Consideration
623 The cases are replete with general statements of principle about the notion of the "market" in competition law and the process of defining markets for competition law purposes. The statements of principle are applicable to the present case. But, being directed to concepts, they do not provide a ready answer to the question of where a market is to be regarded as located in circumstances where the acquired product is a service provided or delivered transnationally. It would be true to say, as the primary Judge appeared to suggest (at [317]), that the language employed in the cases to describe these concepts can be interpreted in ways that might be thought to support each of the competing positions that were advanced by the parties.
624 Thus, to speak of a market as a "field of rivalry" (see, for example, QCMA at 517) or as "the geographic space in which rivalry and competition take place" (Australian Competition and Consumer Commission, Merger Guidelines - November 2008 (Commonwealth of Australia, Canberra, 2011 at [4.6])) is to provide a convenient metaphor for something that is "an analytical tool devised by economists, not a feature of the real world": Seven Network Ltd v News Ltd (2009) 182 FCR 160; [2009] FCAFC 166 (Seven Network) at [827]. But in order for the metaphor to be meaningful, especially when the market concept is instrumental to establishing legal liability, more is required. The quest for meaning is advanced by overlaying other economic concepts. One indispensable concept is substitution. Indeed, in Queensland Wire, Mason CJ and Wilson J at 188 referred to substitution as the defining feature of a market. As I have noted, the primary Judge approached the task of market definition through what he described as the "prism" of substitution. There is no suggestion in these appeals that this was an error. All parties advocated the correctness of that approach.
625 Substitution, as an economic concept, speaks of the possibilities for substitution brought about by the influence and exercise of market power. In the present case, as in a number of other cases before the Court, the economic evidence proceeded with an acceptance that the hypothetical monopolist test is an accepted methodology for market definition. The test involves determining whether a hypothetical monopolist supplier in a given area could profitably impose a small but significant non-transitory increase in price - a SSNIP. Starting with the firm and product in issue, the market borders are expanded to include all sources of close substitutes that would defeat the increase. The product and geographic dimensions of the market are represented by the smallest area over which the hypothetical monopolist (who can also be conceived of as all suppliers of a particular product acting collusively) can profitably impose the increase: Metcash at [247]-[251]; Seven Network at [625]; Re Tooth & Co Ltd (1979) 39 FLR 1 (Re Tooth) at 38-39.
626 In QCMA, the Trade Practices Tribunal said (at 517):
It is the possibilities of such substitution which set the limits upon a firm's ability to "give less and charge more". Accordingly, in determining the outer boundaries of the market we ask a quite simple but fundamental question: If the firm were to "give less and charge more" would there be, to put the matter colloquially, much of a reaction? And if so, from whom? In the language of economics the question is this: From which products and which activities could we expect of relatively high demand or supply response to price change, ie a relatively high cross-elasticity of demand or cross-elasticity of supply?
627 How is substitution played out? This is typically seen as substitution between one product and another and between one source of supply and another. In QCMA, the Tribunal described (at 517) a market by reference to the dynamics of substitution in the following way:
So a market is the field of actual and potential transactions between buyers and sellers amongst whom there can be strong substitution, at least in the long run, if given a sufficient price incentive. Let us suppose that the price of one supplier goes up. Then on the demand side buyers may switch their patronage from this firm's product to another, or from this geographic source of supply to another. As well, on the supply side, sellers can adjust their production plans, substituting one product for another in their output mix, or substituting one geographic source of supply for another. Whether such substitution is feasible or likely depends ultimately on customer attitudes, technology, distance, and cost and price incentives.
628 The Tribunal's description of a market not just as a "field of rivalry" but as a "field of actual and potential transactions between buyers and sellers amongst whom there can be strong substitution" cannot be read as a casual observation. The Tribunal's description focuses attention on the area in which transactions occur as setting the boundaries of the market. This understanding also assists in identifying the location of a market. I do not think that, by the word "transactions", the Tribunal was referring to some limiting legal concept such as might apply when one is determining, for example, the place where a contract is made. Rather, the Tribunal must be taken as using the word "transactions" in a broader economic sense to refer to the exchanges between buyers and sellers to buy and sell the relevant market product (ie, the substitutable products within the market), and the switching and substitution that takes place in the course of those exchanges. Thus, speaking geographically, a market is the field or area in which these exchanges take place, and over which switching and substitution by and between buyers and sellers occur. This focus serves to distinguish the area in which and over which these activities take place from the product or products that are the subject matter of these activities. Hence the relevance of speaking of separate product and geographic dimensions of the market.
629 In Re Tooth, the Trade Practices Tribunal referred (at 38) to a market as corresponding to the activities and geographic area within which, if given a sufficient economic incentive, switching occurs by buyers on the demand side and sellers on the supply side.
630 In its First Annual Report for the year ended 30 June 1975, the Trade Practices Commission described (at [3.54]) the geographic dimension of a market in the following way:
Geographic Market
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. The internal organisation of an enterprise may itself cast light on the dimensions of the geographic market …
(Emphasis added.)
631 This description (which was also apparently given in the Trade Practices Commission's Second Annual Report) was accepted by Spender J in QIW Retailers Limited v Davids Holdings Pty Limited (No 3) (1993) 42 FCR 255; [1993] FCA 287 at 267.
632 Similarly, in Australian Competition and Consumer Commission v Cement Australia Pty Ltd (2013) 310 ALR 165; [2013] FCA 909, Greenwood J described (at [972]) the geographic dimension of a market as:
… the area within which sellers offer the product for sale and includes areas of substitutable geographic sources of supply buyers might turn to in switching their patronage to another supplier of the same or substitutable product. …
633 In Queensland Wire, Dawson J (at 198-199) remarked that the concept of a market is sometimes dealt with in a more complex manner than is necessary. His Honour 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 exchange by employing price as the mechanism to reconcile competing demands for resources … In setting the limits of a market the emphasis has historically been placed upon what is referred to as the "demand side", but more recently the "supply side" has also come to be regarded as significant. The basic test involves the ascertainment of the cross-elasticities of both supply and demand, that is to say, the extent to which the supply of or demand for a product responds to a change in the price of another product. Cross-elasticities of supply and demand reveal the degree to which one product may be substituted for another, an important consideration in any definition of a market. This is reflected in s 4E of the Trade Practices Act …
634 These descriptions of the geographic market, although expressed in various ways, are all consistent with the description in QCMA of a market as the field of actual and potential transactions between buyers and sellers amongst whom there can be strong substitution.
635 The Commission placed considerable emphasis on the geographic element of the product dimension of the markets in question as determining, as well, their respective geographic dimensions and locations. Thus, it referred to the notion of an origin/destination "pair", such that the mere fact that a port of destination is located in Australia meant, without more, that the market was located, at least, in Australia. Relatedly, the Commission placed considerable emphasis on the ways in which competition may be expressed or manifested. It relied on the fact that expressions or manifestations of rivalry in the form of the quality of services provided in Australia as part of the "price product service package" meant that the market was, at least, located in Australia.
636 In order to evaluate this approach it is necessary to distinguish between the following concepts.
637 The first distinction is the one to which I have already referred. It is the distinction between the transactions between buyers and sellers to buy and sell the market product, and the product itself. The product is the subject matter of these transactions but there is no necessary correlation between the location of the product (including where a service is physically provided or delivered) and the location of the field of transactions between buyers and sellers in respect of the product. The distinction between the two concepts should not be eroded simply because the product is a service which has a geographical element in terms of the place or places at which the service is or is to be physically provided or delivered.
638 The second distinction concerns the product itself. Here, the product is the service of flying cargo from a point of origin, say Hong Kong, to a point of destination in Australia, say Sydney. The product includes what the primary Judge described as ancillary services, for example ground handling and inquiry services. The product was regarded by the primary Judge as a suite of services, encapsulated by the expressions "air cargo services" or "air freight services". The Commission does not dispute the characterisation of the product as a suite of services. Indeed, it relied upon it. It submitted, correctly, that the product is a suite of services that is only demanded and supplied as such.
639 The point of present significance is that while an airline might compete in its product offering through, for example, superior ground handling services at the port of destination, the product is not ground handling services at that port of destination. Thus, the market is not - to continue the example - a market for ground handling services in Sydney. Rather, it is the suite of services.
640 The primary Judge recognised this distinction, when his Honour said at [321]:
The ground handling services provided at the destination airport in Sydney are, no doubt, a part of a general suite of services making up the service provided by an airline but ground handling services at Sydney are not themselves a substitute for air cargo services from Hong Kong to Sydney any more than a tyre is a car. …
641 The third distinction concerns the ways in which competition may be expressed or manifested. In Seven Network at [585], Dowsett and Lander JJ identified a number of propositions about competition and markets emerging from the cases, including the proposition that "competition in a market is the sum of activity engaged in by persons in promoting the sale [and supply] of the goods or services with which the market is concerned": see also Dandy Power Equipment Pty Ltd v Mercury Marine Pty Ltd (1982) 64 FLR 238 at 259. This is a statement about the way in which competition might be expressed or manifested. However, the place where competition is expressed or manifested is not necessarily the place where the market is located. This is because a market is not simply a field of rivalry but, as QCMA reminds us, a field of rivalry of a particular kind. It is the field of actual and potential transactions between buyers and sellers amongst whom there can be strong substitution. Thus, when the product is a particular suite of services, the market is the field of actual or potential transactions between buyers and sellers for the sale of the suite. This is not necessarily the place at which the suite of services or some part of it is physically provided or delivered in a way that will enable its qualities and attributes to be appraised or evaluated. This was the point made by the primary Judge when he said (at [322]):
If it was sufficient in the case of an international transportation market that it was located where rival services were provided then it would follow that the market in Australia for packaged tours of Europe would be located wherever the tour bus happened to be. …
642 Nevertheless, on appeal, as before the primary Judge, the Commission's case was that if rivalry in an aspect of the "price product service package" was played out in the manner of delivery at a particular location, and that location was Australia, it followed inexorably that there was a market in Australia. Moreover, this was so even though the aspect of the "price product service package", as so delivered, was not in fact "the product" with which the market is concerned. I am unable to accept those propositions.
643 The focus of market definition must be the identification and analysis of substitution possibilities. It is helpful to consider this from the model provided by the hypothetical monopolist test. Once the hypothetical monopolist controls all the alternative locations from which buyers can obtain the product in question (relevantly, the suite of services) and to which buyers would switch in order to avoid the imposition of a SSNIP, it can profit-maximise by imposing the SSNIP because the buyers of the product have nowhere left to turn to avoid its effect. As the airlines submitted, this proposition holds good regardless of where the services are, in fact, provided. Thus, when the product is a service, the hypothetical monopolist test is not concerned with where the service is provided but with where substitution occurs in respect of the sale and purchase of the service. And substitution occurs on the field of actual and potential transactions between buyers and sellers amongst whom there can be strong substitution.
644 The hypothetical monopolist test, as applied in the present case, is but an aid to arriving at the appropriate market definition. It is not a substitute for evidence. Recently, in Australian Competition and Consumer Commission v Australia and New Zealand Banking Group Ltd (2015) 324 ALR 392; [2015] FCAFC 103, the Court observed (at [138]):
Whilst a market is an analytical or economic tool designed to analyse the particular asserted anti-competitive conduct, a market definition must nonetheless be based on findings of fact: Singapore Airlines Ltd v Taprobane Tours WA Pty Ltd (1991) 33 FCR 158 at 174; 104 ALR 633 at 648 per French J. The premise of that proposition is that it has economic and commercial reality. It must accordingly not be artificial or contrived. Economists frequently construct economic models to analyse complex commercial or economic events or scenarios. But a model is unlikely to be a useful analytical tool if based on unrealistic assumptions that materially depart from the real world facts and circumstances involving commercial behaviour in which the events to be analysed occur. A court should be loathe to accept or act on a market definition which is an artificial construct that does not accurately or realistically describe and reflect the interactions between, and perceptions and actions of, the relevant actors or participants in the alleged market, that is, the commercial community involved.
645 I stress this matter because the primary Judge's conclusion that the relevant markets were not markets in Australia was based on findings that are peculiar to the case at hand. That is the case that must be considered on appeal. It is possible that different findings of fact may have resulted in a different evaluation and a different market definition. But, apart from his Honour's finding that shippers were market participants (which I discuss below), there is no challenge to the findings of primary fact that informed his Honour's conclusions on market definition. The challenge is to his Honour's analysis and evaluation, based on those facts.
646 The primary Judge found that the understandings were arrived at in meetings held, or through conduct occurring, at the port of origin, whether that was in Hong Kong, Singapore or Indonesia. None of the price-fixing conduct occurred in Australia. It was appropriate for the primary Judge to focus attention, firstly, on the port of origin because this was the place where the impugned conduct occurred. As Allsop J observed in Australian Competition and Consumer Commission v Liquorland (Australia) Pty Ltd [2006] ATPR 42-123; [2006] FCA 826 at [437]:
… Market definition is to be approached by beginning with the problem at hand and asking what market identification best assists the assessment of the conduct and its asserted anti-competitive attributes…
647 For the purposes of present analysis, the product dimension of each relevant market is not in dispute. Once again, I emphasise that it concerns a suite of services that is only demanded and supplied as a suite. This is the market "product".
648 As I have noted (at [605] above), the primary Judge found that the participants in the relevant markets were airlines, freight forwarders and shippers (either exporters at origin or importers at destination) whose cargo volume was sufficiently significant for airlines to be commercially motivated to pursue it.
649 With respect to the supply side, the primary Judge found that there would be no airlines seeking to enter the relevant markets if the incumbents imposed a SSNIP.
650 With respect to the demand side, the primary Judge made the following findings. The range of airlines available to be selected was limited by the need to have a presence in the port of origin where possession of the cargo to be transported was taken from the freight forwarder. The service of taking possession of the cargo in the port of origin with a view to flying it to a port of destination in Australia could not be performed anywhere but in the port of origin. Each local cargo sales office of Air NZ and Garuda at the port of origin published from time to time standard rates as "tariff" or "rate" sheets or schedules. Contract rates (as opposed to standard rates) were negotiated between freight forwarders and staff of the airlines at the local sales office. Even though shippers were market participants, and some may have made, in Australia, decisions about which airlines they would use, the shippers nevertheless continued to use freight forwarders who provided indispensable services in respect of transport issues with which the airlines would not deal. Leaving aside extremely rare occurrences (typically involving live animals), airlines carrying cargo from a port of origin such as Hong Kong (and, by implication, the other ports of origin relevant to the present question) generally dealt directly only with freight forwarders situated in Hong Kong (or the other ports of origin) or in nearby environs. The contractual relationship was between the airline and the freight forwarder, who "cut" or "raised" the air waybill at the port of origin. The airline and the freight forwarder were the parties to the air waybill whose terms governed the carriage of the cargo. In most cases, the freight forwarder at the port of origin was obliged to pay the airline for air cargo transport charges in the local currency of the port of origin, although some shipments may be "charges collect", in which case the freight forwarder at the port of destination paid the charges on delivery. The freight forwarder's obligation to pay the charges was not conditional on it receiving payment from the consignor or consignee.
651 These findings show that all the sources of supply of the market product (the suite of services) are located at the port of origin, not the port of destination. This is where the prices for the product are set and where the product is bought and sold. It is, as I have said, the place where the impugned conduct occurred. These findings thus point persuasively to the field of actual or potential transactions between buyers and sellers being the port of origin in respect of each relevant market, not some other place. Indeed, the Commission does not dispute that there was, in each case, a market at the port of origin. The Commission's case is that there was also, correspondingly, a market in Australia because part of the suite of services was physically provided or delivered in Australia and that the manner in which this part was physically provided or delivered is an aspect of the way in which airlines compete with each other. For the reasons given in [636]-[642] above, I do not accept that these considerations demonstrate that there is a market in Australia. They conflate and confuse the product and geographical dimensions of the markets under consideration.
652 When s 45(2) of the TPA refers to the purpose, effect or likely effect of substantially lessening competition, it means competition "in any market" in which a relevant party supplies or acquires, or is likely to supply or acquire, or would (but for a provision in the contract, arrangement or understanding) supply or acquire or be likely to supply or acquire, goods or services. The inquiry is not simply: Where does the party supply or acquire goods or services or where is the party likely to supply and acquire, or would, or would be likely to supply or acquire, goods and services? It is not enough to attract the operation of s 45 of the TPA to do no more than point to the fact that goods or services are supplied or acquired in Australia. Rather, the provision is directed to "the market" in which the goods or services are supplied or acquired, and to competition in that "market". Thus, the setting of "the market" is crucial. This then becomes the point of reference to which questions concerning the supply or acquisition of goods or services are then directed.
653 Here, the "goods or services" is the suite of services supplied by airlines in transporting cargo from a port of origin to a port of destination. The word "supply" is defined in the TPA to include, in relation to services, "provide, grant or confer": s 4(1) of the TPA. The word "acquire" is defined to include, in relation to services, "accept": s 4(1) of the TPA. In Cook v Pasminco Limited (2000) 99 FCR 548; [2000] FCA 677, Lindgren J remarked (at [26]) that the definitions of "supply" and "acquire" are symmetrical, such that a supply of goods must occur as part of a bilateral transaction or dealing under which the other party acquires them. These remarks are apposite when one is speaking of a market for goods. They also hold true when one is speaking of services and a market for services. It is appropriate and correct to speak, in the present case, of the suite of services being "supplied or acquired" in a market that is circumscribed as the port of origin, even though part of the services are to be physically provided or delivered at a port of destination, such as Sydney. It is also appropriate and correct to speak of "competition in a market" whose geographic area is a port of origin, even though the competition may be played out in the manner in which an aspect of the services is physically provided or delivered at a port of destination. This is because the sum of the activities engaged in by competitors in promoting the sale and supply of the goods or services with which the market is concerned, wherever those activities take place, is properly aligned to "the market" itself, in which the opportunities for substitution and switching are to be found. In the present case, when dealing with the suite of services, those opportunities are not found at the port of destination; on the findings of fact made by the primary Judge, they are only found at the port of origin where the suite, as such, is supplied and acquired, even though, having been supplied and acquired - which, on any view, can be no later than the point of departure of the aircraft from the port of origin - elements of the acquired suite are physically provided or delivered at a distant place.
654 The same framework applies when considering barriers to entry - yet another metaphor in the world of abstractions. Accepting that facilities, permits and authorisations (and the like) are required so that the suite can be supplied, and may constitute barriers to entry, it does not follow that the geographical dimension of the market is defined by the location of the facilities and/or the sources from which the permits and authorisations are acquired. In other words, if, for example, the requirement for cargo handling facilities at a port of destination is a barrier to entry, this does not mean that the geographical dimension of the market then becomes a place or includes one of a number of places where the barrier is, metaphorically speaking, perceived to have been erected. It is important to appreciate that, in order to participate in the market, suppliers have already surmounted the barrier (they have, for example, acquired the facilities) and stand in the market (here, at the port of origin) with their competitors to supply the suite of services to buyers who seek to buy that suite. To take the primary Judge's example of the tour bus in Europe (see [607] and [641] above), various and different licences may be necessary to operate the tour bus on its journey through Europe. The need to obtain these licences may be a barrier to entry. But the market in Australia for packaged tours in Europe does not then become also a market in each of those other places.
655 I return to consider the Commission's submissions I have summarised at [612]-[622].
656 With respect to the Commission's first submission (see [617] above), I do not read the primary Judge's reasons for judgment as seeking to establish some new "test", still less a "test" not supported by authority. In my respectful view, the primary Judge focused, correctly, on the geographic area in which the market product - the suite of services - was bought and sold, and over which switching and substitution by and between buyers and sellers occurred. Further, I do not think that the Commission's submissions accurately represent the primary Judge's reasoning by arguing that his Honour determined the locus of the market by reference to a single aspect of the transaction between buyers and sellers, namely the place where possession of the cargo for transport is taken. The primary Judge plainly placed some significance on the fact that the service of taking possession at the port of origin, with a view to flying the cargo to the port of destination, cannot be performed anywhere but at the port of origin. But his Honour's reference to that consideration was in connection with the airlines' broader submission that the geographical dimension of the market is the area in which sellers of the product operate and to which buyers can practicably turn for such goods or services. The primary Judge's reference to the place where possession of the cargo is taken was part of his Honour's acceptance that the range of airlines which are available to be selected in any route specific market is limited by the need to have a physical presence in the port of origin for that purpose.
657 With respect to the Commission's second submission (see [618] above), it may be accepted that, in the markets in question, there is no demand "for the mere commencement of a transport service at the port of origin". But the primary Judge's reasoning did not proceed on the flawed basis that there was such demand.
658 With respect to the Commission's third submission (see [619] above), the primary Judge's reasoning did not involve the inconsistency alleged. Plainly, the competing airlines supplying the suite of services had to supply services at the port of destination. This fact was not lost on the primary Judge who accepted and recognised that the product was a suite of services. Thus, the airlines supplying the services at the port of destination (which were part of the suite) were necessarily the same airlines supplying the services at the port of origin (which were also part of the suite). The Commission's submission raises a false dichotomy.
659 With respect to the Commission's fourth submission (see [620] above), the primary Judge's conclusion on the geographical dimension of the market is not inconsistent with the general observations made in Fortescue Metals Group about markets, which are quoted by the Commission. Fortescue Metals Group does not stand for the proposition that the geographical dimension of the market is determined merely by the fact that the product dimension of the market is characterised, in part, by a geographical element. In that case, the Tribunal did refer (at [1022]) to actual sales patterns, the location of customers and the place where sales take place, as relevant considerations for determining the geographic market. The Tribunal's reference to the location of customers must be considered having regard to the circumstances of the particular case. In the present case, it is to be remembered that, on the primary Judge's findings of fact, customers (shippers) generally acted through intermediaries in Hong Kong and the other ports of origin. Thus, for all relevantly practical purposes, the customers (shippers) were, through their intermediaries, located at the port of origin in the markets in question.
660 With respect to the Commission's fifth submission (see [621] above), there is nothing to suggest that the primary Judge's consideration of the geographic dimension of the market was informed only by short-run considerations.
661 It is necessary for me to say something more about the shippers. Air NZ and Garuda each contended that the primary Judge erred in failing to find that the relevant markets were limited to transactions between airlines and freight forwarders, and in finding that some large importers in Australia were participants in the relevant markets.
662 The gravamen of Air NZ's submission was that the primary Judge's finding that shippers were participants in the relevant markets was based on theoretical considerations and not on evidence. Air NZ pointed, in particular, to the primary Judge's finding (at [309(d)]) that shippers with significant cargo volume, wherever located, were 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 exercise of market power at the relevant origin airport.
663 Air NZ also pointed to what it said was an absence of evidence that large shippers imposed a real constraint on airlines operating out of Hong Kong and Singapore. Moreover, it submitted that, to the extent that there was evidence on the question, it tended to demonstrate that large shippers did not exert a substantial constraining effect on airlines.
664 Garuda's submissions were to the same effect. It submitted that the primary Judge's finding that a number of shippers controlled significant volumes of cargo (to the extent that the finding was made in respect of cargo on routes from Hong Kong or Indonesia to ports in Australia) was not open on the evidence. Garuda submitted that there was no evidence of the existence of such shippers and that the primary Judge made no finding or reference to evidence that a shipper actually made a switching decision. Thus, Garuda submitted, it was not open to the primary Judge to find that shippers (of the kind the primary Judge had in mind) made decisions, including decisions in Australia, about which airlines they would use. Further, the economic significance of any such decision could not be assessed. Garuda submitted that the primary Judge should have held that the Commission had not established that there were any switching decisions of significance to market definition that had been made in Australia in respect of any of the routes in issue.
665 Garuda's submissions referred to evidence concerning a particular dealing between itself and an importer from the Sunshine Coast who wished to ship watches from Jakarta to Brisbane. Garuda submitted that this evidence showed that freight forwarders and shippers were not in the same market. Garuda also referred to a number of general findings made by the primary Judge concerning the role and function of freight forwarders. The thrust of these submissions was that the services supplied by airlines to freight forwarders, and the services supplied by freight forwarders to consignors or consignees, were in different markets having regard to differences in the scope of services provided and in their pricing.
666 The primary Judge's finding was that 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. The cases recognise that the behaviour of participants at one functional level may have a constraining effect on the behaviour of participants at another functional level that is sufficient to warrant the inclusion of all those activities in the market the subject of attention: see, for example, the discussion in Metcash at [252]-[268]. However, I do not think that the primary Judge's conception of the relevant markets in the present case was one in which two distinct functional levels - one between airlines and freight forwarders and the other between freight forwarders and shippers - should be combined (or "collapsed") into the one market. After depicting the "vertical structure of the industry" with the progression Consignor Origin freight forwarder Airline Destination freight forwarder Consignee, the primary Judge said (at [269]):
This arrangement may not be so vertical as it appears. It is true that goods travel from the top to the bottom but often enough the consignee (as importer) may be the instigator of the particular shipment which will create the situation of importation. The services provided by the freight forwarders are necessary accompaniments to the services provided by the airlines themselves and neither makes much sense without the other.
667 Later, the primary Judge described (at [299]) the freight forwarders as intermediaries having fluctuating control over the cargo whose carriage they arranged. His Honour saw this characteristic as reflecting "economic substance". The primary Judge also reasoned that the shippers were "significant economic actors". His Honour said that he could not imagine a universe of discourse in which a rational business would ignore the role of shippers.
668 It seems to me that, after considering a large body of evidence directed to the relationship between airlines, freight forwarders and shippers, it was open to the primary Judge to define the market in a way which included all three as participants, without seeking to draw the "bright line" functional levels which the airlines urge.
669 It also seems to me that it was open to the primary Judge to draw broad conclusions about the markets in question based on how, generally, airlines, freight forwarders and shippers interact in relation to the supply and acquisition of air cargo services. The primary Judge noted, on a number of occasions in the reasons for judgment, the airlines' submissions that the Commission had not established with respect to each relevant market, that shippers played the role that the primary Judge found them to play in other corresponding air cargo markets. But, having been alive to this issue, the primary Judge rejected the airlines' submissions. He considered the airlines' view as contrary to common sense. The primary Judge also reasoned (at [307]):
… There is no reason to think that the structural features of the air cargo business on different routes are different. In particular, there is no reason to think that the airlines on significant routes are not involved in the giving effect to international trade nor that international trade involves importers and exporters.
670 Air NZ and Garuda each referred to aspects of the evidence which, they argued, showed that the primary Judge erred in concluding that shippers were participants in the relevant markets. While an appellate court can draw its own inferences from primary facts, the cautionary observations of Davies J in Australian Meat Holdings Pty Limited v Trace Practices Commission [1989] ATPR 40-932; [1989] FCA 25 should be borne in mind. There, his Honour referred to an appeal court being taken to "snatches of the evidence only". His Honour observed (at 50,091):
The Court must be careful not to draw inferences from evidence which is disputed by other evidence and which forms only part of the material before the Court.
671 That observation is applicable to the present appeals. The primary Judge's discussion and findings on the functional dimension of the relevant markets at [266]-[309] of the reasons show that his Honour gave careful consideration to a very large body of evidence concerning the role of shippers. His Honour was able to conclude that across the Asia Pacific area the airlines recognised that shippers had demand for capacity and that airlines actively followed the position of shippers, recognising that they were "the economic foundation of the market".
672 I am not persuaded that the broad conclusions reached by the primary Judge about the participation of shippers in the relevant markets (see the quotation at [605] above) were not open to be made in light of the evidence to which his Honour specifically referred.
673 But even if the primary Judge was in error in this respect, the error is without significance to the central issue in the appeals. The appeals are not about the functional dimension of the relevant markets, but their geographical dimension. Air NZ expressly conceded that the primary Judge's findings as to the functional dimension of the relevant markets do not affect the result in the case. This concession was properly made. The airlines' respective cases in support of the primary Judge's conclusion that each market was not a market in Australia did not rely on the absence of shippers as relevant market participants. Equally, the Commission's case on appeal did not rely on the inclusion of shippers as market participants. As I have noted on a number of occasions, the Commission's case on appeal rested on the twin propositions that services were physically provided or delivered at the port of destination in Australia and that it was at these locations, amongst others, that the airlines were able to express or manifest rivalrous behaviour.
674 In making this observation, I do not wish to be taken as indicating that, under other circumstances, the finding that shippers were market participants would be of no significance. Once again, the case under appeal is one involving specific findings of fact. Here, the primary Judge was at pains to stress that, notwithstanding the participation of shippers, the market was not as "vertical" as might otherwise be suggested by the fact of that participation. This, no doubt, was because the shippers acted through freight forwarders who were intermediaries. I refer, again, to the primary Judge's finding that airlines carrying cargo from a port of origin, such as Hong Kong (and, by implication, the other ports of origin relevant to the present question) generally dealt directly only with freight forwarders situated in Hong Kong (or the other ports of origin) or in nearby environs.
675 In support of the primary Judge's finding that shippers were participants in the relevant markets, the Commission, in the course of oral submissions in reply, referred to some evidence (specifically, the evidence of Glenna Frances Cluff who was employed by Toshiba (Australia) Pty Limited as a Logistics Import Specialist) to support the proposition that a switching decision could be made in Australia. With respect to this evidence, it is important to understand that the question is not so much where the person who makes a switching decision is located, but where and how a switching decision, once made, is implemented. Here, the effect of the primary Judge's findings was that the relevant switching decisions were implemented by intermediaries at the port of origin, not at the port of destination.
676 The Commission also referred to some evidence given by Dennis Ian Nelson who had worked for BAX Global in the "freight forwarding industry" in Australia. This evidence was to the effect that, for major inbound routes (for example, from the USA to Sydney), Mr Nelson and his staff could negotiate a price and service standards for freight forwarding services with local consignees in Australia without reference to BAX Global at the port of origin. This evidence was given at a high level of generality. It was not directed to the markets in question. I do not accept that snippets of evidence of this kind seriously call into question the correctness of the primary Judge's overall evaluation of how, on the evidence before him, the markets in question functioned.
677 I should also briefly refer to the decision of the High Court of New Zealand in Commerce Commission v Air New Zealand Ltd (2011) 9 NZBLC 103,318; [2011] NZHC 1285 (Commerce Commission v Air New Zealand). In that case, which concerned price-fixing in relation to inbound and outbound air cargo services to New Zealand, it was found that there was a "market in New Zealand". Section 3(1A) of the Commerce Act 1986 (NZ) defines a "market" as a market in New Zealand for goods or services as well as other goods or services that, as a matter of fact and commercial common sense, are substitutable for them. The hearing in that case proceeded on the basis of agreed facts. It is clear that the court's finding that there was a market in New Zealand was informed by its findings on the facts before it concerning the influence of downstream market effects. In the present case, the primary Judge observed (at [334]) that, in Commerce Commission v Air New Zealand, the court was willing to infer a number of matters concerning downstream effects which he could not embrace having had the benefit of a trial. Thus, the present case is immediately distinguishable on the facts.
678 The finding on market definition in Commerce Commission v Air New Zealand was also informed by the approach of the New Zealand Court of Appeal in Port Nelson Ltd v Commerce Commission [1996] 3 NZLR 554 in which it was said (at 560):
Generally a market will be identified 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.
679 In Commerce Commission v Air New Zealand, the court expressed its preference (at [189]) for this general approach to market definition rather than "prescriptive definitions limiting a market to the outer geographic boundary of the immediately substitutable products". This general approach places less emphasis on substitutability as the cornerstone of market definition than is accorded by s 4E of the TPA as interpreted by the Australian case law.
680 Finally, I should record my view that, on existing authority, there is no warrant for incorporating in the process of market definition an effects-based doctrine that would extend the geographic reach of a market to Australia on the basis that persons in Australia are or might be adversely affected by conduct outside Australia that is sought to be impugned.
681 On the facts found by the primary Judge, none of the relevant markets was a market in Australia. In my respectful view, the primary Judge did not err in the conclusion to which he came, based on those facts. It follows that each appeal should be dismissed.