics 555; Harris RG and Jorde TM, “Market Definition in the Merger Guidelines: Implications for Antitrust Enforcement” (1983) 71(2) California Law Review 464 cited in Re Fortescue Metals Group Ltd [2010] ACompT 2; (2010) 271 ALR 256, 421 [1033] (Full Tribunal).

25    Concerning the first aspect at (1) above, the respondent airlines asserted that only airlines and freight forwarders based in Hong Kong were within the market, on the basis that, “any withdrawal [of] patronage from one airline to another had to occur necessarily in Hong Kong”. As to the second issue, the Commission submitted that in the case of large importers to Australia, the choice of carrier might be made by the importer rather than by the freight forwarder. In such a case, the Commission contended, the switching decision occurred where the importer was located, which location might well be in Australia. Such importers might be characterised as market participants because of their capacity to impose discipline on the upstream air cargo market. His Honour accepted that in some cases, decisions as to choice of airline might be made by persons other than freight forwarders in Hong Kong, including large importers (who might be in Australia) or large exporters (who would generally not be in Australia). Such shippers might continue to use freight forwarders, notwithstanding the fact that the latter would not be involved in choosing relevant airlines. His Honour concluded at [263] that:

The market participants in the various markets into Australia therefore included air carriers, freight forwarders and some large importers in Australia and exporters in Hong Kong.

26    The primary Judge considered that although some importers might have been located in Australia, others might well have been located at international headquarters in Europe or, presumably, elsewhere in the world. His Honour said at [264]:

I accept that it is quite possible in the case of significant shippers that the actual decisions about importation, including when it arises, the issue of which carrier to use, need not occur at the origin.

His Honour went on to say:

Because it will be relevant when examining the functional aspects of the relevant markets it is to be observed, however, that regardless of where the event consisting of the subjective decision to switch from one airline to another might be made, the only place in which it could be given effect would be Hong Kong. The range of choices from amongst which a person might choose an airline to fly cargo from Hong Kong to Sydney is inherently limited to those firms having operations in Hong Kong. Even if there were a supply side substitute it would still have to be provided in Hong Kong.

27    Thus his Honour accepted that a decision as to choice of airline might be made somewhere other than Hong Kong, but considered that such a decision could only be given effect in Hong Kong. This conclusion led his Honour to the proposition that the only airlines which could be chosen were airlines having operations in Hong Kong. Having regard to the evidence concerning regulation of air services entering Australia, one might equally have said that it was necessary to use an airline “having operations in” the relevant destination ports as well as in Hong Kong. In our view that proposition was mandated by the geographical aspect of the product dimension. His Honour seems to have attributed significance to the airlines being present in Hong Kong and not to their necessary presence in Australia. This approach reflected his Honour’s view that a switching decision could only be made in Hong Kong.

28    The primary Judge seems not to have drawn any general conclusions as to the geographical dimension of the market, save for identifying the services provided and their geographical locations, and the market participants and their geographical locations. His Honour seems rather to have deferred further consideration of the matter until his consideration of the question as to whether the market was a market in Australia. On the evidence, and having regard to his Honour’s observations concerning it, we infer that the geographical dimensions were as follows:

(1)    the suppliers of air cargo services were airlines presently providing cargo services between Hong Kong and Australia;

(2)    the purchasers of such services were freight forwarders in Hong Kong, some large importers in Australia and exporters in Hong Kong; and

(3)    the services were provided in Hong Kong, at the relevant ports in Australia and on the routes flown between those locations.

29    As we understand the respondent airlines’ submissions, they would exclude the third proposition from any consideration of the geographical dimension on the basis that it adds nothing to the economic analysis. However, in this case, geographical considerations are closely linked to the other dimensions. The product dimension involves the delivery of goods in Australia from Hong Kong. Similarly, the functional dimension reflects geographical considerations. Geographical considerations are also concerned with the right to fly into, out of and over various places. These issues loom large in this case. That geographical considerations arise in connection with the product and functional dimensions demonstrates that the consideration of each dimension cannot be undertaken in isolation.

30    His Honour then turned to the functional dimension. In effect, his Honour sought to identify those entities in the supply chain, between manufacturer and ultimate consumer, which might be sufficiently able to influence switching behaviour in connection with the supply of the relevant cargo services. He identified the “vertical structure” of the industry as:

(1)    Consignor;

(2)    Origin freight forwarder;

(3)    Airline;

(4)    Destination freight forwarder; and

(5)    Consignee.

His Honour recognised that either the consignor (an exporter of goods) or the consignee (an importer of goods) might be the instigator of the shipment, and that the services provided by freight forwarders were necessary accompaniments to the services provided by the airlines.

31    The primary Judge noted that the “nature” of consignors and consignees might “vary significantly”. Some might be involved in single exportation or importation transactions, involving individual chattels, such persons having only transitory connection with the international air cargo system. Others might be “significant” exporters and importers of chattels in size and volume. The evidence indicated that airlines, in general, regarded significant importers and exporters as targets for marketing activities, and as the ultimate source of business. His Honour considered that it was obvious that airlines would compete to obtain volumes of cargo directly from large shippers. At [287] his Honour concluded that:

[A]cross the Asia Pacific area the airlines recognised that shippers had demand for capacity. Indeed, they actively followed the position of shippers, recognising that these were the economic foundations of the market.

32    At [309] his Honour said:

Accordingly, I draw the following conclusions about the functional dimensions of the market:

(a)    the participants in the relevant markets were airlines, freight forwarders and shippers (be they exporters at origin or importers at destination) whose cargo volume was sufficiently significant for the airlines to be commercially motivated to pursue it;

(b)    shippers of that kind often (but not always) made decisions about which airlines they would use. Where the shipper was an importer in Australia this decision was likely to be made in Australia;

(c)    shippers of that kind continued (aside from the situation of integrators) to use freight forwarders who provided an indispensable set of services for dealing with the ancillary transport issues which the airlines themselves would not deal with. Relationships erected in the case of shippers of this kind were often tripartite. In some cases the tripartite nature of what was taking place was consummated with a contract but this was not a necessary nor even particularly common feature;

(d)    shippers of that kind, wherever located were therefore capable, at least in theory, of operating as a constraint on airlines’ cargo rates because of their ability, again in theory, to switch to alternate sources of supply and to outflank any exercises of market power at the relevant origin airport; and

(e)    smaller shippers who had no view about which airline to use and who left matters entirely to their freight forwarders were not participants in any of these markets.

33    The primary Judge then addressed the specific question of “Market in Australia?” At [311] his Honour identified the economic issues for determination as:

(a)    whether the fact that the airlines competed against each other in Australia in the provision of:

(i)    carriage through Australian airspace;

(ii)    ground handling services in Australia; and

(iii)    handling enquiries about lost and damaged cargo in Australia;

was sufficient to locate the markets in Australia;

(b)    whether the fact that the source of some of the demand for the services was ultimately in Australia was sufficient to locate part of each market in Australia;

(c)    whether the market in Hong Kong was constrained by the abilities of importers in downstream markets in Australia to switch to alternate sources of supply, and if so, whether it was appropriate to characterise the downstream markets as part of the upstream market; and

(d)    whether, in light of the foregoing, the market was in Australia.

34    Issues (a) and (b) are framed so as to suggest that the question is whether the relevant factor, alone, would be sufficient to locate the markets in Australia. We note that issue (a) is distinguished from issue (b) in that the former addresses the location of the markets, whilst the latter addresses the locations of parts of the markets.

35    His Honour dealt with issues (a) and (b) together, under the heading “Source of demand in Australia”. At [313]-[314], his Honour said:

313    As I have already said I accept that a separate market for air cargo services existed for the carriage of cargo between Hong Kong and each airport in Australia. Part of the service provided was provided in Australia in the form of carriage through Australian airspace, ground handling services at destination airports and the service of handling enquiries about lost and damaged cargo. There is no doubt that the airlines competed against each other in providing these services and that the competition physically took place in Australia. Further there were substantial importers in Australia whose custom the airlines tousled to obtain.

314    Although the contracts of carriage were entered into in Hong Kong by the freight forwarders with each airline, as a practical matter, substantial importers in Australia had the capacity to influence or even direct the decision as to which airline was to be used. On the other hand, less substantial importers had no such influence.

36    At first instance the Commission submitted that the area of close competition or the field of rivalry extended to Australia where destination services were provided, “in a competitive environment”. In contrast, the respondent airlines submitted that the relevant activity within a market was substitution, the market being the area in which sellers of the product operate, and to which purchasers may practicably turn for such product.

37    The primary Judge considered that the central thrust of “QCMA” concerned substitution, and “through that prism”, the present problem was to be analysed. The reference to “QCMA” is to the decision of the Trade Practices Tribunal in Re Queensland Co-operative Milling Association Ltd (1976) 8 ALR 481. Below, we discuss that decision in some detail.

38    His Honour had previously rejected the existence of any supply side substitutes. With respect to demand side substitution, he considered that the choice would have to be from amongst airlines which had a presence in Hong Kong where possession of the cargo would be transferred. This view reflects the proposition that such transfer for carriage to an Australian port could not occur at any place other than Hong Kong. One might also say that because the purpose for which the service is acquired is delivery of the cargo to an identified port in Australia, that service could not be delivered anywhere other than at that port, so that service could only be supplied by an airline which could fly into that port. The Commission submitted that the relevant service was provided, not only at Hong Kong, but also along all points on the route between Hong Kong and the relevant Australian airport including, we infer, the services provided at that airport. The Commission submitted that a repeat customer might decide to switch airlines in order to obtain superior ground handling services at the relevant Australian destination airport. His Honour accepted at [320] that there was competition in respect of ground handling, “although of a somewhat constrained nature given that ground handling was very often only provided by the home carrier in a particular port”.

39    Paragraphs [321]-[326] contain the major thrust of his Honour’s reasoning concerning s 4E. Given his satisfaction that there was competition in respect of ground handling, his Honour said at [321] that the relevant enquiry was:

… not about switching in some loose sense but rather, as the text of section 4E requires, substitution. The correct question is where are the relevant substitutable services provided to consumers of those services.

40    His Honour then pointed out that ground handling services provided in Sydney were not, themselves, a substitute for air cargo services from Hong Kong to Sydney, a proposition which we accept. However, acceptance of cargo by the airline in Hong Kong would not, in itself, be a substitute for such air cargo services. His Honour went on to observe that if there were “true substitutes” in Sydney, namely ground handling services, this would lead to the proposition that a person wishing to move cargo from Hong Kong to Sydney could do so by giving possession of the cargo to the airline at Sydney, an obviously absurd proposition. We understand this proposition to mean that if ground handling services in Sydney were a true substitute for the supply of the whole suite of cargo services, the result would be that the customer could move cargo to Hong Kong by delivering it to the airlines in Sydney. We agree that the result is absurd. However we do not understand the Commission to submit that the fact that the airlines compete in the supply of ground services in Australia leads, of itself, to the finding that the market is in Australia, and only in Australia. His Honour’s approach seems to be that for the purposes of s 4E, a market must be either entirely in Australia or not in Australia at all. He then sought to identify one incident in the overall provision of cargo services which might be the single factor which determined that location. The apparent anomaly referred to at [322] is based on the same approach.

41    At [323] the primary Judge queried the “analytical significance” of the Commission’s submission that there was a market participant in Australia. By this, his Honour seems to have meant that the location of market participants in Australia had no significance in the analysis required by ss 45, 45A and 4E. His Honour understood the Commission to have submitted that if switching decisions were made in Australia by such a participant, then those decisions would not be made in Hong Kong “where the airplanes were”. His Honour observed that:

However, the question is not where the switching decision is subjectively made as an act of cognition but, instead and in contradistinction, where it is given effect. If that be correct, the location of some importers in Australia is irrelevant … .

42    His Honour then proceeded on the basis that the proposition was correct, and that the Commission’s reliance on the presence of market participants in Australia was of no assistance in establishing the location of the market. His Honour’s view as to such correctness seems to be based upon his observations at [319]-[321] as follows:

    the range of airlines available to be selected in a relevant, route-specific market is limited by the need that such airlines have a presence in Hong Kong where they will take possession of the cargo from a freight forwarder;

    in every cargo transaction there is a “legal moment” when possession is transferred, and that event can only occur in Hong Kong;

    the place where a customer chooses between airlines is “strictly limited to Hong Kong”;

    the relevant inquiry is as to where the relevant substitutable services are provided to the consumers of those services; and

    that question is not to be answered by reference to the location at which some of the services are delivered.

43    We suspect that two assumptions underlie this reasoning. They are:

    that the relevant market cannot be partly in Australia; and

    that delivery of the cargo to the airlines in Hong Kong in some way affects supply by the airlines of the whole suite of services which comprises the product.

44    At [211] his Honour seemed to doubt the correctness of the assumption made in various cases that a market may be partially in Australia for the purposes of s 4E, and that such partial presence is sufficient to satisfy the requirements of that section. However we do not discern, in his Honour’s reasons, any justification for the apparent rejection of that possibility. As to the second assumption, we accept that delivery of possession of the cargo may have legal consequences as between the parties but so, too, may the making of the contract for the provision of cargo services, wherever that event occurs. It is difficult to see why the delivery of possession, rather than the making of the contract should be identified as the effective delivery of the whole suite of services, assuming that it is necessary to identify one discrete location as being the location of the market.

45    His Honour also rejected the proposition that the “ultimate demand” for cargo services was in Australia, or that such proposition was sufficient to locate the market in Australia. His Honour acknowledged that Professor Church, an economist who gave expert evidence at trial, had accepted that demand for air cargo services to Australia was driven by a demand for imported goods in this country. The primary Judge accepted Professor Church’s comment that:

the demand in the market for air cargo is located in large part in Australia and there will be economic effects of an increase in the price of air cargo to Australia in Australia.

However his Honour did not understand Professor Church to be locating the relevant markets in Australia.

46    The remaining point concerned the extent to which the Hong Kong market might be constrained by importers in downstream markets in Australia. His Honour dismissed this proposition on the basis that there was no evidence to support it.

47    At [335] and [336]-[338] his Honour concluded that the relevant Hong Kong markets were not in Australia. His Honour also concluded that, for the same reasons, the Singapore and Indonesian markets were not in Australia.

48    Concerning the primary Judge’s reasoning, we make a number of points. First, his Honour seems to have moved from the proposition at [212] that substitution occurs in a market to the proposition or assumption that the market contains only substitution decisions, and not the circumstances on which such decisions are based. Secondly, his Honour has chosen the point at which effect is given to a “switching decision” as alone governing the question of whether the relevant market was in Australia for the purposes of s 4E. This choice presumably reflects the view that the market consists only of such decisions. The reason for selecting the location at which the decision is given effect as answering the s 4E question, rather than the place at which the decision was made, appears to be that it is the latest time at which the decision can be changed. At one level his Honour’s reasons seem to treat the question concerning the geographical dimension of the market as being the same question as is posed by s 4E. However, at another level, his Honour seems to treat it as a narrower question. Whilst the locations at which services are supplied, and customers are located are treated as relevant matters in connection with the geographical dimension, they are completely discounted in considering whether the market is in Australia. This outcome seems to be the result of his Honour’s view that the only content of the market is substitution.

The grounds of appeal

49    The primary Judge’s decision concerning “market issues” was the subject of four of the Commission’s grounds of appeal against both Air NZ and Garuda, Air NZ’s notice of contention grounds 1-5; and Garuda’s notice of contention, grounds 1-5. Grounds 6-9 of Garuda’s notice of contention were abandoned.

50    Put briefly, the Commission challenged the primary Judge’s finding that:

(1)    although parts of the service were performed in Australia (flying through Australian air space, ground handling services in Australian cities, and handling enquiries) and although there was constrained competition in respect of ground handling, these matters only formed part of a suite of services ([313], [320]-[321]); and

(2)    the relevant question is where a switching decision is given effect ([323]); the only place where a decision to switch from one airline to another could be given effect would be Hong Kong ([264]); that is the place where the airline takes possession of the cargo from the freight forwarder ([319]).

51    However an important aspect of the Commission’s case, at trial and on appeal, was and is that there were, as his Honour found, major shippers in Australia which dealt directly with the airlines, although they may also have retained their freight forwarders. By their notices of contention, the respondent airlines challenge that finding. As it is of some importance to the Commission’s case, we should deal with it before considering that case.

The respondent airlines’ challenge to the finding that large importing shippers were part of the market

52    The respondent airlines challenged the finding by the primary Judge that some large importers (shippers) in Australia were market participants in the relevant markets. They submitted that the relevant markets should have been limited to transactions between airlines and freight forwarders. There are five difficulties with this submission.

53    The first difficulty is that it did not engage to any extent with questions of agency. We did not receive any submissions from the respondent airlines concerning the proper characterisation of the relationship between the freight forwarders and the shippers in Australia and, in particular, why the freight forwarders could not be characterised as agents for the shippers. If the freight forwarders were the agents of the shippers in Australia for the purpose of the relevant transactions then the market could not be limited to transactions between the airline and an agent, excluding the principal. The same point could be made if they were agents of the airlines.

54    Although the primary Judge did not reach a conclusion concerning agency, he did conclude that (i) as a matter of economic substance the freight forwarders were intermediaries having fluctuating control over the cargo whose carriage they arranged at [299]; and (ii) the airlines regarded the goods that they carried as belonging to the shippers at [300]. These findings were not challenged by the respondent airlines.

55    The term “agent” can be slippery but it is usually used to “connote an authority or capacity in one person to create legal relations between a person occupying the position of principal and third parties”: International Harvester Co of Australia Pty Ltd v Carrigan’s Hazeldene Pastoral Co [1958] HCA 16; (1958) 100 CLR 644, 652 (the Court). In this sense, an intermediary is not necessarily an agent: Dal Pont GE, Law of Agency (3rd ed, LexisNexis Butterworths, 2014) 55 [2.23].

56    It is impossible to generalise from the evidence in this case to reach a conclusion as to whether every shipper engaged a freight forwarder as an agent. Questions of agency fall to be considered in the context of particular relationships. Indeed there was evidence that suggested that some airlines described the freight forwarders as agents for the airlines in the airlines’ relationship with the customer importers. Ultimately, although some relationships between shippers and freight forwarders may properly have been characterised as relationships of agency, others might not. Nevertheless, there was significant evidence, at least in some cases, of relationships of agency within the market. For instance, evidence from one freight forwarder (Mr Chan for National Freight Corporation, trading as Exel) described how the freight forwarder would receive blank master air waybills from the air cargo carriers prior to delivering the cargo with space left blank for the signature of the “shipper”. The freight forwarder (Exel) often listed itself as, and signed as, the shipper only because it had consolidated shipments from a number of shippers ( [51]).

57    The second difficulty with the respondent airlines’ submission is the finding of the primary Judge that airlines took note of the shippers’ businesses, even if the airlines denied doing so. As the primary Judge said at [283]:

Oral testimony given during the trial suggested that airlines were interested in what the shippers were doing even if they denied that there was any direct contact with them. Mr Gregg’s evidence was to this effect in relation to Air NZ, and Mr Haddad gave similar evidence for Garuda.

58    Mr Gregg’s evidence was that he paid close regard to the weekly and monthly reports he received from his cargo managers. And Mr Haddad, the Australian Cargo Manager for Garuda Indonesia, agreed in cross-examination that it was a “fairly commonplace occurrence for [him] to have discussion with freight agents in relation to the flight that the shipper wanted their cargo on”.

59    The third difficulty with the respondent airlines’ submission is the evidence that in some significant cases the airlines did not merely take note of the shippers’ business but dealt directly with shippers, and not merely with freight forwarders. Although this was not always the case, and although the shippers would often defer to the freight forwarders, as long as the requirements of the shippers were met, the desire of some airlines to establish and maintain relationships with the shippers is a strong indicator of the participation of the shippers in the market.

60    There was evidence from the Customs and Shipping Manager of Robert Bosch (Australia) Pty Ltd who was responsible for monitoring the performance of Robert Bosch’s freight forwarders. He said that where perishables needed to be shipped within a tight timeframe, Robert Bosch would enter into tri-partite arrangements with its freight forwarder and the airline in the form of a memorandum of understanding which would guarantee Robert Bosch its required capacity on a particular airline and priority over other shippers’ cargo (affidavit of Gunther Ressig at [106]-[107]).

61    The former Group General Manager (Freight) at Qantas, also described the existence of some tri-partite arrangements between Qantas, freight forwarders and shippers (affidavit of Stephen Michael Cleary at [53], [61]-[70], [78]-[81]). These arrangements were usually recorded in memoranda of understanding after face-to-face meetings attended by all three parties ([61]-[62]). Negotiations related to price, capacity and service standards ([62]). The Group General Manager (Freight) observed that the memorandum of understanding allowed Qantas to, “establish a relationship directly with a shipper. Qantas Freight would then be in a position to use its relationship with the shipper to reduce the risk of the freight forwarder taking the shipper’s freight to another airline” ([68]).

62    Evidence from one freight forwarder, Burlington Northern Air Freight (subsequently BAX Global), by the State Manager of its Perth, Adelaide and Sydney offices, (affidavit of Dennis Ian Nelson at [62]), was also that:

BAX Global’s major multi-national customers would regularly hold meetings with employees of BAX Global and representatives of the airlines to review our performance against these KPI’s and to ensure that there was sufficient airline capacity available. In order to maintain the relationship, it was necessary for there to be dialogue between all three parties, so that the airline knew when the customer was likely to require capacity and so the customer knew of any expected capacity constraints on the airline.

63    There was also evidence from Ms Cluff who was employed by Toshiba (Australia) Pty Ltd as a Logistics Import Specialist. In her affidavit, at [126], she said that, “On occasions, there have been tri-partite discussions between Toshiba Australia, the freight forwarder and Qantas”. She also said that on numerous occasions she liaised with Qantas directly. Ms Cluff said ([99]) that on occasions Toshiba even considered bypassing the freight forwarder altogether. The reason this never eventuated was that Toshiba Australia decided it would be difficult to handle all of the services provided by the freight forwarder, and Toshiba Australia thought it would be able to obtain a better price from the airline when it contracted a freight forwarder, given the greater total volume of goods when Toshiba Australia’s goods were combined with those of the freight forwarder.

64    The fourth difficulty with the respondent airlines’ submission is the conclusion of the primary Judge at [269], described above, that the functional dimension of the market was not a strictly vertical one, and that in some cases the consignee as importer would instigate the transaction. This finding was not challenged. Where the consignee was a large shipper who was a “significant economic actor” (to use the phrase of the primary Judge at [305]) the inference drawn by the primary Judge is irresistible. That inference is that it is hard to imagine a universe of discourse in which a rational business would ignore the role of such a shipper.

65    The fifth difficulty with the respondent airlines’ submission is that it was essentially based upon an assertion of a lack of evidence to support a conclusion that large shippers imposed a real constraint on airlines operating out of Hong Kong and Singapore. The primary Judge’s conclusion on this point was by inference based upon the evidence before him. He referred to documentary evidence and testimony that airlines regarded significant importers and exporters as targets for their marketing activities; he referred to direct contact that airlines had with customers; he referred to the tri-partite arrangements made by airlines not bi-partite arrangements with freight forwarders which excluded importers; he referred to internal Air NZ emails which recognise price pressure from exporters; he referred to internal reports of airlines about losing the custom of particular shippers ([272]-[308]). On appeal there was very little analysis by the respondent airlines of this considerable body of evidence which was before the primary Judge. This is an excellent example of a circumstance where, “the appellate court does not typically get taken to, or read, all of the evidence taken at the trial” and where the primary Judge, “has advantages that derive from the obligation at trial to receive and consider the entirety of the evidence and the opportunity, normally over a longer interval, to reflect upon that evidence and to draw conclusions from it, viewed as a whole”: Fox v Percy [2003] HCA 22; (2003) 214 CLR 118, 126 [23] (Gleeson CJ, Gummow and Kirby JJ).

66    We reject these grounds of contention.

The submissions concerning the market issues

67    Senior counsel for the Commission orally summarised his market submissions as involving two issues (app ts 4). First, the primary Judge used an incorrect test to determine whether the relevant markets were markets in Australia. Secondly, the findings by the primary Judge in respect of (i) the product dimension; (ii) the locus of competition; and (iii) the functional dimension of the relevant markets, established that the relevant markets were markets in Australia for the purposes of the relevant sections of the Trade Practices Act.

68    The Commission alleged that the primary Judge focussed erroneously upon where the switching decision was given effect (app ts 8). At [24]-[32] it submitted that this approach was erroneous because:

(1)    it erroneously focussed upon one aspect of the transaction, the transfer of possession or the place of contracting;

(2)    it erroneously assumed that there is a demand for mere commencement of the service at origin;

(3)    it was falsely premised upon the availability of the airlines in Hong Kong, where there is also a need for supply of the service at destinations in Australia;

(4)    contrary to the decision in Re Fortescue Metals Group Ltd, the test erroneously excluded factors relevant to the location of the market including the place where the customers are located, the place where the airlines direct their marketing, and the place where the airlines maintain their operations; and the test erroneously excluded the temporal dimension of the market and that competition is a “process and not an outcome”, when the competition was for the carriage of air cargo on multiple flights over periods of time.

69    On numerous occasions, the Commission submitted that the place where the services were provided to consumers was relevant in determining whether the market was in Australia (eg [4.4], [29], [34], [36]-[38]). Senior counsel submitted (app ts 8) that by focussing on the place where switching decisions were given effect, the test ignores the location of customers. He gave the example of a situation where all customers for a product were located in Australia but the product was supplied from overseas. He submitted that such a situation would surely be a market in Australia. He submitted that this was the situation in the present case because all the shippers could be in Australia (ie the consignees requesting the carriage of cargo to them) and, “the freight forwarder would be entering into the contractual arrangement with the shipper at origin, but the demand would be solely located in Australia”.

70    The Commission submitted that whether or not a switching decision was given effect in Australia, the market must be a market in Australia, since the product dimension of the market was the supply of air cargo services from an overseas port to a port in Australia ([4.1], [8]). Senior counsel for the Commission submitted that the service of carrying the cargo from an origin to a destination has no meaning in economics if it is focussed solely at the origin (app ts 10). He submitted that the expert evidence of Professor Church (on a point where Professor Gilbert disagreed), should be accepted as supporting the conclusion that wherever the product description includes a geographic place (app ts 23-27), the geographic dimension of the market will be located at that place.

71    In support of this submission, the Commission also submitted that the services which made up the product involved geographic elements so that there was a “one to one” correspondence between the product (or “price-product-service”) dimension of the market and the geographic dimension ([16]) and that competition physically took place in Australia ([17]). These geographic elements of the product description included: (i) delivery to a destination port; (ii) ground handling services at destination ports; and (iii) enquiry services at destination. The Commission observed that an exercise of market power might occur in Australia by an airline withdrawing those services provided in Australia, diminishing their quality or introducing collection charges ([23]).

The proper approach to whether a market is “in Australia”

72    In broad summary, our approach to whether a market is “in Australia” involves two related issues. The first issue involves defining the market. As we have explained, the established approach to market definition is to consider three “dimensions” of the market: the product, geographic, and functional dimensions. This approach is simply a heuristic tool to assist in understanding the market. The Trade Practices Act does not divide the market in this way, and in the exercise of market definition the three dimensions are not mutually independent.

73    The second issue involves characterising whether the defined market is “in Australia”. In many cases, a determination of the geographic dimension of the market will reveal whether the market is “in Australia”. But the question is not whether the “geographic dimension of the market” is in Australia. Nor is it whether the effect of a switching decision is in Australia. The question is whether, as a matter of characterisation, the market is in Australia. For this reason, all of the matters identified by the Commission should properly be considered.

74    This approach is textual. It is supported by the legislative background to s 4E. It is also contextual and purposive. And it is supported by authority. Our consideration of the issue below is divided into the following sections:

(1)    legislative background;

(2)    the flexibility of the concept of a “market in Australia” and purposive considerations;

(3)    difficulties with the primary Judge’s approach;

(4)    the proper approach to the words “market in Australia”;

(5)    identification of the market; and

(6)    characterisation of whether the identified market is “in Australia”.

75    One significant area of authority which is not considered below is the body of antitrust decisions from the United States. The primary Judge held that United States authorities were not of any assistance because there is a substantial difference in the “effects doctrine” under the Sherman Antitrust Act 15 USC (1890), a doctrine which was developed as early as the decision of Learned Hand J in United States v Aluminum Co of America 8 F 2d 416 (2d Cir 1945). The Commission did not challenge this. Neither at trial nor on this appeal did the Commission submit that the appropriate approach to whether a market was “in Australia” was to test whether the relevant effect of anti-competitive conduct was felt in Australia. The parties all avoided any reference to, or discussion of, the copious United States authority on this issue. That authority is, however, not uniform: Brown GW, “Relevant Geographic Market Delineation: The Interchangeability of Standards in Cases arising under Section 2 of the Sherman Act and Section 7 of the Clayton Act” (1979) Duke Law Journal 1152.

(1) Legislative background

76    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.

77    When the Trade Practices Act was first enacted, the term “market” was defined in s 4 as follows:

“market” means a market in Australia.

78    According to the explanatory memorandum which accompanied the draft Bill at [87]:

The extent to which the legislation will operate extra-territorially is indicated in clause 5. The definition of market in clause 4 is also relevant in this regard.

79    The Trade Practices Amendment Act 1977 (Cth) repealed that definition and inserted s 4E in its present form, save that the words, “unless the contrary intention appears”, were not present. Those words were inserted by the Trade Practices (Misuse of Trans–Tasman Market Power) Act 1990 (Cth) to allow for the existence of trans-Tasman markets.

80    At [4] of the explanatory memorandum which accompanied the 1977 Bill, it was said that:

A “market” for goods and services has been defined to include substitutable or competitive goods or services … .

In the second reading speech, the Hon John Howard MP, as he then was, referred to an earlier Bill introduced in 1976. It had lapsed when Parliament was prorogued: see Commonwealth, Parliamentary Debates, House of Representatives, 8 December 1976, 1476 (John Howard). In the second reading speech in connection with the 1976 bill, Mr Howard referred to a committee report, saying that the bill would implement some of its recommendations. The report in question was the Swanson Report: see Trade Practice Act Review Committee, Report to the Minister for Business and Consumer Affairs (Australian Government Publishing Service, Canberra, 1976). Concerning the meaning of the term “competition”, the committee adopted the language used by the Trade Practices Tribunal in QCMA. The committee said:

4.20    The concept of market involves the performance of a function in relation to a product, being goods or services, within a geographical area. Markets within the chain of distribution, from manufacturing to wholesaling and retailing, depend upon factors internal to the industry concerned, and distinction may in particular cases be blurred or non–existent. Product and geographic markets, on the other hand, depend upon factors extraneous to the industry. Their boundaries are determined by the relationship between such factors as price, product substitutability, desired use and distance from supply, to name some. Because of the variable nature of such factors, the boundaries of product and geographic markets are necessarily flexible.

4.21    The Committee considers that no advantage would be gained by attempting to define exhaustively the term ‘market’. No definition could produce a formula capable of certainty, having regard to the variable nature of the factors discussed in the paragraph above. Importantly also, the Committee has regard to the fact that persons involved with particular cases wish the matters in dispute to be judged on the particular facts, as they may present them, and not by artificial rules designed to achieve what we would suggest is an illusory certainty.

4.22    There is, however, one aspect of the definition of ‘market’ about which we consider the Act should give useful legislative guidance; namely, in relation to product substitution. The Committee therefore recommends that the Act should require that, in the determination of a ‘market’ for particular purposes, regard shall be had to substitute products, being products which have a reasonable interchangeability of use and which have high cross-elasticity of demand, i.e. where a small decrease in the price of a particular product would cause a significant quantum of demand for a similar product to switch to the product in question.

81    We have set out the history of the definition of the word “market” in order to demonstrate that the enactment in 1977 of s 4E was not intended to change the meaning of the term “a market in Australia”, save to the extent that the amendment may have extended the concept of “market” by the addition of the inclusory provision which requires that regard be had to substitute products as one of the factors to be considered in the flexible assessment of the factors comprising a market. In other words, the amendment seems to have been intended to extend the range of goods or services to be taken into account in the market identification exercise.

(2) The flexibility of the concept of a “market in Australia” and purposive considerations

82    The reference to a “market” in s 4E is to an abstract concept, not to a physical place. In one sense the description is a metaphor. It conjures an image of a “marketplace” involving a single location where buyers and sellers meet. But the legislation is plainly not limited to this narrow conception which developed well before the era of global commerce in which the Trade Practices Act was enacted. Moreover, the history of s 4E, and its purpose, manifest Parliament’s intention that the section be interpreted flexibly and in light of the purposes of the Trade Practices Act.

83    The Swanson Committee recognised the flexibility of the boundaries of product and geographic dimensions of a market due to the variable nature of factors such as price, product substitutability, desired use and distance from supply ([4.20]). One reason why the evaluative exercise contemplated by the Swanson Committee was intended to be flexible is that the Committee was conscious of the dangers of definitions that are too wide or too narrow. At [4.19] the Committee said:

If the market is too widely defined it may be that the requisite effect upon competition cannot ever be shown, to the detriment of those seeking relief from a restrictive agreement or practice. Alternatively, if the market is too narrowly defined it may result in hardship to and unnecessary limitations upon business actions … .

84    Air NZ submitted ([26]) that defining a market too widely would have the result that the requisite effect upon competition would never be shown, citing French, Spender and O’Loughlin JJ agreeing in Singapore Airlines Limited v Taprobane Tours WA Pty Ltd [1991] FCA 808 [167]; (1991) 104 ALR 633, 642. This can be accepted. But, equally, the narrower the approach that is taken to whether a market is “in Australia”, the less likely it will be that any effect on competition, no matter how significant to Australians, would fall within ss 45 and 45A.

85    The need to avoid too narrow an approach to whether a market is “in Australia” for the purposes of s 4E is reinforced by the objective of the Trade Practices Act as set out in s 2: “to enhance the welfare of Australians through the promotion of competition and fair trading and provision for consumer protection”.

86    The explanatory memorandum to the Trade Practices Bill 1974 (Cth) indicated that the words “in Australia” were intended to limit the extraterritorial operation of the Trade Practices Act: Explanatory Memorandum, Trade Practices Bill 1974 (Cth) 19 [87]. However, the explanatory memorandum does not indicate the extent to which those words were intended to constrain. In a submission to the Dawson Committee for Trade Practices Review in 2002, Professor Brunt submitted that one feature of the Trade Practices Act that is “glaringly inappropriate to the global challenge” is s 4E. She said that a more expansive definition was required because the current definition failed to take into account “international forces”. In cases involving mergers she thought that the courts should take account of the international forces by considering the impact of actual and potential behaviour of overseas buyers and sellers upon Australian transactions. Hence, she proposed an amendment that “market” should mean a market in relation to Australia.

87    The expanded definition proposed by Professor Brunt was not adopted. It may be, therefore, that the Trade Practices Act does not protect consumers from the effects of international forces upon Australian consumers of international buyers and sellers. If so, then whether it should do so as a matter of policy would be a question for Parliament. But this case directs attention to a considerably greater restriction upon the protection given by the Trade Practices Act to the welfare of Australians. This case concerns the direct participation of Australians in the relevant transactions, including (as a matter of economic substance) Australian shippers through the intermediaries of freight forwarders, and it concerns the performance of important elements of the services in Australia.

(3) The proper approach to the words “market in Australia”

88    As the primary Judge pointed out, there has been virtually no judicial consideration of the requirement in s 4E that a market be “in Australia”. In those circumstances, it is surprising that in argument before us, little or nothing was said concerning the proper construction of that section. The respondent airlines tended to assume that the location of the market in Australia would depend upon its geographical dimension. The Commission took a somewhat different view, looking to both the product dimension and the geographical dimension, although it also made a submission that there needed to be a one-to-one correspondence between these two dimensions. Thus, at [18] of its written submissions, the Commission submitted that the fact that part of the services that formed the product dimension of the relevant markets was provided in Australia, and that competition for the supply of those services took place in Australia, led to the conclusion that the market was in Australia.

89    Neither of these approaches accords with the text or purpose of s 4E. In our view, properly construed, the exercise required by s 4E is twofold. First, as the Commission submitted, it requires identification of the relevant market. That identified market will include, “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”. Section 4E does not prescribe the process by which the market is to be identified. We have already discussed the traditional way in which the process is approached by reference to three inter-related “dimensions”, having regard to the economic analysis in question, in this case, that posed by ss 45 and 45A.

90    Secondly, s 4E requires characterisation of whether or not the identified market is in Australia. In many cases the geographical dimension of a market might easily resolve the characterisation issue of whether the market is “in Australia”. But it would be an error to approach the question of characterisation simply by asking whether one dimension of the market was “in Australia”. The entire market must be considered, and the other dimensions may be relevant to the question.

91    The identification of the “market” and the characterisation of whether it is in Australia are therefore flexible questions which require evaluative judgement in light of the purpose of the relevant provision.

(4) Difficulties with the approach of the primary Judge

92    There are six difficulties with applying the respondent airlines’ (and the primary Judge’s) approach as a test to determine whether the market is “in Australia”.

93    The first problem, which we have already identified, is the lack of any textual foundation in s 4E for this approach. Section 4E is not concerned only with whether the supplier is located in Australia. Nor is it concerned only with whether the effect of a substitution is located in Australia. It is concerned with whether a market is located in Australia.

94    The second problem is understanding what is meant by locating the place where a substitution decision is given effect.

95    The primary Judge found, and the respondent airlines submitted, that substitution took effect where possession was taken of the cargo. We can see only two possible explanations for adopting that point as determining the location of the market. The first is that it is the point at which the airline commences to supply the relevant services as this is where the choice of supply takes effect. The second is that it is said to be the last point at which any change of mind can occur (app ts 110). We find neither of these explanations to be convincing.

96    As to the first explanation, almost inevitably a choice for supply of the service will have been made prior to both the making of the contract for carriage and delivery of possession. The product is a complete suite of services. There is no basis for treating the delivery of possession as being more important than any other part of the supply. In any event, the customer will be, for all practical purposes, committed to the choice for supply of the service before delivery of the cargo. Delivery of the cargo to the carrier will inevitably follow the choice. There was no reference to any evidence that shippers or freight forwarders frequently resile from a choice which had been notified to the chosen carrier, let alone after a contract had been made. Hence there is no logical basis for describing the point of delivery to the carrier as the point at which the choice took effect.

97    As to the second possible explanation, it assumes that the first act of performance becomes the point of no return. But it is hard to see why the place of loading the cargo is the point of no return rather than, for instance, the place where the customer becomes irrevocably bound to pay for the services of the airline, subject to repudiating the contract and paying damages. From a commercial, legal, and moral point of view, the point of no return is the making of the contract, or possibly communication of the choice. The location of the making of the contract evidenced by the air waybill might bear no relationship to any of the underlying economic realities of the transaction. Nor did the respondent airlines submit that the place where the switching was given effect was that of the proper law of the contract.

98    His Honour’s approach seems to be based upon the proposition that the process of substitution necessarily provides the answer to the enquiry prescribed by s 4E. The process of substitution may be relevant to market identification, and may therefore have an effect on whether the market is “in Australia” but s 4E is not confined to substitution. Further, as we have explained, significant difficulties attend the identification of the point at which substitution occurs or may occur. These difficulties flow from the quite arbitrary choice of the criterion for deciding the s 4E question.

99    Ultimately, the respondent airlines submitted that the place where the switching decision is given effect is the location of the supplier. Although the location of the supplier need not correspond with the point of no return for a switching decision by a customer, the respondent airlines’ submissions assimilated the primary Judge’s test of the place where a substitution is given effect with the question of the location of the supplier.

100    The assimilation of the place where substitution is given effect and the location of the supplier may have reflected the emphasis placed on the latter by the experts Professor Church and Professor Gilbert. We put to one side the evidence of Dr Williams upon whose evidence the Commission did not rely in this appeal. He had accepted that his approach to market definition was based on a test which had been rejected by the Australian Competition Tribunal (ts 2407-2408): see Application by Services Sydney Pty Ltd [2005] ACompT 7; (2005) 227 ALR 140, 169-170 [122] (Gyles J, Keane and Walker). Air NZ pointed to the agreement between Professors Church and Gilbert that the, “geographic dimension of a … market identifies the location and identity of competing suppliers in the relevant product market”, and that the, “geographic dimension of a market describes the locations of the suppliers of the products that consumers consider to be close substitutes for each other” (Air NZ submissions [49]).

101    Any attraction of this test (which focusses upon the location of the suppliers) is dispelled when one attempts to apply it. Where is an international airline that supplies a service located? Is it the place of its incorporation? Is it the location of its international head office? Is it the location of all of its operations? Is it, as the respondent airlines submitted, as simple as pointing to Hong Kong (or “the vicinity of the Hong Kong airport”: app ts 103) as the place where the service begins because at that place the planes will be loaded with cargo? Or is it, as the Commission submitted, all the places where significant parts of the service are performed in circumstances where the airlines have, and require, permits to operate in Australia and do, in fact, operate in Australia?

102    The experts who gave evidence at trial were divided in their answers to this question. The approach of Professor Gilbert (ts 2458), appeared to be that the supplier is located at the origin of the service it provides, or the region in which the product is supplied. The approach of Professor Church (ts 2366-2367), appeared to be that it is located at all points between origin and destination of the service it provides.

103    In summary, the second problem with the primary Judge’s approach is that if the location of the market were to be defined solely by reference to the location of the supplier, it would still be necessary to determine that location. It is difficult to understand why a supplier could not be located at any place at which it operates.

104    The third problem with a test which focusses upon the location of the supplier, or place where substitution takes place, is that as a matter of principle it is hard to see why these should be the only relevant matters. Once it is accepted that the concept of a market is an abstract concept, or possibly a metaphor, rather than a description of an activity which takes place in a physical location or marketplace, then the concept of a test based upon a single factor which identifies a physical location appears inapt. In a world where commerce is often electronic, customers rarely travel to a marketplace where the seller is physically located. Indeed, suppliers may have very little physical presence at all. The physical location of an itinerant supplier could be a website which might be located on a server, or series of servers, anywhere. Or a location might be large warehouses in a country wholly unknown to any customer in another country who places an electronic order. In some cases, the location of a seller, as a matter of economic reality, might be the location of the customer.

105    The fourth problem with an attempt to find a single “location of the seller”, or a single place where a switching decision is given effect is that this approach is contrary to the legislative history of the provision and the desire for flexibility. It treats this matter as the sole criterion for locating the market, contrary to the flexibility which was intended in the definition in s 4E. The place of substitution or switching is a consideration in determining the dimensions of the market and may therefore have an effect in determining whether a market is “in Australia”. But it is not the only consideration.

106    As we have explained, s 4E was introduced following the Swanson Report in August 1976. The Swanson Committee recognised the flexibility of the boundaries of product and geographic dimensions of a market due to the variable nature of factors such as price, product substitutability, desired use and distance from supply ([4.20]). The Committee concluded that no advantage would be gained by attempting an exhaustive definition since, “persons involved with particular cases wish the matters in dispute to be judged on the particular facts, as they present them, and not by artificial rules designed to achieve what we would suggest is an illusory certainty” ([4.21]).

107    The fifth problem with a test which searches only for a single “location of the seller” or a single place where a switching decision is given effect is that in situations in which this test is most easily applied it can give rise to a meaning which is so narrow that it creates results which are contrary to the policy of the Trade Practices Act as set out in s 2: “to enhance the welfare of Australians through the promotion of competition and fair trading and provision for consumer protection”. As we have explained above, the narrower the meaning that is given to a “market in Australia” the less likely it is that any effect on competition, no matter how significant, will fall within ss 45 and 45A.

108    The sixth problem with a test which searches for a single “location of the seller” or a single place where a switching decision is given effect is that it is not consistent with the authorities which we discuss below. Rather, the authorities suggest that many other considerations are relevant to the question.

109    Ultimately, to the extent that it is meaningful to speak of a “location of a seller” or a “place where switching is given effect” such considerations will be relevant factors in determining the geographic dimension of a market. But they will not be conclusive. The abstract concept of a market involving buyers and sellers does not require that a seller have a single location. Nor does the language or the purpose of s 4E require this. We will return to this question. Other relevant factors may include the location of customers, the place where the goods or services are marketed (as in Atlantic Container Line AB v European Commission [2002] All ER (EC) 684) or even the sphere of geographic operation of a service which delimits possible customers. The determination of the geographic dimension of a market for the purposes of ss 4E, 45, and 45A is an evaluative exercise to be conducted consistently with the words of those sections, and their purpose. It cannot be reduced to a single monolithic test based only on one factor.

(5) Identification of the market: authority and application

110    Identification of a market is closely related to the concept of competition. As a framework for economic analysis, the market will be defined in a way which reflects the nature of the problem to be solved. In connection with alleged anti-competitive conduct contrary to a provision of the Trade Practices Act, the chosen market will be the framework in which the case will be decided. For present purposes, the impugned conduct is said to be contrary to ss 45 and 45A of the Trade Practices Act. Broadly speaking s 45 prohibits a corporation from entering into, or giving effect to, a contract, arrangement or understanding which has, or is likely to have the purpose, effect or likely effect of substantially lessening competition. Section 45A provides that a contract, arrangement or understanding which provides for price fixing is to be deemed to have the proscribed purpose, effect or likely effect. Section 45(3) provides that for the purposes of ss 45 and 45A, the word “competition” means competition in any market in which a relevant corporation supplies or acquires goods or services. Hence s 4E is engaged.

111    We have emphasised the point that, contrary to the respondent airlines’ submissions and the approach taken by the primary Judge, the identification of a market, and even the identification of its geographic dimension, is not determined by a single component such as “location of a seller” or “the place where a substitution is given effect”. The course of Australian authority supports this approach. It is also supported by the approach taken to similar questions in analogous fact scenarios in New Zealand and Europe.

Australian authority

112    Both competition and markets are discussed in detail in the decision of the Trade Practices Tribunal in QCMA. In QCMA at 517, the Tribunal said that 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. Inherent in this proposition is the assumption that a transaction may include an agreement to supply and acquire goods or services, and the actual supply and receipt of the goods or services. In our view, both aspects occur within the market. The choice might be evidenced by the contract, but performance is an essential aspect of competition in that market. It is part of the transaction. It follows that the geographic dimension of the market is not confined to the location of the seller or the place where a substitution is given effect.

113    In other words, the identified market will contain all of those matters which comprise the “richness” of the concept of competition, including transactions and potential transactions, and the circumstances and conduct which lead to the possibility of further transactions and to the transactions themselves.

114    As appears from the reasons of Spender J in QIW Retailers Ltd v Davids Holdings Pty Ltd (No 3) (1993) 42 FCR 255 at 267-268:

A market may be seen as having three dimensions: the product market, the geographic market, and the relevant functional level. This was explained by the Trade Practices Commission in its Second Annual Report (1975):

“The process of market definition consists in seeking to isolate the area of effective competition in which the parties operate. This requires the Commission to assess three dimensions – the product market, the geographic market and the functional level.

The product market consists of the goods or services supplied by the parties, together with competing goods or services which could reasonably be used by most customers as substitutes for them …

The geographic market can be described as the geographic area or areas in which sellers of the particular product operate and to which purchases can practicably turn for such goods or services. The geographic market is a function of a variety of factors, including the pattern of demand and the value of commodity in relation to the cost of transporting it or, in the case of services, the degree of inconvenience involved in obtaining them from another source …

In industries where functional levels are clearly distinct, the levels on which the parties respectively operate are a dimension to consider when determining the relevant market.”

The dimensional character of markets was also acknowledged in [Singapore Airlines Ltd v Taprobane Tours W.A. Pty Ltd (1991) 33 FCR 158] at 174-175 per French J.

Professor Brunt at 102 states that a market has product, space, function and time dimensions. The time dimension relates to “how much time is needed for customers and suppliers to make their adjustments in response to economic incentives”. Professor Brunt also notes that these dimensions may not be independent.

The relevant market for the purposes of the economic analysis in question will be defined by reference to those dimensions.

115    In Queensland Wire Industries Pty Ltd v Broken Hill Proprietary Co Ltd [1989] HCA 6; (1989) 167 CLR 177 Mason CJ and Wilson J referred to QCMA at 188 as, “explaining that the defining feature of a market is substitution”. At 195, Deane J said:

Section 4E confines “market” for the purposes of the Act to “a market in Australia” and provides that, when the word “market” is used in relation to any goods or services, it “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”. Section 46(4) provides that a reference in s 46 to a market is a reference to a market for goods and services. The Act does not otherwise seek to define what is meant by the word “market”. That is not surprising since the word is not susceptible of precise comprehensive definition when used as an abstract noun in an economic context. The most that can be said is that “market” should, in the context of the Act, be understood in the sense of an area of potential close competition in particular goods and/or services and their substitutes … .

116    At 199, Dawson J said:

A market is an area in which the exchange of goods or services between buyer and seller is negotiated. It is sometimes referred to as the sphere within which price is determined and that serves to focus attention upon the way in which the market facilitates the exchange by employing price as the mechanism to reconcile competing demands for resources … .

117    In Queensland Wire, the relevant question was whether there was a market, rather than whether an identified market was “in Australia”. We accept that, as Mason CJ and Wilson J observed, market definition primarily addresses substitution. However it does not follow that a market is comprised only of substitution or substitution possibilities. The description of competition in QCMA demonstrates the width of the concept of competition which takes place in the market. A decision as to substitution is the outcome of such competition. The views expressed by Deane J are similar to those expressed in QCMA.

118    At first blush, it may seem that Dawson J was focussing upon actual transactions, rather than upon the circumstances leading to each transaction, and the decision to enter into it. However his Honour’s focus was not on the transaction, but upon the negotiation leading to the transaction. Negotiation, and the transition from negotiation to choice, and from choice to acquisition of the relevant product, is the consequence of the competition which occurs in the market, not the sole manifestation of such competition. His Honour’s reference to the determination of price should be similarly understood.

119    In Seven Network Ltd v News Ltd [2009] FCAFC 166; (2009) 182 FCR 160 at 283-284 [585], Dowsett and Lander JJ (Mansfield J concurring), noted eight propositions concerning competition in a market, which propositions emerge from a consideration of the various decisions, as follows:

(1)    competition in a market is the sum of activity engaged in by persons in promoting the sale of the goods or services with which the market is concerned;

(2)    competition may express itself as rivalrous conduct in any aspect of the “price product services” package or in all of them;

(3)    in particular, firms may compete by devising new products or services, new technology, more effective service or improved cost efficiency;

(4)    the most important aspect of market structure is the height of barriers to entry;

(5)    in identifying and examining a market it is necessary to consider the extent of product differentiation and promotion;

(6)    it is also necessary to consider “vertical relationships” and any “formal, stable and fundamental” arrangements between firms which restrict their capacities to function as independent entities;

(7)    market participants may expect competition from other market participants, from other existing entities, or from entities “as yet unborn”; and

(8)    a market may exist where there is a potential for close competition even if there is no actual competition at the relevant time.

120    The following passage from Heydon JD, Trade Practices Law (Lawbook Co., subscription service) at [30.245] has frequently been cited with apparent approval:

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 of 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.

121    At [30.240], the author put that same proposition in a different way as follows:

The market is where sellers meet buyers. It is the mechanism accommodating the transactions which pass goods and services from one person to another. It also encompasses the events leading up to those transactions. There are two streams of prior conduct meeting at the point of final transaction: the conduct of the seller and the conduct of the buyer. Both streams must be considered in order to form a picture of the market.

See also Australian Competition and Consumer Commission v Cement Australia Pty Ltd [2013] FCA 909; (2013) 310 ALR 165 at 215 [249] (Greenwood J).

122    In Singapore Airlines Ltd v Taprobane Tours WA Pty Ltd the Full Federal Court considered whether Singapore Airlines had abused market power contrary to s 46 of the Trade Practices Act by (i) restricting sales to Taprobane (a Western Australian travel wholesaler) of wholesale flights to the Maldive Islands; and (ii) charging Taprobane prices for Singapore Airlines’ services that were higher than prices charged to other wholesalers. An issue for the Full Court was whether Singapore Airlines had a substantial degree of power in the relevant market. Justice French (Spender and O’Loughlin JJ concurring) said at 174:

The concept of “market” has long involved elements of both activity and location. The ordinary meanings set out in the Shorter Oxford English Dictionary include “the meeting together of people for the purchase and sale of provisions or livestock, publicly exposed at a fixed time and place”, and the “action or business of buying and selling”. At common law it was a “franchise right of having a concourse of buyers and sellers to dispose of commodities in respect of which the franchise was given” … .

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.

The descriptive element is seen in the discussion of “market” in the report of the US Attorney-General’s National Committee to study the Anti-Trust Laws (1955), p 222:

“For purposes of economic analysis the ‘market’ is the sphere of competitive rivalry within which the crucial transfer of buyer’s patronage from one supplier of goods to another can take place freely.”

Clearly, his Honour understood the term “market” to describe not just “manufacturing, wholesale and retail sales” but rather a range of economic activities defined by reference to those functions.

123    At 175, French J said:

An early commentary on the Trade Practices Act from an American perspective described the notion of the market as designed to assist the court to determine whether the conduct in issue is likely to be anti-competitive in fact and, if so, whether it is likely to be sufficiently anti-competitive to fall within the statutory prohibition. The possible markets in any given case might encompass a continuum of product and geographic ranges, the selection of the outer limits of which would depend, in a s 46 context, on how much market power is thought to be significant … . [Professor Breyer] (now Judge Breyer) suggested that the word “market” is best understood not as denoting an objective feature of the world but as designed to set in motion a process related to the effective application of the particular statutory provision under consideration. That process may lead to the drawing of different lines and different circumstances depending upon the purpose of the provision in question.

124    The present case does not concern s 46, but these observations are nonetheless relevant. In QCMA, the Trade Practices Tribunal said at 516 that, “[c]ompetition is a process rather than a situation”. In our view this means that competition is the conduct of buyers and sellers and the myriad circumstances in which such conduct occurs. When the Tribunal spoke of a market being the “area of close competition”, the “field of rivalry”, or the “field of actual and potential transactions between buyers and sellers”, it meant that the market, properly identified and defined, is the “space” in which the competitive process takes place. As French J said in Singapore Airlines at 174, in competition law, the term “market” describes a range of economic activities defined by reference to particular economic functions, the class or classes of products and the geographic area within which those activities occur.

125    Justice Heydon put the same proposition in a slightly different way when he said, extracurially, that the market is the mechanism which accommodates transactions which pass goods and services from one person to another, and the events leading up to those transactions. These are the streams which meet at the point of the final transaction. Thus the market contains much more than contracts and performance of contracts. It also includes much more than choices or substitution.

126    Another decision relied upon by the Commission was the decision of the Trade Practices Tribunal in Re Application by Concrete Carters Association (Victoria) (1977) 31 FLR 193; (1977) 16 ALR 387. It is a case where the definition of the market for a product was limited by the geographic area within which the suppliers (concrete carters) could travel.

127    In that case, the applicants, members of the Concrete Carters Association (Victoria), applied to the Tribunal for review of a decision of the Trade Practices Commission. The Trade Practices Commission had refused to grant them authorisation to make a contract or enter into an arrangement or understanding with producers of pre-mixed concrete in certain areas of Victoria. The Tribunal was required to consider whether the benefit to the public would outweigh the detriment constituted by any lessening of competition. Determination of the relevant market was the “essential first step” (216).

128    The normal practice was for delivery of pre-mixed concrete from fixed plants to varying construction sites. Since the plants were not normally mobile, the delivery or cartage service involved the use of trucks containing a “barrel” which kept the mix in constant movement. The Tribunal considered the average and maximum radius of a journey (201-211) and concluded that the geographic dimension of the cartage market was the Greater Melbourne region. The geographic boundaries of the market had been limited by the capacity of the trucks with a product that must be kept volatile.

129    Definition of the geographic dimension of a market also arose in Davids Holdings Pty Ltd v Attorney General of the Commonwealth (1994) 49 FCR 211. The context in which the issue arose was whether the acquisition of Queensland Independent Wholesalers by Davids Holdings meant that Davids Holdings would be in a position to “dominate a market for goods or services” contrary to s 50(1) of the Trade Practices Act. Davids Holdings was the holding company for a group of companies whose activities included wholesaling and distribution of grocery products in Queensland and northern New South Wales. The independent retailers for whose business they competed were Coles, Woolworths and Franklins. Davids Holdings argued that the geographic dimension of the market was all of Australia or, alternatively, the eastern States of Australia.

130    At first instance, the primary Judge defined the geographic dimension of the market as the market for the supply of grocery products by independent wholesalers to independent retailers in Queensland and northern New South Wales. See QIW Retailers Limited v Davids Holdings Pty Limited (No 3) above. At 267, his Honour quoted from an annual report of the Trade Practices Commission which said:

The geographic market can be described as the geographic area or areas in which sellers of the particular product operate and to which purchasers can practicably turn for such goods or services. The geographic market is a function of a variety of factors, including the pattern of demand and the value of the commodity in relation to the cost of transporting it, or in the case of services, the degree of inconvenience involved in obtaining them from another source. (Emphasis added).

131    The Full Court of the Federal Court upheld his Honour’s approach and conclusions. Justice Von Doussa (with whom O’Loughlin and Drummond JJ agreed) said that the finding had flowed essentially from findings of fact which included that the geographical extent of the market is essentially a function of transport costs and the need for distribution centres to be close to their customers (227). The Full Court also emphasised that in Queensland and northern New South Wales each of the two companies competed fiercely with the other including conduct which “targets customers of the other and offers incentives to induce a change of supplier” (228).

The position in New Zealand: Commerce Commission v Air New Zealand Ltd

132    Both the Commission and the respondent airlines relied upon the decision of Asher J in the High Court of New Zealand, Commerce Commission v Air New Zealand Ltd (2011) 9 NZBLC 103, 318. That case was tried on agreed facts by Asher J, sitting with a lay member, economist Professor Richardson. The decision demonstrates that in considering the question of market identification, the geographic dimension of the market is not limited to some deemed location of the supplier or deemed place of substitution.

133    The underlying facts in the Commerce Commission case bear some similarity to those in this case. The Commission alleged that a number of airlines had colluded to raise the price of air cargo by charging fuel and security surcharges on the cargo shipments in and out of New Zealand. In the course of addressing a series of issues, the High Court held that the airlines had supplied services in competition with each other in New Zealand. One of the central reasons for the decision was that the demand for the cargo would often emanate from a New Zealand importer ([189]). A significant part of the demand for the services was in New Zealand ([190]). There are, however, two substantial differences between the Commerce Commission decision and this case.

134    First, there is a different emphasis on substitutability in the market definition process. In the Commerce Commission case, the High Court observed at [118] that s 3(1) of the Commerce Act 1986 (NZ) defined a “market” as a “market for goods and services that may be distinguished as a matter of fact and commercial common sense”. The High Court also followed authority in the New Zealand Court of Appeal that, “the test is not substitutability as such, although that will ordinarily be an important consideration” ([119]). At [188] the High Court rejected substitutability as a necessary determinant of geographical boundaries. Instead, the Court preferred an approach which generally identified a market by reference to the, “activities of those engaged in commerce, the structures underlying their activities and the perceived susceptibility to change in the medium-term future” ([189]). As the Australian authorities above demonstrate, this is a departure from the role of substitutability in s 4E as applied in Australia.

135    Secondly, as the primary Judge observed at [334], the Court in Commerce Commission was prepared to infer a number of matters concerning downstream substitution effects (such as those arising from consumers seeking to locate a different source of supply of the goods imported through freight forwarders). In this case there was no basis upon which a conclusion could be reached that sufficient downstream substitution had occurred (or could be aggregated) such that it would operate as a constraint upon the upstream air cargo market ([333]).

136    Despite the substantial differences between the New Zealand decision in Commerce Commission and Australian authority, one aspect of the New Zealand High Court’s decision is worthy of note. It is the relevance it placed upon customers in the assessment of the geographic dimension of the market, relying upon the decision of the Australian Competition Tribunal in Re Fortescue Metals Group Ltd. In particular, the New Zealand High Court relied upon the following passage at [1022]:

The geographic market is the area of effective competition in which sellers and buyers operate. What is relevant, as a starting point, are actual sales patterns, the location of customers and the place where sales take place, and any geographical boundaries that limit trade. (Emphasis added)

137    As the High Court explained at [184], “while the starting point for the geographic market definition exercise is the location of the points of supply, the boundaries of the geographic market are drawn to include customer locations”. The same conclusion had been reached by the High Court in Commerce Commission v Ophthalmological Society of NZ Inc (2010) TCLR 994 [192(d)]. In that case, it was the locations of customers for cataract surgery that determined the boundary of the geographic dimension of the market rather than the location of the supply.

The approach in Europe

138    As with the New Zealand authority to which the parties referred, a comparison may also be drawn with the approach taken by the European courts to Article 82 of the Treaty Establishing the European Community (Consolidated Version), Rome Treaty, 25 March 1957, European Union (now Article 102 of the Consolidated Version of the Treaty on the Functioning of the European Union, 13 December 2007, European Union, 2008 O.J. C 115/47 as amended by the Treaty of Lisbon Amending the Treaty on European Union and the Treaty Establishing the European Community, signed 13 December 2007, European Union, 2006 O.J. C 321 E/37 (entered into force 1 December 2009)). That Article includes the following:

Any abuse by one or more undertakings of a dominant position within the common market or in a substantial part of it shall be prohibited as incompatible with the common market insofar as it may affect trade between Member States.

139    Article 82 therefore requires that an undertaking have a dominant position “within the common market or in a substantial part of it”. An integral issue is whether the undertaking has a dominant position “within the common market or in a substantial part of it”.

140    In Atlantic Container Line AB v Commission (TAA) [2005] 4 CMLR 20, the Court of First Instance (third chamber) considered an agreement (the Trans-Atlantic Conference Agreement [1959–1962] O.J. Spec. Ed. 87) reached between shipping companies on the fixing of prices for inland transport services provided within the Community. The Trans-Atlantic Conference Agreement contained a number of rules concerning other aspects of transport including the remuneration of freight forwarders.

141    One defence raised by the companies was that they did not have a dominant position within the common market or in a substantial part of it. The Commission rejected this submission. The Commission concluded that the relevant market for sea-transport services was, “the market for containerised liner shipping ... between ports in northern Europe and ports in the United States and Canada” ([84]). The Commission considered the geographic dimension of the relevant market for maritime transport services as follows ([519]):

The geographic [dimension of the] market [consists] of the area in which the maritime transport services defined above are marketed, that is, in this case, the catchment areas of the ports in northern Europe. Such a geographic market is commensurate with the scope of the [Trans-Atlantic Conference Agreement’s] inland tariff and constitutes a substantial part of the common market.

142    In other words, even at the stage of determining the geographic dimension of a market, the Commission was not concerned with the physical location of a supplier or the location of a place of substitution, but with the area in which the services were marketed.

143    The Commission held that the geographic dimension of the market was limited to the catchment areas of the ports in northern Europe and did not extend to the Mediterranean ports of southern Europe even though northern European ports were for some (but very few) shippers substitutable for some Mediterranean ports. The principal catchment areas were Ireland, the United Kingdom, Denmark, the Netherlands, Belgium, Luxembourg, most of Germany, and northern and central France ([67]-[68]).

144    On appeal, the Court of First Instance (third chamber) upheld the decision of the Commission that the geographic dimension of the market was limited to the catchment areas of the ports in northern Europe ([855]-[857]). The Court explained at [852]-[853], that there is a difference between the, “geographical component of the transport services constituting the relevant market” and, “the definition of the relevant geographic market [ie the geographic dimension of the market]”.

145    The Court of First Instance at [852]-[853] also explained that the geographic component of the transport services was, “the points of origin and of destination of the transport services concerning the transatlantic route”. But the geographic dimension of the market (described as the “geographic market”) was as the Commission had defined it (and as we have quoted above) limited to the catchment areas of the ports in northern Europe. The Court explained at [853] (by reference to Case 27/76 United Brands v Commission [1978] ECR 207 [11], and Case T-65/96 Kish Glass v Commission [2000] ECR II-1885 [81]) that the definition of the geographic dimension of the market:

is intended to determine the territory on which the undertakings concerned are engaged in the supply of the services in question, on which conditions of competition are also sufficiently homogeneous and which may be distinguished from neighbouring geographical areas because, in particular, the conditions of competition there are significantly different.

146    Another argument by the liner companies before the Court of First Instance was (as described at [884]) that inland transport services between northern Europe and the United States should have been included as part of the market for maritime transport services. The liner companies alleged that these inland transport services should have been included because they alleged that the Commission had found at [84] of its decision that the geographic dimension of the market was, “sea routes between the ports of northern Europe and those of the United States of America and Canada”. Hence, the liner companies argued, they did not occupy a dominant position in the market as so defined.

147    This argument was also rejected by the Court of First Instance. The Court reiterated the fallacy upon which this argument depended. The fallacy was that [84] of the Commission’s decision did not define the relevant geographic dimension of the market. It defined only the geographic component of the relevant maritime services. In the course of rejecting this submission, the Court of First Instance reiterated that “the geographic market … [consists] of the area in which the maritime transport services … are marketed, that is, in this case, the catchment areas of the ports in northern Europe” ([887]).

The identified market in this case

148    When all the dimensions of each market are taken into account (and since we have rejected the respondent airlines’ submission that shippers in Australia did not form part of the market), the proper identification of each of the markets in relation to Hong Kong (which were the focus of all submissions) is as follows. The services to each port in Australia, including transport, involved a market between airlines, freight forwarders, exporters and (at least) including some large shippers, for a suite of air cargo services between Hong Kong and that port in Australia.

(6) Characterisation of whether the identified market is “in Australia”

149    We move then from the issues involved in the identification of a market to the connected issue of whether the market is in Australia.

150    In considering whether an identified market is “in Australia” for the purposes of s 4E, one must understand the nature of the concept which one seeks to locate. Although the issue of whether the market is in Australia is intimately connected to the identification of the market, as we have said it is not any single dimension of the market that must be considered in determining whether it is “in Australia” but the market itself. Unless the overall dimensions of the market (and perhaps the relevant content of the market) are known, it may be difficult to determine whether it is in Australia.

151    Our point, in short, is that one addresses the characterisation question by reference to a “market” not by reference only to some part of the market identification exercise. Market identification is part of an economic analysis. The characterisation question posed by s 4E serves quite a different purpose. It limits the extent to which Parliament has chosen to exercise its legislative power in relation to a market.

152    A somewhat similar question was addressed by this Court in GTK Trading Pty Ltd v Export Development Grants Board (1981) 4 ALD 389; 40 ALR 375. The Full Court (Evatt, Deane and Ellicott JJ) considered whether lobsters, caught by Australian fishermen in waters lying between the coast and the edge of the continental shelf, were produced or processed in Australia. The waters above the continental shelf were not within the definition of the term “Australia” in s 15B of the Acts Interpretation Act 1901 (Cth). At 396 their Honours said:

In the context of an Act which, as we interpret it, is wide enough to confer benefits in relation to the export of fish and lobsters, the words “in Australia” in s 4 should not, in our opinion, be given a strict geographical meaning but one wide enough to cover fish and lobsters which are caught in waters above the Australian continental shelf.

153    The “strict geographical meaning” was, “in Australia, including coastal waters, but not waters beyond coastal waters and above the continental shelf”. We cite this case only as indicating that a “strict geographical” meaning need not be adopted if the proper construction of the relevant statutory provision suggests otherwise.

154    There is another reason why this characterisation exercise should not be limited by any one dimension of the market. The notion of a market is purely conceptual. Market identification is based upon evidence as to actual conduct by buyers and sellers, and actual transactions. It also recognises potential market participants. Part of the relevant conduct will be choices and the reasons for them. To an extent the market concept is a metaphor. It is difficult to see how one can locate a metaphor in a strictly geographical sense.

155    The text of s 4E provides that, “unless the contrary intention appears, market means a market in Australia” (emphasis added). The preposition “in” may have a wide variety of meanings. The Oxford English Dictionary (2nd ed, Oxford University Press, 1989) describes the word in its “general sense” as:

… expressing the relation of inclusion, situation, position, existence, or action, within limits of space, time, condition, circumstances, etc …

That dictionary then states, under the heading “Of position or location”:

Primarily in (of position) is opposed to out of; anything that is in a given space is not out of it, and vice versa.

Another definition is:

Of place or position in space or anything having material extension: within the limits or bounds of, within (any place or thing).

Examples of the word’s use with the names of countries and other geographical features are given.

156    Two points emerge from that definition. First, the superficially trite proposition that “in” is the opposite of “out”, and vice versa, supports the view that s 4E does not require proof that the market was only in Australia. It is only necessary that it be shown that the market is not “out of Australia”. The second point is that, as we have said, since the word “market” is used as a metaphor, it is difficult to see how the question can be purely geographical. Unless a metaphor can be geographically located in Australia, a broader approach should be taken to the meaning of the expression “in Australia”. The Trade Practices Act should be construed so as to facilitate achievement of its stated objective. It is difficult to see how a narrower interpretation of s 4E would assist in such achievement. The better approach is, in effect, to “visualise” the metaphorical market, having regard to all of its dimensions and its content, and then to consider whether it is within Australia, in the sense that at least part (perhaps a substantial or significant part) of it must be in that “location”.

157    As part of the characterisation exercise, we therefore accept the Commission’s submission that the test for the market in Australia cannot focus exclusively upon matters such as the presence of the supplier at the origin of the service, and that among other factors it is necessary to identify where suppliers must operate in order to satisfy the relevant demand for the product (here, a suite of services).

158    However we reject the Commission’s submission that the sphere of operation of a service is conclusive of the geographic dimension of the market. In other words, we reject the Commission’s submission that the geographic dimension of the market, in this case, includes Australia simply because part of the suite of services, even an important part, is provided in Australia. There need not be “one to one correspondence” of the product dimension and geographic dimension of the market, a concept rightly rejected by Professor Gilbert.

159    The fallacy of assimilating the product description with the geographic dimension of the market was also exposed by the Court of First Instance (third chamber) in Atlantic Container Line AB v Commission (TAA). Such assimilation is also inconsistent with the approach taken by French and Spender JJ in Singapore Airlines Ltd v Taprobane Tours WA Pty Ltd. The geographic dimension of the market in Taprobane Tours was Australia. It was not Bali, Fiji, Tahiti, New Caledonia, Hawaii, Penang, the Philippines, the Solomon Islands, the Barrier Reef islands, Guam, and Mauritius. The submission is not supported by any Australian authority. And it is similar to the submission described as a “fallacy” by the Court of First Instance in the European Court of Justice.

160    The central question posed by s 4E, read with ss 45 and 45A, is whether the Hong Kong to Australian port markets for the suite of air cargo services (and involving the participants to which we have referred) were markets “in Australia”. We identify seven overlapping reasons for concluding that the identified markets for the suite of air cargo services between Hong Kong and Australian ports should be characterised as being “in Australia”. These reasons apply even if the question is to be treated as one of locating the “geographic dimension” of the identified market. But, as we have explained, the proper construction of s 4E involves (i) identifying the relevant market including each of its dimensions; and (ii) characterising whether that market is in Australia.

161    In summary, our seven reasons are as follows.

162    First, as we have explained, a market could be “in Australia” even if the market were also in another country. Hence, the issue here is not a question of deciding between whether the market is in Australia or Hong Kong.

163    Secondly, neither the legislative text of s 4E of the Trade Practices Act, when read with ss 45 and 45A, nor the authorities, preclude the consideration of such factors as the presence of customers in Australia to whom the services were marketed, and the fact that the services involved performance in Australia. Rather, as we have explained, the legislative text assumes that any relevant aspect will be considered in the characterisation exercise.

164    Thirdly, a significant and important part of the operation of the “suite of services” being provided was in Australia. As we have explained in our discussion of QCMA, actual supply also occurs in the market. The cargo was transported to Australia, ground handling services were provided in Australia and there were enquiry services in Australia (which involved tracing delayed or lost shipments and identifying and dealing with damaged shipments). The airlines competed in the supply of these services. It cannot seriously be suggested that the freight forwarders in Hong Kong had no interest in the quality of the service provided to the customers in Australia.

165    No doubt the delivery of cargo to the airline in Hong Kong evidenced a choice between or amongst substitutable services, although the making of the contract to carry would also have done so. It is true that for the carriage of cargo from Hong Kong to an Australian destination, the cargo must be delivered to the airline in Hong Kong, unless the same airline has brought the cargo to Hong Kong, a fact situation which seems not to have been considered. However, as we have pointed out, the geographical aspect of the product dimension dictates use of an airline which flies between Hong Kong and the relevant Australian port. The whole of the suite of services is no more supplied in Hong Kong than it is at that Australian port. Taking delivery of the cargo in Hong Kong is no more important in the provision of that suite of services than is the flight itself, and the delivery of the cargo at the Australian port.

166    Fourthly, the suite of services provided by the airlines involved barriers to entry in Australia. Those barriers included matters such as availability of landing slots, licences to operate in Australia, permission to build or to use facilities for perishable cargo, permission to operate additional flights and so on.

167    Fifthly, the services were marketed in Australia to shippers who (as a matter of economic reality) were customers of the airlines. The airlines competed for business in Australia. Some shippers were in Australia and were capable of operating as a constraint on the fixing of cargo rates. Although the primary Judge did not need to decide whether the freight forwarders were agents for the importing shippers in Australia, he concluded that as a matter of economic substance they were intermediaries, having fluctuating control over the cargo the carriage of which they arranged.

168    Sixthly, the purpose of s 4E supports the evaluative conclusion that the market in this case is a market “in Australia”. That purpose must be applied to the context which involves (i) marketing to persons in Australia who, as a matter of economic reality are customers; and (ii) of a suite of services which include performance in Australia. The legislative purpose is, “to enhance the welfare of Australians through the promotion of competition and fair trading and provision for consumer protection”.

169    Seventhly, the conclusion that the market is “in Australia” is consistent with the conclusion reached upon similar fact patterns in New Zealand and in Europe.

170    For these reasons we conclude that the markets for the suite of air cargo services from Hong Kong to Australian ports, involving airlines, freight forwarders, exporters and (at least) including some large shippers, were markets “in Australia”. It is common ground that the conclusion in relation to the market for air cargo services from Hong Kong also applies to the Singapore and Indonesia markets.

THE RESPONDENT AIRLINES’ NOTICES OF CONTENTION

171    The notices of contention from the respondent airlines, particularly Garuda, were not always easy to follow. Air NZ had 11 issues in its notice of contention. Garuda had 94. The issues in the notices of contention were not always clearly matched to submissions. Garuda’s written submissions, for example, sometimes raised various themes which were only loosely tied, in a heading, to grounds of the notice of contention. The Commission submitted, in written submissions, that Garuda had not addressed grounds 17, 19, 25(b)-(d), 27, 29-30, 33, 34, 36(e), 45-46 and 69. In reply, Garuda did not attempt any particular consideration of those matters. It accepted only that grounds 19, 29, and 30 had been abandoned. Our reasons which follow are structured around the written and oral submissions. Where those submissions made reference to grounds of contention then we have set those out.

Alleged inconsistency between the Trade Practices Act 1974 (Cth) and the Air Navigation Act 1920 (Cth)

172    This issue was the subject of various parts of Garuda’s notice of contention (grounds 10-18 and 77-78). It was not directly raised in the submissions of Air NZ so it is convenient to focus on the submissions of Garuda.

173    As Garuda explained the point in its submissions, the issue concerned whether ss 45 and 45A of the Trade Practices Act should be construed to resolve conflict with the provisions of the Air Navigation Act 1920 (Cth) (the “Air Navigation Act”) and Australia’s treaty obligations. A construction in this manner was said to be one which did not proscribe the making or giving effect to any contract, arrangement or understanding made or entered into between international airlines concerning tariffs for scheduled international air services.

174    The relevant arguments made on this point at trial were summarised by the primary Judge at [129] as follows:

(1)    the background to the regulation of international commercial aviation showed that at the time of the introduction of the Trade Practices Act, the majority of bilateral air transport agreements provided for rate fixing machinery;

(2)    a consideration of the terms of the Air Navigation Act showed that the Trade Practices Act could never apply to international commercial aviation;

(3)    alternatively, the Trade Practices Act could not apply where Australia had entered into an Air Services Agreement (“ASA”) with another nation which permitted or required rate or tariff fixing; and

(4)    consequently, the Trade Practices Act could not apply to the conduct alleged against the respondent airlines in the period 2001-2006.

175    In 1974, at the time when the Trade Practices Act was enacted, the Air Navigation Act contained the following provisions.

176    Section 12 provided:

12.    International airline licences

(1)    An international airline of a country other than Australia shall not operate a scheduled international air service over or into Australian territory except in accordance with an international airline licence issued by the Director-General in accordance with the regulations.

(2)    An international airline licence shall not be granted to an international airline of a country other than Australia unless that country and Australia are parties to the Air Transit Agreement, or to some other agreement or arrangement, whether bilateral or multilateral, under which scheduled international air services of that other country may, subject to the agreement or arrangement, be operated over or into Australian territory.

177    Section 13 provided:

13.    Suspension or cancellation of international airline licences

The Minister may suspend or cancel an international airline licence issued to an international airline of a country other than Australia if and only if –

(a)    the airline or any aircraft operated by the airline fails to comply with a provision of this Act or the regulations or the terms of its licence; or

(b)    the airline fails to conform to, or comply with, any term or condition of the relevant agreement or arrangement referred to in the last preceding section.

178    Section 22(1) made it an offence for any person to contravene or fail to comply with a provision of the Air Navigation Act. The penalties include imprisonment and fines.

179    One of these treaties, or “agreements or arrangements” within the meaning of ss 12(2) and 13(b), was the 1969 Air Services Agreement between Australia and Indonesia entitled Agreement between the Government of the Commonwealth of Australia and the Government of the Republic of Indonesia for Air Services Between and Beyond their Respective Territories, 7 March 1969, [1969] ATS 4 (the “Australia-Indonesia ASA”). The Australia-Indonesia ASA has the following features:

(1)    Article 2 provides for the rights for designated airlines (including Garuda) to make stops (in cities including Sydney, Darwin, and Melbourne as provided in Annex, Section II) for purposes including putting down and taking international cargo; and

(2)    Article 6 then provides as follows:

(1)    The tariffs on any agreed service shall be established at reasonable levels, due regard being paid to all relevant factors including the cost of operation, reasonable profit, characteristics of service (such as standards of speed and accommodation) and the tariffs of other airlines for any part of the specified route. These tariffs shall be fixed in accordance with the following provisions of this Article.

(2)    Agreement on the tariffs shall, whenever possible, be reached by the designated airlines concerned through the rate-fixing machinery of the International Air Transport Association. When this is not possible, tariffs in respect of each of the specified routes shall be agreed upon between the designated airlines concerned. In any case the tariffs shall be subject to the approval of the aeronautical authorities of both Contracting Parties.

(3)    If the designated airlines concerned cannot agree on the tariffs, or if the aeronautical authorities of either Contracting Party do not approve the tariffs submitted to them in accordance with the provisions of paragraph (2) of this Article, the aeronautical authorities of the Contracting Parties shall endeavour to reach agreement on those tariffs.

(4)    If agreement under paragraph (3) of this Article cannot be reached, the dispute shall be settled in accordance with the provisions of Article 9 of this Agreement.

(5)    No new or amended tariff shall come into effect unless it is approved by the aeronautical authorities of both Contracting Parties or is determined by a tribunal of arbitrators under Article 9 of this Agreement. Pending determination of the tariffs in accordance with the provisions of this Article, the tariffs already in force shall apply.

180    It was common ground that a “tariff” includes all charges, including all the levies and surcharges in question in this appeal.

181    The primary Judge’s reasons comprehensively described the background to the Australia-Indonesia ASA. These reasons can be summarised as follows.

(1)    Following a failure to achieve a multilateral agreement in the 1944 Chicago Convention on International Civil Aviation, signed 7 December 1944, 15 UNTS 295 (entered into force 4 April 1947), international aviation operations were conducted by bilateral agreements of which the first and most important was between the United States and United Kingdom (Bermuda I).

(2)    Bermuda I made provision for the two States to agree, within their respective constitutional powers and obligations, upon rates to be charged by air carriers, and in the event of disagreement a process be used which included the processes of the International Air Transport Association (“IATA”, which is the international trade association for airlines). The rates fixed were to be reasonable having regard to all relevant factors including cost of operation, reasonable profit and the rates charged by any other air carriers.

(3)    IATA obtained antitrust immunity because provisions of the Federal Aviation Act of 1958, 49 USC 1301 (1958) authorised the Civil Aviation Board to disapprove arrangements that were adverse to the public interest. After 1979, the Civil Aviation Board had discretion to grant antitrust immunity.

(4)    At the time of the introduction of Pt IV of the Trade Practices Act in 1974, the vast majority of bilateral air transport arrangements provided for rate fixing machinery of the nature of Bermuda I ([148]). Although this point was disputed by the Commission, Garuda provided a schedule which showed that between 1946 and 1984 Australia entered 36 different “agreements or arrangements” within the meaning of s 13(b) and s 12(2), by entering treaties with those countries. All but five of these treaties contained tariff provisions of a similar nature to those described below. In 1974, at the time of the enactment of the Trade Practices Act, there were 24 of these treaties in force.

The reasoning of the primary Judge

182    The primary Judge’s conclusions on this point were as follows.

(1)    Although a breach of Article 6 of the Australia-Indonesia ASA by Garuda was not a breach of the Air Navigation Act (and hence was not an offence), such a breach would make Garuda liable to lose its licence with the consequence that it would have to either cease operation, or commit a serious offence. Section 13(b) of the Air Navigation Act does, therefore, impose an obligation upon Garuda to comply with the Australia-Indonesia ASA ([152]).

(2)    The Air Navigation Act in 1974 did not require or authorise rate fixing of the kind contemplated in Bermuda I ([156]). It required carriers to comply with any bilateral Air Services Agreement (“ASA”). But it was silent upon the terms of the ASAs ([158]). The operation of the Air Navigation Act was not inherently inimical to the operation of Pt IV of the Trade Practices Act ([160]).

(3)    However, the combined operation of Article 6(2) with ss 12 and 13 of the Air Navigation Act had the effect, in their practical operation, of requiring collusive behaviour of the kind prohibited by Pt IV of the Trade Practices Act when it was enacted ([165]).

(4)    Assuming, without deciding, that the effect of (3) is that the application of Pt IV of the Trade Practices Act to Garuda would put Australia in breach of its obligations at public international law ([181]), it is not possible to interpret price fixing under the Air Navigation Act as an act or thing that is “specifically authorised or approved” by that Act so that it would fall within an exemption in s 51 of the Trade Practices Act ([183], applying Re Ku-ring-gai Co-operative Building Society (No 12) Ltd (1978) 36 FLR 134.

(5)    The conflict in practical operation between Pt IV of the Trade Practices Act and the effect of the Air Navigation Act on the Australia-Indonesia ASA is that an adjustment must be made to give the two statutory schemes a harmonious operation. The appropriate adjustment is that s 13 of the Air Navigation Act does not authorise the Minister to cancel or suspend a licence for breach of an ASA when the conduct constituting the breach was required by Pt IV of the Trade Practices Act ([185]).

(6)    The amendments to s 51(1) of the Trade Practices Act by the Competition Policy Reform Act 1995 (Cth), which introduced s 51(1C), reinforce this conclusion because that new subsection required the authorising provision to make express reference to the Trade Practices Act and there was no express reference in the Air Navigation Act to the Trade Practices Act ([193]). Even if the Trade Practices Act had not extended to international civil aviation prior to 1995, the effect of ss 51(1A) and 51(1C) was to extend it to international civil aviation thereafter ([204]).

(7)    Neither Indonesian domestic law nor Hong Kong law requires compliance with the relevant ASA so there is not a conflict where one State prohibits that which another State requires ([186]).

(8)    The terms of Garuda’s licence did not require it to engage in rate fixing ([195]) but such a requirement would not be an authorisation or approval under the Air Navigation Act ([194]).

183    The primary Judge also considered, in some detail, the Air Navigation Regulations 1947 (Cth) (“Air Navigation Regulations”), including reg 258 that permits the Director-General (rather than the Minister) to cancel a notice in circumstances which include a failure to comply with conditions including Article 6. Senior counsel for Garuda accepted that this did not add anything to the effect of the Air Navigation Act for the purposes of the submission (app ts 190). It suffices to put the regulations to one side.

184    As explained above, the primary Judge concluded that there was no inconsistency between the terms of the Air Navigation Act and the Trade Practices Act. He held that there was only inconsistency in their practical operation. He also concluded that one reason why the inconsistency in practical operation was not resolved by excluding international commercial aviation from the Trade Practices Act was due to s 51 of the Trade Practices Act and, after 1995, ss 51(1A) and 51(1C) which provided for limited exemptions from the operation of Pt IV, in circumstances which did not include the operation of Article 6 of the Australia-Indonesia ASA.

The respondent airlines’ submissions

185    As we have explained, the submissions on this issue were made by Garuda. Garuda’s approach on this appeal was effectively that inconsistency existed:

(1)    between the terms of the Trade Practices Act and the Air Navigation Act;

(2)    between the terms of the Trade Practices Act and the practical effect of the Air Navigation Act due to the terms of the ASA; and

(3)    between the terms of the Trade Practices Act and the practical effect of the Air Navigation Act due to the terms of Garuda’s airline licence.

186    Garuda also submitted that s 51 of the Trade Practices Act was concerned with circumstances to be taken into account when considering contraventions of a provision of Pt IV of the Trade Practices Act. It was not concerned with whether Pt IV applied at all in cases of conflict with a different statute. To put the matter another way, the Air Navigation Act was not a statute which provided “authorisation or approval” for a particular act which would otherwise contravene Pt IV of the Trade Practices Act. Instead, the Air Navigation Act was a specific statute that required conduct that the general terms of the Trade Practices Act would otherwise have prohibited.

187    Each of these submissions is rejected in the sections below. We then explain how even if some inconsistency had existed between the terms of the Trade Practices Act and the practical effect of the Air Navigation Act, this would not have necessarily implied into s 45 of the Trade Practices Act words which excluded the operation of international commercial aviation.

There is no inconsistency in the terms of the Trade Practices Act 1974 (Cth) and Air Navigation Act 1920 (Cth)

188    As to whether there was inconsistency at the level of the statutes we consider, with respect, that the primary Judge was correct in his conclusion and reasoning that there is no inconsistency in principle between the terms of s 13(b) of the Air Navigation Act and the Trade Practices Act. As French CJ and Kiefel J said in Firebird Global Master Fund II Ltd v Republic of Nauru [2015] HCA 43; (2015) 326 ALR 396, 416 [87] (footnote omitted):

For a court to conclude that a later statute impliedly repeals an earlier statute the court must be satisfied that the two statutes are so inconsistent that they cannot stand or live together. This will be so only if the provisions of the two statutes cannot be reconciled.

189    The effect of s 13(b) of the Air Navigation Act was that Garuda was exposed to the possibility of loss of its licence, at the discretion of the Minister, if it failed to comply with conditions in the ASA. But the Air Navigation Act did not create new legislative duties when, in 1969, Australia entered the ASA with Indonesia. Instead, the operation of the existing duty in s 13(b) was expanded because the entry into the ASA meant that the Minister’s discretion then included a power to take action where Garuda had failed to comply with any terms or conditions of the Australia-Indonesia ASA.

190    There is a distinction between (i) the Australia-Indonesia ASA expanding the circumstances in which the licence might be cancelled; and (ii) the Australia-Indonesia ASA itself creating a new duty. As to (ii), the Australia-Indonesia ASA could not itself create a new duty imposed upon an entity like Garuda who is not a party to it. Nor could it create a new duty by somehow amending the terms of the Air Navigation Act. Speaking of this suggestion of an international instrument amending domestic legislation, French CJ said in Maloney v The Queen [2013] HCA 28; (2013) 252 CLR 168, 182 [15]:

Obligations imposed by international instruments on States do not necessarily take account of the division of functions between their branches of government. The difficulty is compounded when the interpretation of the international instrument is said to have been subject to change by reference to practices occurring since the enactment of legislative provisions implementing it into domestic law. Such practices may, by operation of Art 31(3) of the Vienna Convention, be taken into account in interpretation of the treaty or convention for the purposes of international law. They may lead to its informal modification. However, they cannot be invoked, in this country, so as to authorise a court to alter the meaning of a domestic law implementing a provision of a treaty or convention. (Footnotes omitted)

191    There is therefore no inconsistency between the terms of the Air Navigation Act and the Trade Practices Act. Any inconsistency that could exist could only arise as a result of the practical operation of the Air Navigation Act rather than the terms of s 13(b).

192    Garuda also relied upon the decision in Commissioner of Police for New South Wales v Eaton [2013] HCA 2; (2013) 252 CLR 1 for its submission that there was an inconsistency between the Air Navigation Act and the Trade Practices Act which should be resolved in favour of disapplying the Trade Practices Act in circumstances of international aviation. That case involved a conflict between s 80(3) of the Police Act 1990 (NSW) and s 84(1) of the Industrial Relations Act 1996 (NSW). The Police Act gave the Commissioner of Police a power to dismiss a probationary police officer at any time without giving reasons. The Industrial Relations Act provided for the review of a dismissal that was “harsh, unreasonable or unjust”. This included dismissal of public sector employees who were defined to include a member of the New South Wales Police Force. In a joint judgment the majority, Crennan, Kiefel and Bell JJ, held that the Police Act prevailed: the Industrial Relations Act was a general statute whose provisions were not intended to prevail over the earlier, special provisions of the Police Act without manifesting that intention very clearly.

193    Apart from the lack of any inconsistency at the level of the statutes in this case, the conclusion in Eaton also concerned a wholly different circumstance from this case. These differences, as described in the majority judgments, include:

(1)    The language of the relevant provisions in the Police Act pointed strongly against a power to review a dismissal with the decision of the Police Commissioner described as one that can be made “at any time”, “without giving any reason”, and the description of the officer as “probationary”. In contrast, there is no language in the Air Navigation Act which suggests any exclusion of the terms of Pt IV of the Trade Practices Act and the language of the Australia-Indonesia ASA, when introduced in 1969, is directed to the parties to that treaty (Australia and Indonesia) not to international airlines.

(2)    Anomalies arose within the Police Act if the Industrial Relations Act had qualified the broad power of the Commissioner. For instance, it would put probationary police officers in a better position than non-probationary officers. No internal inconsistency would exist in the Air Navigation Act if the Trade Practices Act did not exclude international commercial aviation.

(3)    The only point of comparison between the statutes in Eaton and those in this case is that Eaton involved a question whether a later general statute (ie the Trade Practices Act) was inconsistent with an earlier specific statute (ie the Air Navigation Act). But Garuda’s submissions did not seek to draw this comparison. Garuda submitted that the amendment to the provisions in the Air Navigation Act (including a 1989 amendment to s 13(b) to extend to the Minister a power to vary, as well as suspend or cancel, a licence) meant that the Air Navigation Act was somehow a later statute than the Trade Practices Act.

194    At some points in its written submissions before this Court, Garuda also referred to the Air Navigation Regulations, apparently in support of the submission that there was an inconsistency because the regulations required the conduct which Pt IV of the Trade Practices Act prohibited. In written submissions, Garuda relied upon the decision of the High Court in R v Halton; Ex parte AUS Student Travel Pty Ltd [1978] HCA 26; (1978) 138 CLR 201. That case was concerned with a prosecution for an offence of charging less than an approved tariff which, in that case, had been approved by IATA. The relevant regulation in the Air Navigation Regulations was reg 106A which prohibited discounting from the approved tariff. Several points can be made about Garuda’s reliance upon that case.

195    First, as the primary Judge explained, Halton was not concerned with price fixing ([155]). It was concerned with charging less than an agreed tariff. That prohibition against discounting is different from an alleged “requirement to engage in price fixing”.

196    Secondly, as the primary Judge observed, reg 106A (renumbered ANR 19 of 1999) was amended on 20 December 2000 by the Air Navigation Amendment Regulations 2000 (No 3) (Cth) which permitted, but did not require, lodgement by an airline of its tariff which the Secretary could in his or her discretion approve, reject or vary by subjecting it to conditions.

197    Thirdly, as the primary Judge also observed at [1247], in September 1981 the Commonwealth formally announced that it would not prosecute the offence involving discounting from an agreed tariff.

There is no inconsistency in the practical effect of the statutes due to Article 6, ASA

198    Although there is no conflict between the terms of the Air Navigation Act and the Trade Practices Act, Garuda submitted that there are potential inconsistencies between the terms of the Trade Practices Act and the effect or operation of the Air Navigation Act due to the terms of Article 6(2) of the Australia-Indonesia ASA. The potential inconsistency in effect was said to arise because an international airline is exposed to the possibility of the loss of its licence under s 13 of the Air Navigation Act for failure to follow the procedure in Article 6(2) when that procedure is prohibited by ss 45 and 45A of the Trade Practices Act.