FEDERAL COURT OF AUSTRALIA

Australian Competition and Consumer Commission v Australia and New Zealand Banking Group Limited [2015] FCAFC 103

THE COURT ORDERS THAT:

1.    The appeal be dismissed.

2.    The cross-appeal be allowed.

3.    The appellant pay the respondent’s costs of the appeal and cross-appeal.

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

REASONS FOR JUDGMENT

THE COURT

1    This appeal concerns the proper characterisation, for competition law purposes, of the activities of and interaction between participants in the market for mortgage loans. Loan providers, such as banks, distribute or “sell” mortgage loans to borrowers through both internal and external distribution channels. The internal or “in-house” channels include, most significantly, bank branches. The external or independent channels relevantly include mortgage brokers. Independent mortgage brokers provide advice and assistance to their clients about the loan products potentially available to them from a range of loan providers. Brokers receive commission from a bank if a borrower successfully applies for a loan from the bank through them. In the course of their interaction with prospective borrowers, bank branch employees also provide a level of advice and assistance, albeit only in relation to their bank’s own loan products. When providing this advice and assistance through its branches, is the bank competing with mortgage brokers in a market for the supply of loan arrangement services? Or are the advice and assistance activities of bank branches in this respect properly characterised as merely activities ancillary to the distribution of loan products in the market for the supply of loan products?

2    The Australian Competition and Consumer Commission (ACCC) commenced proceedings against Australia and New Zealand Banking Group Limited (ANZ) alleging that it contravened s 45(2)(a)(ii) and s 45(2)(b)(ii) of the Trade Practices Act 1974 (Cth) (the Act) as in force at the relevant time. The ACCC alleged that, in April 2004, ANZ entered into an agreement with a particular mortgage broker, Mortgage Refunds Pty Ltd (Mortgage Refunds), which contained a provision which had the purpose or effect of substantially lessening competition in the market for the supply of loan arrangement services to members of the public by loan providers, franchisees and brokers. The impugned agreement concerned the maximum amount of money that Mortgage Refunds could offer to refund to its customers when its customers successfully applied for an ANZ mortgage loan through Mortgage Refunds. It was alleged that the agreement had the purpose, effect or likely effect of fixing, controlling or maintaining a discount, allowance, rebate or credit in relation to loan arrangement services provided by Mortgage Refunds in competition with ANZ. The ACCC relied on s 45A of the Act, which deemed the agreement, if proved, to have the purpose or effect of substantially lessening competition.

3    Critical to the ACCC’s case was the allegation that ANZ and Mortgage Refunds competed against each other, and other banks and brokers, in the market for the supply of loan arrangement services.

4    The ACCC was unsuccessful. The primary judge found that ANZ, through its branches and other internal channels, did not participate in any market in which brokers, such as those engaged by Mortgage Refunds, provided loan arrangement services to potential borrowers. ANZ was not, in any relevant sense, in competition with Mortgage Refunds. Therefore, s 45A of the Act was not engaged.

5    The ACCC appeals from the judgment of the primary judge. It raises a number of appeal grounds, though most of them are ultimately directed, in one way or another, at his Honour’s rejection of the ACCC’s case that ANZ and Mortgage Refunds relevantly competed in a market for the supply of loan arrangement services. The ACCC challenges that finding primarily on the basis that, in its submission, the primary judge made findings of fact that were contrary to, or against the weight of, the evidence, or failed to make findings that were supported by the evidence.

6    ANZ cross-appeals against one finding made by the primary judge. That finding was that the refunds paid by Mortgage Refunds to borrowers that were the subject of the agreement with ANZ were “rebates” within the meaning of s 45A of the Act.

Uncontroversial facts

7    Most of the primary facts are not in dispute. The factual controversies at trial and on appeal relate mainly to the inferences or conclusions that can or should be drawn from the evidence, in particular statements made in a number of internal ANZ documents tendered by the ACCC and the evidence of so-called “market participants”. The inferences able to be drawn from that evidence were also critical to the probative value of the expert opinion evidence led by the ACCC from an economist in relation to the nature of the relevant market or markets.

8    ANZ’s banking business includes the provision of finance to borrowers for the acquisition of residential properties. These loans are made or structured in various ways, though repayment is generally secured by a mortgage over the property purchased. At the relevant time, ANZ had a division or department called ANZ Mortgage Group which was responsible for making such loans. That division or department was responsible for, amongst other things, designing particular “loan products”, identifying lending criteria and, importantly, receiving and approving (or declining) loan applications.

9    At the relevant time, there were a number of different ways that ANZ interacted with prospective borrowers so as to offer its loan products and receive loan applications. To use the industry jargon, it had a number of different “distribution channels” in relation to its loan products, including mortgage loans. Some of the distribution channels were internal or “in-house”. The main internal distribution channel was ANZ’s network of bank branches, where bank officers were available to provide advice and assistance in relation to loan applications from prospective borrowers.

10    ANZ also appears to have had other internal channels, either in existence or in contemplation at the time of the relevant impugned conduct, though they ultimately appear to have only been marginally relevant to the issues at trial. They included specialist loan officers or sales persons called Personal Mortgage Managers and a call centre called Mortgage Direct. There were references in the evidence to a proposal to appoint franchisees of a “mobile lending” business called “Mortgage Solutions”, though this had not occurred at the time of the impugned conduct. These reasons will primarily refer to the ANZ branches as the main in-house distribution channel.

11    As one would expect, all the internal or “in-house” channels only offered and provided advice and assistance in relation to ANZ loan products. The ANZ branches and other “in-house” channels were managed by a separate division or business unit of ANZ called Personal Banking Business.

12    The external distribution channels employed by ANZ were primarily independent finance or mortgage brokers. These brokers offered clients or customers advice and assistance in relation to an often wide range of loan products from a wide range of loan providers. They were not in any respect tied to ANZ or ANZ loan products. Brokers were trained and accredited by ANZ and other loan providers to submit loan applications on behalf of their customers. The brokers derived their income from commissions paid by loan providers, including ANZ, when loan applications by their customers were approved. They were generally not separately remunerated by their customers in respect of any advice or assistance provided to them. They were not rewarded at all unless a loan application submitted by the broker on behalf of a customer was approved by the loan provider. The amount of commission was generally calculated as a percentage of the amount of the approved loan. It was not reflective of the amount of advice or assistance that preceded the submission of the loan application.

13    The nature of the advice and assistance provided by both the internal ANZ channels and the external brokers is addressed in more detail later in these reasons. Suffice it to say at this stage that a critical issue is whether the services provided to prospective borrowers by the internal and external channels were substantially similar such that there was, or was likely to be, close substitution between them.

14    On 7 July 2001, ANZ entered into an agreement, called the “originator agreement”, with a company named Australian Financial Group Ltd (AFG). AFG carried on an independent finance broking business in which it acted for clients in arranging mortgage loans and other financial transactions. Under the originator agreement, ANZ appointed AFG to market certain ANZ loan products as an independent contractor. AFG was not to represent that it was acting in any capacity other than as an independent contractor and had no authority to bind, or purport to bind, ANZ without ANZ’s prior consent in writing.

15    AFG’s obligations under the originator agreement included ensuring that applications from its clients to ANZ were in an approved form and accurately recorded the customer’s instructions. AFG was obliged to exercise care and skill in marketing ANZ’s loan products.

16    Importantly, a clause in the originator agreement prohibited AFG from inducing, or causing anyone to induce, any person to apply for any loan products by “offering gifts or prizes or other inducements of whatever kind, without the prior written approval of [ANZ]”.

17    Whilst AFG was able, under the originator agreement, to market ANZ loan products and submit loan applications to ANZ on behalf of its customers, the terms of the originator agreement made it clear that AFG was not prevented from engaging in any other business or undertaking, or transacting any business with any other bank, insurance company or other financial institution.

18    The originator agreement provided that only “authorised officers” of AFG who had been approved by ANZ were able from time to time to act on behalf of AFG in performing AFG’s obligations under the agreement. The authorised officers, being employees, contractors or agents of AFG, were named in a schedule to the agreement. The schedule was able to be amended from time to time by agreement between ANZ and AFG.

19    In March 2004, AFG entered into an agreement with Mortgage Refunds. Under this agreement, persons nominated by Mortgage Refunds were to be accredited for the purpose of receiving and processing loan applications to ANZ Mortgage Group for its loan products. The primary judge inferred that these nominated persons were to be authorised officers of AFG for the purposes of the originator agreement with ANZ. It was these individual brokers who dealt directly with, and acted as intermediaries between, prospective borrowers and ANZ in relation to the submission of loan applications to ANZ.

20    AFG, Mortgage Refunds and the individual brokers all ultimately derived their income, directly or indirectly, from commission paid by ANZ and other loan providers when loan applications submitted by customers through them were approved. Insofar as ANZ was concerned, where a successful loan application was lodged through a broker associated with Mortgage Refunds, the general method of paying commission was that it first paid the commission in respect of the successful loan application to AFG. AFG then paid a share of that commission to Mortgage Refunds and Mortgage Refunds paid a share to the individual broker who had assisted the customer to lodge the successful loan application.

21    Mortgage Refunds attracted customers by offering a deal whereby it agreed to refund to the customer part of the commission that was otherwise payable by the loan provider to Mortgage Refunds’ broker in respect of successful loan applications lodged through its brokers. Whilst the individual brokers associated with Mortgage Refunds were the main point of contact between the borrowers and ANZ, the agreement to refund part of the commission if the loan application was successful was between Mortgage Refunds and the borrower. When Mortgage Refunds received its share of the ANZ commission from AFG, it would deduct the refund amount from it and pay it to the customer before paying the individual broker, who dealt with the customer, his or her share of the commission.

22    It would appear that in early 2004, ANZ became aware of, and took exception to, the “mortgage refund” offer. On 22 March 2004, ANZ wrote to AFG advising AFG that it had become aware that a business accredited under AFG was promoting a “mortgage refund” offer involving the rebate to the customer of part or all of the commission paid by ANZ. The letter identified a number of individuals accredited under AFG who were involved in this business. These individuals were all mortgage brokers associated with Mortgage Refunds. The letter referred to the clause in the originator agreement between ANZ and AFG which provided that an approved originator must not induce any party to apply for any loan products by offering gifts, prizes or other inducements without the prior written approval of ANZ. The letter pointed out that no written approval had been provided by ANZ to the named individuals to offer such a refund or rebate and that, as a result, ANZ was cancelling their accreditation. The result was that the individual brokers were no longer eligible to submit applications to ANZ.

23    Following this letter, discussions ensued between representatives of ANZ, AFG and Mortgage Refunds. Following these discussions, ANZ wrote to the directors of Mortgage Refunds. The letter, which is dated 29 April 2004, should be set out in full because, on the ACCC’s case, it recorded or evidenced the impugned agreement which had the purpose, effect or likely effect of fixing, controlling or maintaining a discount, allowance, rebate or credit.

Re Mortgage Refunds Pty Ltd

I refer to our recent discussions regarding accreditation with ANZ of third party introducers who operate under your company.

Under the terms of the agreement held between ANZ and the Australian Finance Group, an Approved Originator must not “induce or cause to induce any party to apply for Loan Products by offering gifts or prizes or other inducements of whatever kind, without the prior written approval of the Bank”.

We consider that the current business model that you advertise and operate to be in breach of this agreement and we have not given prior written approval. On this basis accreditation was cancelled.

After our discussion of this matter with yourself and representatives from AFG, we hereby give specific agreement to the following in regards to ANZ Loan Products:

•    The maximum refund that can be provided to the customer in relation to an ANZ Loan Product is to be no greater than the amount of the Loan Approval Fee as determined by the ANZ Bank. The amount of this fee may be altered at anytime and at the Bank’s sole discretion.

•    You may not advertise this refund in any form without the specific written agreement of the ANZ Bank, this is to include use of the ANZ logo in any advertisement. At this time agreement is not given.

•    ANZ retains the right to withdraw or amend this agreement, at it’s sole discretion, at anytime.

We trust that the above accurately reflects our discussions and would ask that the enclosed duplicate of this letter be signed by the relative officers of your company as acceptance of the terms above.

Upon receipt of the signed acceptance we will reactivate the accreditation of your loan writers.

Should you have any queries regarding this matter please do not hesitate to contact me.

24    ANZ’s proposal in this letter was accepted by Mortgage Refunds and as a result the brokers named in the earlier letter, all of whom were associated with Mortgage Refunds, were re-accredited.

25    Accepting, for present purposes, that this letter evidenced an agreement which fixed the level of discount that Mortgage Refunds could pay to a person who successfully applied for an ANZ loan through it, three important questions relevant to the issues raised by this appeal arise.

26    First, can the discount properly be characterised as a “rebate” for the purposes of s 45A of the Act? The primary judge found that it could. This finding is the subject of ANZ’s cross-appeal.

27    Second, if the refund was a rebate for the purposes of s 45A of the Act, what precisely were the “goods or services” to which this rebate related? Specifically, did the rebate relate to the supply of loan products, or did it relate to the supply of services comprising advice and assistance to prospective borrowers? The significance of this question is to the third and more fundamental question.

28    The third question is, did Mortgage Refunds and ANZ compete in a market for the supply of the goods or services to which the rebate related? If the rebate related to the supply of services by Mortgage Refunds, did Mortgage Refunds compete with ANZ in a market for the supply of such services?

29    As will be considered in more detail later in these reasons, the ACCC characterised the services the subject of the agreement as a suite of services, described as “loan arrangement services,” that were provided by loan providers (and their agents and franchisees) and independent mortgage brokers to persons seeking to acquire loan products. The ACCC claimed that ANZ and Mortgage Refunds competed with each other and other loan providers and brokers in the market for the supply of such loan arrangement services.

30    As will also be considered in more detail later in these reasons, the primary judge rejected the ACCC’s characterisation of the relevant market and rejected the contention that ANZ and Mortgage Refunds competed in any market for loan arrangement services. The essential question that needs to be addressed in this appeal is whether, having regard to the evidence, the primary judge erred in rejecting the ACCC’s case in this regard.

Relevant Statutory Provisions

31    At all times relevant to this matter, ss 45(2) and (3) of the Act, as then in force, provided as follows:

(2)    A corporation shall not:

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

(i)    …

(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.

(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.

32    Section 45A(1) of the Act, contained a deeming provision for the purposes of s 45. It provided as follows:

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.

33    As clearly spelt out in s 45(3), “competition” for the purposes of ss 45 and 45A, means competition “in any market”. Section 4E of the Act provides a limited inclusive definition of “market” in the following terms:

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.

ACCC’s case at Trial

34    To properly consider both the primary judge’s findings and the ACCC’s challenge to them, it is necessary to have close regard to the ACCC’s case as pleaded and to the evidence it led at trial.

The ACCC’s pleading

35    In its pleading (the Further Amended Statement of Claim or FASC), the ACCC alleged (at paragraph 6) that at all material times there had been a demand from persons throughout Australia seeking to acquire loan products (“customers”) for services to assist them in choosing and acquiring loan products (“loan arrangement services”). Paragraph 7 of the FASC contains critical particulars of the so-called “loan arrangement services”. It is in the following terms:

7.    Loan arrangement services:

7.1.    are supplied in relation to either the loan products of a single loan provider, or the loan products of more than one loan provider;

7.2.    at all material times comprised:

7.2.1.    advice as to the respective features of available loan products;

7.2.2.    advice as to which loan products were available to persons in the customer’s circumstances;

7.2.3.    advice as to which loan products best suit the customer’s needs;

7.2.4.    assistance to complete and lodge applications in a manner that meets the requirements of a loan provider;

7.2.5.    facilitation or liaison in the transaction for the acquisition of a loan product between the customer and those sections or divisions of a loan provider responsible for providing loan products; and

7.2.6.    submission of an application for a loan product for a customer to a loan provider, or the relevant section or division of the loan provider.

36    Paragraph 7.1 is important. It is not in dispute that one difference between the services that were supplied to prospective borrowers by loan providers, such as ANZ, and the services provided by brokers, such as the brokers associated with Mortgage Refunds, is that, as one would expect, ANZ only advised and assisted customers in relation to ANZ loan products. Brokers, on the other hand, provided advice and assistance, at least in the first instance, in relation to the loan products supplied by many different loan providers. A critical issue at trial, and on appeal, is the importance and significance of this difference.

37    The ACCC’s pleading acknowledged this difference. Paragraphs 9 and 10 of the FASC stated that when loan providers and their agents and franchisees supply loan arrangement services, they do so only in relation to their own products or those of associated companies. Paragraph 13 of the FASC stated that when brokers supply loan arrangement services, they generally do so in relation to all loan products of those loan providers for which they are accredited.

38    As has been said, the significance of this difference was an important issue at trial. It is an important issue on appeal. It goes to the question whether any services provided by ANZ branches to prospective borrowers, its customers, were close substitutes for the services supplied by independent brokers. That, in turn, is relevant to determining whether the respective services were provided in the same or different markets, and whether ANZ and Mortgage Refunds competed in any market for such services. The ACCC’s case was that, despite this difference, the loan arrangement services supplied by loan providers were close substitutes for the services supplied by brokers. They were accordingly supplied in the same market.

39    The ACCC’s pleaded case was that ANZ supplied loan arrangement services to customers “through ANZ branches and franchises in Australia” (paragraph 17.2 of the FASC). The loan arrangement services supplied by ANZ included services of the sort pleaded in paragraph 7.2 of the FASC, though again it was acknowledged in the pleading that the services related only to ANZ loan products (paragraph 17.3 of the FASC).

40    Likewise, the ACCC alleged that Mortgage Refunds itself, or its employees or agents, supplied loan arrangement services of the sort pleaded in paragraph 7.2 of the FASC (paragraph 18.3 of the FASC). The services supplied by Mortgage Refunds or its “engaged persons” were in relation to loan products generally, not just in relation to ANZ loan products (paragraph 18.3 of the FASC).

41    It was, however, only when Mortgage Refunds offered loan arrangement services “in respect of ANZ loan products” that the ACCC alleged that the services offered by Mortgage Refunds were direct substitutes for the services offered by ANZ (paragraph 19.4 of the FASC).

42    Despite this, the ACCC alleged (in paragraph 20 of the FASC) that ANZ was in competition with Mortgage Refunds in relation to the supply of loan arrangement services generally. It would follow that it was the ACCC’s case that ANZ competed with Mortgage Refunds even when Mortgage Refunds was offering loan arrangement services in respect of loan products offered by other banks, such as Westpac or the Commonwealth Bank. It would also appear to follow from the ACCC’s case, as pleaded, that ANZ competed with other suppliers of loan arrangement services, such as other loan providers (see 8.1 of the FASC) even where the other loan providers did not provide any such services in respect of ANZ loan products.

43    Thus, on the ACCC’s case as pleaded, ANZ competed with, for example, Westpac and the Commonwealth Bank, in relation to the supply of loan arrangement services, even though the services provided by Westpac and the Commonwealth Bank related only to their respective loan products, and ANZ’s loan arrangement services were only in respect of ANZ loan products.

44    Perhaps recognising the possible incongruity or implausibility of this proposition or allegation, the ACCC’s pleading also alleged (additionally or in the alternative) that ANZ only competed with Mortgage Refunds in relation to the supply of loan arrangement services in respect of ANZ loan products (paragraph 21 of the FASC). It was not, however, pleaded, or at least not in clear terms, that there was a separate market or sub-market for the supply of loan arrangement services in respect of ANZ loan products. The only market pleaded by the ACCC was “an Australia-wide market for the supply of loan arrangement services to members of the public by loan providers, franchisees and brokers” (paragraph 15 of the FASC).

45    The only other point to note about the ACCC’s pleading concerns the agreement that was alleged to fall within the terms of s 45A of the Act. As already indicated, the ACCC’s case was that the agreement between ANZ and Mortgage Refunds was contained in, or evidenced by, the 29 April 2004 letter from ANZ to Mortgage Refunds. The ACCC’s case, as pleaded, was that the term of the agreement between ANZ and Mortgage Refunds that fixed the maximum refund allowed to be offered by Mortgage Refunds to its customers had the purpose, effect or likely effect of fixing, controlling or maintaining a discount, allowance, rebate or credit “in relation to loan arrangement services” supplied by Mortgage Refunds in competition with ANZ (paragraph 30.1). Therefore, the alleged rebate (or discount, allowance or credit) was tied to the supply of a service supplied by Mortgage Refunds to its customer. It was not alleged that the refund was “in relation to” the supply of the ANZ loan product to the customer.

Evidence relied on by the ACCC at trial

46    The focus here is on the evidence relied on by the ACCC to make good its allegations in relation to the relevant market.

47    The evidence relied on by the ACCC essentially fell into five categories.

48    First, the ACCC relied on some answers ANZ gave to the ACCC’s interrogatories.

49    Second, reliance was placed on the evidence of a number of so-called “industry witnesses”. These were people who had worked both at banks and as, or with, mortgage brokers. They included Mr Jason King, the principal of Mortgage Refunds, Mr Leon Stark, a broker associated with Mortgage Refunds, Ms Manuela Zacka, a business development manager with AFG, Mr Paul Lahiff, the managing director of Mortgage Choice Limited (Mortgage Choice), a leading mortgage broker group, and Mr Damian Percy, an officer of Bendigo and Adelaide Bank Ltd who had worked in that bank’s mortgage business.

50    Third, the ACCC relied on evidence called from two so-called “market participants” or customers. Both of these individuals, Mr Roger O’Malia and Mr John Black, had been customers or clients of Mortgage Refunds, as well as customers of banks.

51    Fourth, the ACCC relied on the contents of a number of internal ANZ documents.

52    Fifth, the ACCC relied on the opinion evidence of an experienced and highly qualified economist, Dr Vincent FitzGerald.

53    The judgment of the primary judge contains a detailed summary and analysis of the evidence. There is ultimately no utility in including an equally detailed summary here. Attention will, instead, be given to those parts of the evidence that are directly relevant to the ACCC’s grounds of appeal, or were emphasised by the ACCC in its oral and written submissions on appeal.

Answers to interrogatories

54    Paragraphs 7.2 and 17.3 of the ACCC’s pleading specified six types of advice and assistance that comprised loan arrangement services generally, and loan arrangement services provided by ANZ specifically. In paragraph 11 of ANZ’s verified statement in answer to the ACCC’s interrogatories, the following questions and answers appear:

11    Did ANZ, in the period November 2002 to May 2004, in responding to enquiries from persons seeking to apply for an ANZ home loan product, undertake any of the following activities and if so, which:

11.1    advising enquirers as to the features of the respective loan products available from ANZ;

Yes.

11.2    advising enquirers which of ANZ’s loan products are available to persons in the enquirer’s circumstances;

Yes.

11.3    advising the enquirer which of ANZ’s loan products are best suited to the enquirer’s needs.

Yes.

11.4    assisting the enquirer to complete and lodge applications for the loan product selected in a matter that meets the requirements of the Mortgage Division.

Yes.

11.5    acting as liaison or otherwise facilitating the transaction for the acquisition of the loan product between the enquirer and the Mortgage Division, and

No. Mortgages is a business unit of ANZ.

11.6    submitting, or arranging for the submission, of the enquirer’s application for the loan product to the Mortgage Division.

No. Mortgages is a business unit of ANZ.

55    The reference to “Mortgages” in answers 11.5 and 11.6 is a reference to ANZ Mortgage Group, the business unit of ANZ which was responsible for administering the lending of money by ANZ to its customers secured by residential properties (see interrogatory 2). The significance of the existence of this separate business unit is considered later in these reasons. ANZ’s negative answers to questions 11.5 and 11.6 were perhaps a product of the way the questions were framed. It is difficult to see how ANZ could act as a liaison between a customer and ANZ’s Mortgage Division. It could, however, perhaps be said that branch officers (or Personal Banking) acted as a liaison.

56    In any event, the ACCC submitted that these answers to the interrogatories provided evidence that ANZ supplied four out of the six services said to constitute “loan arrangement services” as defined in its pleading.

The industry witnesses

57    The evidence of the so-called industry witnesses was relied on by the ACCC in three ways. First, the evidence of Mr King and Ms Zacka about their discussion with an employee of ANZ about ANZ’s objections to Mortgage Refunds’ payment of refunds was said to demonstrate relevant competition or rivalry between ANZ and Mortgage Refunds. Second, their evidence was said to demonstrate that loan providers and brokers competed with each other in respect of loan arrangement services. Third, the evidence was said to show that the services provided by bank branches were effectively the same, and therefore were close substitutes for, the services supplied by brokers.

58    In relation to the evidence concerning the relevant discussions with ANZ, the evidence of Mr King, who was the principal of Mortgage Refunds, was that, on 30 March 2004, he and a senior ANZ bank manager, Mr Tim Carroll, met and discussed ANZ’s objection to Mortgage Refunds providing refunds to its customers. Mr King recalled Mr Carroll said words to the following effect (at [58] of his affidavit):

We have a problem with your refund model. I’m not sure what we can do about it.

Wayne Ormond [of Refund Home Loans] has rocked the boat. He has been giving refunds to customers. He has pointed the finger at you and we have had to take a stand.

We could look at you not giving any refunds to ANZ customers, or limiting the refund to a certain level for ANZ loans. But I’d have to go back to my guys and see about it.

We’d need you to be offering something the branches could offer as well so it’s a level playing field. If your refund was the same as the $600 establishment fee, then your refund would be matching what the branch staff can waive.

It’s not like you negotiate with the customers in an ad hoc way what amount of refund you’re giving. For example it’s not like you offer them $1000 refund, and they negotiate with you to get $1500.

59    Ms Zacka, who was an employee of AFG, and who was at the same meeting, recalled that Mr Carroll said words to the following effect (at [20] of her affidavit):

I want to get Mortgage Refunds back on board. We need to find a way around this. I need something to take back to Joe Sirianni.

The problem is you’re offering something the branches can’t offer. We need a level playing field.

You could limit your refund for ANZ loans to the amount of the $600 establishment fee, because the ANZ branches can waive that fee. Then you’d be offering the same deal as the branches.

60    The ACCC submitted that Mr Carroll’s statement was both an indication of competition between ANZ branches and Mortgage Refunds’ brokers and the anti-competitive effect of the relevant agreement or arrangement. The ACCC asked rhetorically, if the branches and the brokers did not compete, or were not in the same market, why would the bank seek to limit the refund offered by Mortgage Refunds? Why was there a need for a “level playing field”?

61    ANZ’s, and ultimately the primary judge’s, answer to this rhetorical question is dealt with later in the context of ANZ’s case.

62    The second aspect of the industry witnesses’ evidence was said to be their evidence that lenders and brokers compete.

63    In his affidavit evidence, Mr King stated as follows (at [89]):

It was part of my duties to try to refinance Westpac loans that were originated by brokers. Although the first loan would have a “break fee” to either stop or compensate for early repayment, Westpac would waive this if I churned a broker-originated loan to another Westpac loan. All lenders with a retail presence, in my sixteen year experience with residential lending, employ specialists to bring customers to them and to do so directly rather than through brokers. In all my jobs as a lending specialist and as a broker, I was competing against brokers and other lending specialists to attract and service customers. My incentives when employed by others reflected this and the incentives to Mortgage Refunds contractors reflected this. I am not aware of it ever being suggested by any bank or employer that lending specialists did not compete with brokers for customers’ business.

64    When cross-examined in relation to this statement in his affidavit, however, Mr King appeared to accept that he was speaking about competing to sell loan products. He agreed that his function or duty, initially as a sales and service representative in a Commonwealth Bank call centre and then as a personal lending officer at a branch, was to “sell” as many Commonwealth Bank loans as he could. He regarded himself as competing with other distribution channels who were selling home loans at the same time.

65    In his affidavit evidence, Mr Lahiff stated that the two strongest “channels” for all lenders with a retail presence were the lenders’ “retail branches” and the “broker channel”. By “channel” Mr Lahiff appeared to mean distribution channel. Mr Lahiff referred to the respective competitive strengths of the lenders’ retail network and the broker channel. He then stated (at [87]-[89] of his affidavit):

Over the years, brokers and lenders with retail networks have of course modified their offerings and practices to use their competitive advantages and address relative disadvantages.

Retail banks have either introduced (or where they existed, strengthened) more customer-friendly service providers to match the brokers, such as mobile lenders, mortgage specialists within branches and franchised lenders, so that the superior convenience and service of brokers is countered. Most have success based incentives which echo those of brokers.

Brokers also have taken steps to counter the retail lenders’ advantages, Mortgage Choice and other broker groups have invested heavily in advertising to establish a brand and the sort of trust that major banks have from some customers. Brokers also often invest heavily in sponsorship of, or involvement in, community groups and events, both to generate awareness for the brand and to counter the bank’s ability to profit from existing customer relationships.

66    The suggestion was, or appeared to be, that lenders and brokers changed their practises because they were in competition with each other. Mr Lahiff’s apparent suggestion that lenders and brokers competed was challenged in cross-examination by reference to statements made in the 2006 annual report of Mr Lahiff’s company, Mortgage Choice. This challenge, and the primary judge’s finding in relation to Mr Lahiff’s evidence, are addressed later in these reasons in the context of the ACCC’s appeal grounds.

67    It may immediately be observed, however, that to the extent that Mr Lahiff maintained that lenders competed with brokers, the weight to be attributed to that assertion is at best questionable. It may be that, in a general sense, brokers and lenders compete to be the one to submit a loan application on behalf of a prospective borrower. The broker wishes to earn commission. The lender may wish to avoid the cost of paying commission. But does that mean that the branches and the brokers are competing in a market for the provision of loan arrangement services? Or is that rivalrous behaviour merely an incident of activity between different distribution channels of a lender in the market for the supply of loan products by loan providers?

68    The third aspect of the industry evidence relied on by the ACCC concerned the similarity of the services provided by internal channels, such as branches, and external channels, such as brokers. In this regard, the ACCC relied, in particular, on the evidence of Mr Lahiff, Mr King and Mr Stark.

69    Mr Lahiff had worked at Westpac prior to becoming managing director of Mortgage Choice. In his affidavit evidence, he expressed the following opinion (at [69]):

In my opinion, the only significant difference between the services offered by a Mortgage Choice broker and a Westpac mortgage specialist was that the Westpac employee could offer information and advice about, and assistance in applying for, only Westpac mortgage products, whereas the Mortgage Choices broker could do so in relation to a range of lenders.

70    In cross-examination, it was put to Mr Lahiff that there were three features of the services offered by brokers that the internal distribution channels of the lenders could not offer. They were: first, access to a wide range of lenders and lenders’ products; second, advice free of any pressures or elements of favouritism relating to any one or more lenders; and third, a “path through a … maze of lenders and [their] products”. Mr Lahiff’s evidence was as follows:

Yes. All right. Thank you. And I want to put to you that those features of what brokers offered were features which the distribution channels, the internal distribution channels of lenders, could not offer, do you agree with that?---Certainly in terms of the wide range, I would agree with that.

Yes?---As to the third, of the complexity, I believe that most financial institutions were at pains to describe all of the product features in comprehensible language.

Yes?---And of the second, about being independent, unbiased, you and I are at disagreement about that issue, so certainly the first and the third.

71    Mr Lahifff’s disagreement with the proposition that the services offered by brokers was “unbiased” was based on the fact that the Australian Securities and Investments Commission had told Mortgage Choice that it was not permitted to represent that the advice it gave was unbiased because it did not have all potential lenders on its “panel”.

72    Mr King’s affidavit evidence concerning the services offered by Mortgage Refunds included the following (at [21]):

Mortgage Refunds’ service was an alternative to a borrower seeking a loan directly through a lender. A broker such as Mortgage Refunds distinguished itself from the service offered by a lender by offering access to a wide range of loans and lenders and independent comparative information, whereas a bank would only offer access to, and advice about, its own products. In addition, a broker usually offered a more convenient, flexible and personalised service than a lender, such as being available after hours and travelling to the customer’s home to meet with them.

73    Mr King’s evidence on this topic during cross-examination at trial was as follows:

Well, I want to suggest three particular differences to you and see if you agree with these. This is differences between the broker offering and the in-house channel offering. First, and perhaps foremost, offering access to the wide range of lenders and loan products. Do you agree with that?---Brokers?

Brokers, yes?---Yes.

Thank you. Secondly, independence of lenders and therefore independent advice and information?---Well, yes, we weren’t getting paid by an individual bank, you know, we got paid regardless of what loan you went through, you know, what loan you took.

Yes, but you treated Mortgage Refunds loan writers and understood Mortgage Refunds loan writers to be agents for their clients, not agents for the lenders, didn’t you?---That’s right.

Yes, and the third feature that I want to suggest to you is convenience, flexibility and personalisation of service. Is that something you agree with or disagree with?---No, definitely.

Thank you. And not only were those features features [sic] offered and provided by Mortgage Refunds in its broking activities or those of its loan writers, but you would regard those three features as generally proffered by mortgage brokers throughout the industry, wouldn’t you?---By good mortgage brokers, yes.

Yes?---Yes.

Yes, in 2004 to 2006?---Yes.

Yes, thank you. So is this the position, while a number of the things that both the branches and the brokers were doing coincided, such as providing information, such as helping with application forms, although there was overlap in activity, there was significant difference in the elements of the service offering which differentiated brokers on the one hand from branches on the other hand?---Well, I would say the only difference, really, would be the fact that if you walk into a bank, unless they’re writing other peoples’ paper, you’ve only got the option of writing their home loans so you’re tied into their policies, their procedures, their costs, their interest rates and so forth.

Their products?---The only difference between that and being a broker is the fact that you’ve got 30 or 40 different lenders and all those different policies and procedures and so forth that you can work with to tail it to the individual customer’s needs.

Plus the independence which you agreed to a few minutes ago?---Yes.

Plus the convenience you agreed to a few minutes?---Yes.

74    Prior to joining Mortgage Refunds, Mr Stark worked as a mortgage lender at Suncorp Banking Corporation (Suncorp). Mr Stark’s affidavit evidence concerning the work he did at Suncorp as compared to the work he did at Mortgage Choice was as follows (at [18]):

In my view, the work that I did for Mortgage Refunds was almost identical to the work I performed in my mortgage lending role at Suncorp. The main differences in my view were that I was acting now as a broker who had access to loan products from a whole range of lenders, as opposed to acting ‘in-house’, and so I was able to offer loan products from other lenders in addition to Suncorp products. The other difference was that, as a broker for Mortgage Refunds, I did not have access to the lender’s internal credit department as I did in my role at Suncorp.

75    Like the other industry witnesses, Mr Stark readily agreed, in cross-examination that there were differences between the role he played as a broker and the role he played as a bank officer. He agreed that when he was at Suncorp he was working for and to advance Suncorp, whereas at Mortgage Refunds he understood his function was to act on behalf of the customer or client, rather than to act on behalf of the lender. He also agreed that a borrower who comes to a broker wants advice about a range of options that are available in the marketplace. When a borrower goes directly to a bank “they have obviously done their own research … they have pretty much taken their own matters into their own hands”.

The customer evidence

76    The ACCC also relied on evidence of customers of Mortgage Refunds in support of the proposition that the services offered by lenders or bank branches were fundamentally the same as the services offered by brokers.

77    Mr O’Malia had obtained mortgage finance from Suncorp at a time when Mr King worked as a mobile lending manager at Suncorp. Later, after Mr King had left Suncorp and established Mortgage Refunds, Mr O’Malia approached Mr King at Mortgage Refunds in relation to potential mortgage finance for the purchase of two properties. Mr O’Malia’s evidence was that “Mortgage Refunds provided essentially the same services” as had been provided by Suncorp “except that with Mortgage Refunds we were offered access to loans from more than one lender.” In cross-examination, Mr O’Malia agreed that it was an advantage to him to get comparative information about the available choices from various lenders and it saved him the trouble of doing the comparative shopping himself “by going from one bank to the next, to the next”.

78    The evidence of Mr Black was very similar. Mr Black had obtained mortgage finance from HSBC when Mr King was employed at that bank. He later dealt with Mr King when Mr King was at Mortgage Refunds. Mr Black’s evidence was that (at [13] of his affidavit):

The service Jason [King] provided me at Mortgage Refunds was fundamentally the same as that which he had provided at HSBC. The only difference was that Jason could now offer me information and assistance with a broader range of loan products from a number of lenders and not solely HSBC products.

79    It would appear from Mr Black’s evidence during cross-examination that the difference he identified was of some significance, at least to him. Like Mr O’Malia, Mr Black saw it as a significant advantage that Mortgage Refunds was able to “shop the market” and save him the time and effort of having to research the advantages and disadvantages of loans offered by different lenders himself.

The ANZ documents

80    The documentary evidence was and is significant to the ACCC’s case. The ACCC relied heavily on statements made in a number of ANZ’s internal documents as supporting the inference or conclusion that ANZ saw its branches as being in competition with external brokers. The documents were also the basis for a number of inferences or assumptions relied on by the ACCC’s expert economist, Dr FitzGerald, in expressing his opinions concerning competition in the relevant markets.

81    Following is a brief summary of the key documents highlighted by the ACCC, in its written and oral submissions on appeal.

82    In a document recording a presentation given by Mr Chris Cooper, the chief executive officer of ANZ Mortgage Group on 16 September 2003, the following statements were made:

    The key elements of the Mortgage Group’s “distribution strategy” included establishing “a franchised mobile specialist salesforce that will replicate broker performance to win new customers”.

    “ANZ’s proposed franchise mortgage distribution, in part, will counter broker growth”.

    The evolution of the Australian mortgage industry included, in 2004, “[l]enders’ sales forces compete more closely with brokers (potential to offer product choice)”.

    “With the specialist business model, the natural next step for the Group is to move to a specialist distribution model in order to compete with the external specialists (the brokers)”.

    ANZ has an “over reliance on brokers … exposing the business to sustainability risk and margin pressure”.

83    In a document that records a presentation by an ANZ officer entitled “Channel Neutrality” on 21 March 2005, the following statements were made:

    “ANZ Mortgages sells mortgage products simultaneously through a number of channels. In many instances, these channels are competing to reach the same set of customers.”

    “To ensure all distribution channels compete on an equal footing for customers, ANZ Mortgages has a strict policy of channel neutrality, where all channels receive the same pricing and fee structures”.

84    In an email exchange in early 2004 between an external broker and an officer of ANZ Mortgage Group, the broker and the ANZ officer discussed a complaint made by another external broker that they had been “undermined” by an ANZ officer offering “discounted terms”. In one of his emails, the ANZ Mortgage Group officer stated:

ANZ’s guiding principal [sic] is that we are channel neutral to all our distribution channels re: product, credit & pricing.

…

No one owns the customer except the customer. ANZ and also the broker have equal rights to market to the client.

What is offered by the channel as an inducement to secure the business in [sic] not my concern. In fact it is extremely common for brokers to offer significant incentives to the client to win the business. This is outside my control and recognise that this is a market factor.

85    In an email exchange in April 2004 between ANZ Mortgage Group officers concerning an issue that had arisen concerning an external broker offering a customer a rebate of $2,800 if the customer secured an ANZ loan of $700,000 through the broker, one of the officers stated:

One observation could be that Brokers are capturing too much of the Mortgage Value Chain if they can afford to to [sic] this … This has potential to further erode value in the market place, depending of course how much market share they capture. How do we compete with this sort of stuff?

The economist’s opinion

86    The ACCC tendered two reports of Dr FitzGerald. The second report responded to a report prepared by an economist retained by ANZ, though ultimately ANZ called no evidence from that or any other economist.

87    The judgment of the primary judge contains a detailed summary of Dr FitzGerald’s evidence. It is unnecessary to repeat that detailed summary here. It is sufficient to identify the substance of the critical opinions of Dr FitzGerald recorded in his reports. They are as follows:

(1)    There is a market in which mortgage brokers compete. It is “a market providing services to participants in the market for mortgage loans”.

(2)    “Brokers provide both loan arrangement (or broking) services to prospective borrowers and distribution services to loan providers”. The market is accordingly a two-sided market.

(3)    Brokers compete in a market for intermediary services, essentially a “functional market”. To the loan providers, predominantly the retail mortgage lending arms of banks, brokers represent a channel through which the business of eligible borrowers is brought to them. That is, “a distribution channel for their loan products.”

(4)    In relation to the market for loan arrangement services (as defined in the ACCC’s pleading), brokers compete with the following categories of parties:

(a)    other brokers in the same geographic market who are accredited in respect of loan products that differ from the loan products in respect of which the broker in question is accredited;

(b)    other brokers in the same geographic market which are accredited only in relation to the loan products of a single loan provider;

(c)    persons franchised by a particular loan provider to provide loan arrangement services or broking services, in the same geographic market, in respect of the loan products of that loan provider;

(d)    loan providers which, in respect of their own loan products, engage or employ persons to provide, and who specialise in providing, the following advice and assistance to persons seeking home loans: advice as to the features of the respective loan products available from that loan provider; advice as to which of that loan provider’s loan products are available to persons in those person’s circumstances; advice as to which of that loan provider’s loan products are best suited to the person’s needs; and assistance to complete and lodge applications for the loan product selected in a manner that meets the requirements of that loan provider.

(5)    Brokers did not compete with lenders themselves in respect of loan products, but only with other intermediaries providing loan arrangement services to the lenders, including lenders’ tied channels.

88    An important aspect of Dr FitzGerald’s analysis was that banks, such as ANZ, were generally divided into separate business or economic units (or cost and profit centres). In the case of ANZ, ANZ Mortgage Group was the business unit that was effectively the lender. It developed and provided the mortgage products. It did not, however, generate referrals of eligible prospective borrowers or provide advice or assistance to them. That function was mainly performed by the Personal Banking Business unit which managed the ANZ branch network.

89    The recognition of these separate business units was important to Dr FitzGerald’s opinion because otherwise it made no sense to regard the services provided by ANZ branches as services provided by an intermediary. ANZ could not provide intermediary services to itself unless the Personal Banking division of ANZ was accepted as being, from an economic perspective, a distinct economic entity that provided intermediary services to ANZ Mortgage Group, another separate business unit, as lender.

90    Critical also to Dr FitzGerald’s opinion was the conclusion or opinion that the services provided by a lender’s in-house or tied channel (for example a branch) were not significantly or fundamentally different to the services provided by an independent broker. Dr FitzGerald maintained that, in economic terms, the services provided by branches and brokers were relatively close substitutes. That was the case, according to Dr FitzGerald, even though the branches were only able to, or likely to, provide advice or assistance in relation to the bank’s own products, whereas brokers were able to provide comparative advice concerning the products of numerous lenders.

91    As will be seen, the primary judge rejected this aspect of Dr FitzGerald’s evidence. His Honour found that the services provided by branches and brokers were not close substitutes. This finding is the subject of a ground of appeal and is considered later in these reasons in that context.

ANZ’S case at Trial

92    ANZ’s case at trial was, in short, that there was no loan arrangement services market as contended by the ACCC. The fact that there may have been some common features of the advice and assistance ANZ branch officers provided to prospective ANZ loan applicants, and the services provided by brokers to their clients or customers, did not mean that there was a relevant market for the supply of services comprising those common features. It was ANZ’s case that the advice and assistance provided by ANZ branch officers was provided as part of the sales process in the market for the supply of loan products by loan providers. The services provided by brokers were provided in a different market, being the market for the supply of broking services. ANZ’s internal distribution channels did not compete in that market.

93    ANZ’s defence to the ACCC’s pleading made the following assertions or denials, expressed below in summary and simplified terms, in relation to the matters relevant to the issues that arise in this appeal:

    ANZ’s business included supplying ANZ loan products (paragraph 1).

    Applicants for ANZ loan products (including ANZ home loan products) applied either directly to ANZ, or through, inter alia, accredited mortgage brokers (paragraph 4).

    Mortgage brokers provide broking services (“broking services”) to their customers, namely:

(a)    “the provision of advice and services to assist customers to identify an appropriate or preferable credit supplier or suppliers (which may or may not be or include ANZ) and loan products (which may or may not include ANZ Loan Products or ANZ Home Loan Products) suitable for the needs of that customer; and

(b)    otherwise represent their customer’s interests in dealings with the credit suppliers to whom any loan applications are made” (paragraph 6).

    Where a successful applicant for an ANZ loan product (including an ANZ home loan product) was introduced by a mortgage broker, ANZ paid the broker a fee irrespective of what broking services were provided to the broker’s customer (being the successful loan applicant), calculated as a percentage of the loan (paragraph 7).

    Mortgage brokers do not supply loans, including ANZ loans (paragraph 9(a)).

    ANZ does not provide broking services of the sort provided by mortgage brokers and does not provide “loan arrangement services” as defined in the ACCC’s pleading (paragraph 9(b)).

    ANZ is not, and has not at any time, been in competition with providers of broking services or loan arrangement services (paragraph 9(c)).

    Specifically, ANZ was not in competition with Mortgage Refunds, or any of the brokers allegedly engaged by it, either in relation to the supply of loan arrangement services or loan arrangement services in respect of ANZ loan products (paragraph 27).

    Mortgage Refunds (as opposed to the individual brokers associated with it) did not supply either loan arrangement or broking services to customers in respect of ANZ loans. The individual brokers were not employees or agents of Mortgage Refunds (paragraph 26B).

    The individual brokers associated with Mortgage Refunds were not parties to any agreement recorded in or evidenced by the 29 April 2004 letter from ANZ to Mortgage Refunds (paragraph 35).

94    ANZ did not call evidence from any witness. It did, however, rely, in a positive sense, on some of the evidence given by the ACCC’s witnesses in cross-examination. It also relied on parts of the documentary evidence tendered by the ACCC, as well as additional documents tendered by ANZ.

95    It is unnecessary to detail here the particular oral or documentary evidence relied on by ANZ at trial. Much of it is referred to later in the context of the ACCC’s appeal grounds. It is sufficient to note two aspects of the evidence relied on by ANZ.

96    The first point to note is that ANZ contended at trial, as it does on appeal, that the evidence of both the industry witnesses and Mortgage Refunds’ customers reveal three important differences between the services supplied by mortgage brokers and the activities engaged in by ANZ bank officers.

97    First, the services provided by brokers offered potential customers choice, in that they assisted potential borrowers to select and apply for loan products from a wide range of lenders. The internal distribution channels of banks, on the other hand, only provided advice and assistance in relation to their own loan products. The activities of bank employees, including ANZ employees, therefore could not offer or provide choice to customers.

98    Second, the advice and assistance provided by brokers was independent, in the sense that the brokers were not tied to a single lender. Brokers acted for and on behalf of the customers to find the best loan for the customer’s needs. Bank officers, including ANZ officers, on the other hand, acted primarily in the interests of their bank by selling the bank’s own loan products.

99    Third, the services offered by brokers gave the customers the convenience of a “one-stop shop” where the customer could be provided with advice comparing the advantages and disadvantages of a wide range of products from a wide range of lenders. Bank officers could not reasonably provide this comparative advice service.

100    The second aspect of the evidence to note concerns the documentary evidence. As already indicated, the ACCC’s case relied fairly heavily on statements in various ANZ internal documents that referred to competition between ANZ branches and external brokers.

101    ANZ’s case at trial was, as it was on appeal, that these passages from the ANZ internal documents, when read in context and not in isolation, did not relate to competition between branches and brokers in relation to the supply of loan arrangement services. Rather, they referred or related either to competition in the market for the supply of loan products, or to ANZ’s intention or desire to maintain “channel neutrality” between the various internal and external distributors of its loan products. Channel neutrality meant that, when “selling” or distributing ANZ loan products, the internal and external channels were to receive the same pricing and fee structures in relation to the loan products. ANZ’s policy was to provide a “level playing field” to its distribution channels, both internal and external.

102    In this respect, the documents also showed, in ANZ’s submission, that ANZ and the brokers were not competitors, but rather were partners or collaborators. It was in ANZ’s interests to keep them happy with the relationships because they introduced customers to ANZ that may not necessarily have otherwise presented to ANZ branches. Whilst ANZ may have preferred to distribute more loans through its branches, because it did not then have to pay commission to an external entity, brokers were nonetheless seen as a “necessary evil”.

The Judgment of the primary judge

103    After setting out the broad context and summarising the uncontentious facts (at [4]-[14]), including the facts surrounding the impugned conduct (at [16]-[22]), and outlining the ACCC’s contentions in relation to the relevant market and competition (at [23]-[31]), the primary judge examined the evidence in considerable detail. His Honour considered the evidence of each witness (at [32]-[477]). Within that detailed discussion, his Honour examined some of the relevant documentary evidence. His Honour also gave separate detailed consideration to the internal ANZ documents and emails (at [478]-[520]). Within this detailed consideration and analysis of the evidence of the individual witnesses and the documents, his Honour made a number of findings or observations concerning the effect of, or weight to be given to, parts of the evidence. It is unnecessary to set out all of those findings in detail here, but it will be necessary to consider some of them in the context of the ACCC’s appeal grounds.

104    At [521]-[552], the primary judge summarised the evidence and some of his findings in relation to it. His Honour’s summary included the following:

    With some minor qualifications, the witnesses were accepted as being honest and reliable so far as matters of fact were concerned: see [521].

    To the extent that witnesses made assumptions or drew inferences, those assumptions were not necessarily accepted. His Honour said that he would test the “validity” of the assumptions and inferences having regard to the evidence. His Honour referred, in this context, specifically to Dr FitzGerald: see [521]. His Honour’s findings relating to Dr FitzGerald’s evidence are a significant issue on this appeal and will be addressed in detail later in these reasons.

    His Honour did not give “great weight” to the evidence of Messrs O’Malia and Black (the Mortgage Refunds customers) concerning the similarity of the services provided by Mr King when he was employed by a lender and when he operated as an independent broker: see [522].

    In relation to the “industry witnesses” (Messrs King, Stark, Lahiff and Percy) to the effect that the “only difference” between the services offered by brokers and the services offered “in-house” was that the brokers’ services related to a wider range of lenders and products, his Honour concluded that “[t]hat proposition tends to conceal the extent of the relevant difference”. His Honour concluded that the industry evidence “demonstrates that access to a wide range of products supplied by a wide range of lenders, including the major banks, was essential to the success of a broking business”: see [524].

    On this point, his Honour also expressed the view that Dr FitzGerald failed to appreciate the full range of services offered by brokers. As will be seen, the inference or assumption made by Dr FitzGerald concerning the equivalence of the services provided by brokers and “in-house” or “tied” channels was important to his market opinion. His Honour’s findings in relation to this inference or assumption were critical to his Honour’s rejection of Dr FitzGerald’s market opinion: see in particular [524].

    Mr King’s evidence suggested that the services offered by brokers were significantly different from those offered by in-house and tied channels: see [527]. Brokers offered services which in-house and tied channels did not provide: see [528].

    Mr Stark also acknowledged the differences between the services offered by brokers and the services offered in-house by lenders, in particular in relation to the range of products, the choice of different lenders, and independence: see [529].

    In relation to the evidence of the industry witnesses to the effect that brokers competed with lenders, his Honour noted that “it is not at all clear that they recognised ACCC’s distinction between the supply of loan arrangement services and the supply of loan products.” His Honour inferred that they were speaking of the latter: see [526].

    His Honour did not accept Mr Lahiff’s evidence that brokers (such as those engaged by his company, Mortgage Choice) competed with lenders. That was in large part because Mortgage Choice’s contemporaneous documentation did not support the view that it was in close competition with lenders rather than other brokers: see [532].

    Mr Percy’s view was that the internal channels of Bendigo and Adelaide Bank Ltd competed with brokers in the provision of loan products, not loan arrangement services or broking services: see [533].

    Ultimately, the industry witnesses, in the view of the primary judge, offered little support for the ACCC’s case: see [534].

    The industry evidence also did not support many of Dr FitzGerald’s assumptions and conclusions: see [535]. His Honour identified four examples:

(a)    First, one of Dr FitzGerald’s reasons for concluding that the products (the services provided by lenders and brokers) are substitutable is that the loan products themselves are not significantly differentiated, with the result that “competition will focus largely on the accompanying services.” The industry evidence, however, showed that there were differences between the lending criteria, and that lending criteria and product characteristics vary over time.

(b)    Second, his Honour expressed doubts about the inferences drawn, or the reliance placed on, Dr FitzGerald’s belief that many borrowers obtain significant amounts of information from the internet. Issues relating to the use of the internet were either not explored, or suggested that brokers’ services were significantly different from those provided by a lender’s in-house or tied channels: see [536]-[537].

(c)    Third, Dr FitzGerald seems to have discounted the importance to potential borrowers of a broker’s independence from potential lenders: see [538].

(d)    Fourth, his Honour had difficulty with Dr FitzGerald’s approach to a potential borrower’s purpose when he or she approaches either a broker or a lender. A potential borrower approaches a broker in order to obtain advice and assistance for the purpose of applying for a loan. When a potential borrower approaches a lender directly, he or she will generally be simply seeking a loan from that lender. Brokers offer an alternative to the time-consuming task of borrowers visiting multiple lenders to obtain advice and assistance about available products: see [539].

    His Honour did not accept that the internal ANZ documents relied on by the ACCC were “all one way” as the ACCC contended: see [552]. His Honour concluded that a number of the documents showed that ANZ wanted to increase market share in the lending market, rather than in any market for loan arrangement services or broking services: see [551]. Whilst they also showed that ANZ wanted to reduce reliance on brokers, they equally revealed that ANZ did not want to risk broker support by departing from the principle of channel neutrality: see [551].

105    At [553]-[561], his Honour outlined some principles concerning competition and markets by reference to passages from the reasons or judgments in some of well-known authorities, including Re Queensland Co-operative Milling Association Ltd; Re Defiance Holdings Ltd (1976) 8 ALR 481 at 514-516; 25 FLR 169 at 187-189; Queensland Wire Industries Pty Ltd v Broken Hill Proprietary Company Ltd (1989) 167 CLR 177 at 187-188, 195-196, 198-200; and Arnotts Ltd v Trade Practices Commission (1990) 24 FCR 313 at 331-332.

106    Neither party contended that his Honour’s statement of the relevant principles relating to market definition and competition was in any way, erroneous or deficient.

107    At [562]-[615], the primary judge summarised the submissions of the parties. In the course of so doing, his Honour repeated or referred to a number of factual findings either made earlier in the judgment or elaborated on later. It is unnecessary to rehearse here his Honour’s summary of the submissions. It is also unnecessary to refer to, or repeat, all of the findings made by his Honour in this part of the judgment. The most important findings were as follows:

    His Honour rejected the ACCC’s submission that the agreement between ANZ and Mortgage Refunds recorded or evidenced in the 29 April 2004 letter was motivated by an anti-competitive purpose which showed that ANZ and Mortgage Refunds were in competition. His Honour said (at [581]) that “[h]aving regard to the apparently isolated nature of the conduct and the limited nature of the restraint imposed, I am inclined to the view that ANZ was trying to keep the branches happy rather than seeking to restrain perceived competition.”

    His Honour effectively rejected the ACCC’s submission that the payment of commission was in respect of the service supplied by the broker to the customer. His Honour said (at [589]) that it “would be more realistic to say that the payment of the commission is in respect of services supplied to the lender by the broker” and that “[o]ne might say that the commission was for effectively bringing about a sale of an ANZ loan product.” It is but a short step from this finding to conclude, as his Honour ultimately did, that the relevant market in which the conduct occurred was the market for the supply of loan products, not any market for the supply of loan arrangement services.

    The primary judge rejected ANZ’s submission that the relevant agreement between ANZ and Mortgage Refunds concerning the maximum refund did not relate to any discount, rebate, allowance or credit within the meaning of s 45A of the Act. His Honour held that it was “at least arguable” that the refund fell within the ordinary (dictionary) meaning of “rebate”, which included a deduction from a sum of money to be paid, or a partial refund of money paid (at [610]). His Honour’s finding in this regard is the subject of ANZ’s cross-appeal.

108    At [616] to [657] of the judgment, the primary judge set out his dispositive findings and reasoning. His Honour made five critical findings that were effectively fatal to the ACCC’s case.

109    The first critical finding was that his Honour was not satisfied that ANZ branches supplied loan arrangement services (as defined in the ACCC’s pleading) to potential borrowers: see [631] and [634]. It is common ground that the reference to “lenders” in [631] should be read as “borrowers”. Rather, his Honour concluded that they supplied sales services to ANZ Mortgage Group and/or to ANZ.

110    There are a number of important points that should be emphasised in relation to this finding. These points are elaborated on later in these reasons in the context of the ACCC’s challenge to this finding and his Honour’s reasons for making it. It is sufficient here to make four short points.

111    First, the finding is a finding of fact concerning the nature or proper characterisation of the services provided by ANZ branch staff in the course of their interaction with prospective borrowers.

112    Second, his Honour was not saying that ANZ branch staff did not provide, at various times, some or all of the things that the ACCC alleged constituted “loan arrangement services”. His Honour found, in effect, that ANZ staff and employees did do some or all of those things from time to time: see [622] and [630]. The issue was the proper characterisation of those things.

113    Third, his Honour characterised the things done by ANZ branch staff, which the ACCC described as “loan arrangement services”, as functions or duties performed in the course of selling or distributing ANZ’s loan products: see in particular [619]-[620].

114    Fourth, whilst his Honour said that one could (perhaps loosely) describe the things done by the branch staff as “sales services” provided to ANZ or ANZ Mortgage Group, they could not properly be characterised as the provision of services to prospective customers, as distinct from the performance of functions associated with the supply by ANZ of loan products. His Honour characterised the interaction between an ANZ branch officer or franchisee and a prospective borrower in the following terms (at [625]):

Thus the process does not resemble the supply of a service. Rather, it resembles an interactive procedure between the potential borrower and a representative of the lender, anterior to the making of a loan. It is a process in which a potential borrower must participate in order to obtain a loan product. To characterize it as the provision of services would lead to the similar characterization of any pre-contractual negotiations.

115    The second critical finding made by the primary judge was that ANZ franchisees were effectively in the same position as the branches. The franchisees performed much the same functions as the branches: see [633]. They were “selling ANZ loan products” and thereby supplying “sales services” to ANZ Mortgage Group: see [633].

116    The third critical finding was that mortgage brokers (one of the so-called “external channels”) “performed a different function”: see [635]. They held themselves out as offering, or undertaking to, provide advice, information and assistance to potential borrowers who consulted them concerning a wide range of products, supplied by a wide range of lenders: see [635]-[636]. His Honour thus had “no difficulty in characterising the brokers’ function as involving the supply of services to potential borrowers.” His Honour also found, however, that, like the branches, the brokers were providing sales services to lenders. This service was provided once a customer had decided which loan to apply for. Thus, whilst his Honour does not say so in terms, his Honour appears to accept that brokers were participating in the sort of two-sided market described by Dr FitzGerald. The problem for the ACCC’s case was that his Honour found that ANZ, through its internal channels, was not.

117    The fourth critical finding is really a conclusion that flows from the first three findings. It is expressed in [638] in the following terms:

I have concluded that ANZ branches and franchisees did not participate in any market in which the brokers provided loan arrangement services to potential borrowers. That finding leads to the conclusion that ANZ was not, in any relevant sense, in competition with Mortgage Refunds. Thus s 45A is not engaged. The proceedings must be dismissed.

118    The fifth critical finding perhaps provides an independent or alternative basis for arriving at the conclusion that ANZ branches and franchisees did not compete with brokers in a loan arrangement services market. His Honour found that there were significant differences between the services provided by ANZ branches and franchisees and the services provided by brokers. His Honour thus appears to conclude (at [643]-[648]) that, if ANZ branch officers did supply services to prospective borrowers, such services as were provided by ANZ branches and franchisees were not close substitutes for the services provided by mortgage brokers. His Honour’s reasons for so concluding included what he described as a “fairly basic SSNIP [Small but Significant and Non-transitory Increase in Price] test”: see [644]-[648].

119    His Honour made some other findings concerning the relationship between Mortgage Refunds and the individual brokers associated with it: see [649]-[657]. These findings were not ultimately particularly important to his Honour’s disposition of the matter and are not of any particular significance to this appeal. The only other finding that warrants mention, if only because it is the subject of a ground of appeal, is that his Honour expressed doubt that it was possible, on the evidence, to conclude that Mortgage Refunds, as distinct from the individual brokers associated with it, supplied broking services.

120    To the extent that his Honour made any other relevant findings, they are referred to later in these reasons in the context of the ACCC’s appeal grounds.

Appeal grounds and submissions

121    The ACCC’s notice of appeal contains twelve grounds of appeal. Each of the grounds is directed, in one way or another, at challenging one or more of the critical findings made by the primary judge to which reference has just been made, or his Honour’s reasons for arriving at those findings. Many of the appeal grounds are lengthy and contain detailed particulars of circumstances which, on the ACCC’s case, suggest error on the part of the primary judge. The grounds of appeal maybe summarised as follows:

(1)    The primary judge erred in finding (at [631]) that ANZ did not supply loan arrangement services as defined in paragraph 7.2 of the FASC.

(2)    The primary judge erred (at [656]) when he failed to find that Mortgage Refunds provided loan arrangement services.

(3)    The primary judge erred in rejecting, failing to take into account, or failing to give appropriate weight to, the evidence of Dr FitzGerald that ANZ provided loan arrangement services to customers.

(4)    The primary judge erred when he rejected, failed to take into account, or failed to give appropriate weight to the evidence of the “market participants” in relation to the lack of differentiation between the services provided by loan providers and brokers, the substitutability of those services, and the competition between loan providers and brokers.

(5)    The primary judge erred in failing to find that ANZ’s internal documents showed that ANZ’s internal distribution channels were in competition with brokers.

(6)    The primary judge erred in finding that the services provided by loan providers’ internal distribution channels and the services provided by brokers were not close substitutes.

(7)    The primary judge erred in relying on the proposition (at [622], [625] and [626]) that the differing nature of the legal rights, duties and liabilities as between bank branch and customer, on the one hand, and broker and customer, on the other, was a significant criterion in determining the question whether the services supplied by bank branches and brokers were close substitutes.

(8)    The primary judge erred (at [539], [626]-[628]) in relying upon his own understanding of the purpose for which customers seeking loans attended ANZ branches and his own views about the interaction of such persons and bank staff in the absence of any evidence to support those views.

(9)    The primary judge erred (at [638]-[639]) when he failed to find that ANZ provided loan arrangement services in competition with Mortgage Refunds.

(10)    The primary judge erred in failing to find that the evidence of Mr King and Ms Zacka concerning their conversation with ANZ’s Mr Carroll supported the conclusion that there was competition between ANZ’s branches and Mortgage Refunds or brokers and that the agreement was for the purpose of substantially lessening competition.

(11)    The primary judge erred in failing to find that the agreement between ANZ and Mortgage Refunds contravened s 45(2)(a)(ii) of the Act by virtue of the then applicable s 45A of the Act.

(12)    The primary judge erred when he failed to find that ANZ gave effect to a provision of that agreement which, by virtue of the then applicable s 45A, contravened s 45(2)(b)(ii) of the Act.

122    It is unnecessary to rehearse here the ACCC’s submissions in support of these grounds. The submissions are addressed later in this judgment in the context of each of the grounds. It is sufficient, at this stage, to make the following few points.

123    First, the ACCC’s submissions, both written and oral, did not directly engage with all of the grounds. The submissions did not separately or distinctly address each ground. Rather, for the most part, the ACCC’s submissions challenged the critical findings of the primary judge at a more global or general level. In effect, the ACCC submitted that the critical findings by the primary judge were either contrary to the evidence, not supported by the evidence, or at least against the weight of the evidence.

124    Second, the ACCC did not contend that the primary judge misconstrued either s 45 or s 45A of the Act, or failed to apply the correct principles in relation to the critical issues concerning the identification of the relevant market or markets, or competition or substitutability. Rather, the essence of the ACCC’s case on appeal was that the critical findings made by his Honour, or other aspects of his Honour’s reasoning in support of these findings, were contrary to the evidence, not supported by the evidence, or at least against the weight of the evidence. As part of this attack on the findings, the ACCC complained that the primary judge either ignored, incorrectly rejected, or gave insufficient weight to evidence relied on by the ACCC.

125    ANZ, in its submissions, characterised the ACCC’s case on appeal as amounting to no more than an attempt to reargue the case on the basis of dissatisfaction that the primary judge did not accept, or was not satisfied, that the evidence made out the ACCC’s case as pleaded. ANZ submitted, in essence, that when regard is had to the totality of the evidence, the findings made by the primary judge were open and that there was no warrant for appellant interference.

Consideration of appeal grounds

126    Before addressing the appeal grounds individually, some general observations should be made about two matters.

127    First, despite the fact that the ACCC’s grounds did not directly raise any question of principle in relation to the concepts of market and competition, some general observations should be made concerning those concepts. Some observations should also be made concerning the ACCC’s defined “loan arrangement services” market, the existence of which was rejected by the primary judge. Did this market definition truly reflect the commercial relationships and interactions between banks (including ANZ) and mortgage brokers (including Mortgage Refunds), on the one hand, and borrowers and prospective borrowers, on the other, in the market for residential property loans? Or was it a somewhat artificial construct?

128    Second, it is necessary to say something briefly about the principles to apply where, as here, an appellant challenges on appeal factual findings made at trial. Whilst the critical findings made by the trial judge here, particularly those in relation to competition and markets, were in some respects evaluative conclusions or inferences, they were nonetheless findings of fact or based on findings of fact.

Markets and Competition

129    Lying at the heart of the issues for consideration and determination in this matter are the twin concepts of market and competition. To make out its case that ANZ contravened s 45 of the Act, the ACCC was required to prove, to put it in simple terms, that the relevant agreement between ANZ and Mortgage Refunds had the purpose or effect of substantially lessening competition. Competition in that context means, by reason of s 45(3) of the Act, competition “in any market” in which either ANZ or Mortgage Refunds supplied or acquired goods or services.

130    The ACCC also relied on s 45A to establish that the agreement had the purpose or effect of substantially lessening competition. To rely on s 45A, the ACCC was required to prove that the relevant agreement had the purpose or effect of, relevantly, fixing, controlling or maintaining a rebate in relation to services supplied by both ANZ and Mortgage Refunds in competition with each other. Competition, in that context, must again mean competition in a market.

131    The critical question in this matter was and is whether ANZ and Mortgage Refunds competed in a market for the supply of “loan arrangement services” as defined in the ACCC’s pleading.

132    The Act does not include an exhaustive definition of “market”. Section 4E (extracted earlier) relevantly provides that a market in relation to goods and services includes a market for those goods and services and other goods and services that are “substitutable for, or otherwise competitive with,” those goods and services.

133    Much has been written about the meaning of a “market”. It is neither necessary nor desirable to significantly add to that body of writing here. The word is “not susceptible of precise comprehensive definition when used as an abstract noun in an economic context”: Queensland Wire at 195-196 (Deane J). It may generally be described as the area of actual or potential close competition or rivalry between firms in respect of particular goods or services and their substitutes: see generally Queensland Wire at 187-188 (Mason CJ and Wilson J), 195-196 (Deane J) and 198-200 (Dawson J). The dimensions of that area of economic activity may be defined by reference to function (that is, wholesale, retail or both), product (that is, the nature and characteristics of the goods or services), and geography (for example, local, state or national): Australian Gas Light Company v Australian Competition and Consumer Commission (No 3) (2003) 137 FCR 317 at [378].

134    The leading authorities emphasise the importance of the substitutability of the relevant products (goods or services) in the market. In simple terms, substitutability means that buyers and sellers can and will substitute one product for the other in response to changes in prices. That does not mean that the products have to be identical. As Wilcox J explained in Trade Practices Commission v Australia Meat Holdings Pty Ltd (1988) 83 ALR 299 at 317:

The existence of price differentials between different products, reflecting differences in quality or other characteristics of the products, does not by itself place the products in different markets. The test of whether or not there are different markets is based on what happens (or would happen) on either the demand or the supply side in response to a change in relative price.

135    Whilst relatively easy to describe, at least in a general sense, it is often difficult to identify and define the nature and parameters of a particular market in any given circumstance. Market definition “involves value judgments about which there is some room for legitimate differences of opinion”: Queensland Wire at 195-196 (Deane J). Market identification and definition is not an exact science. It is rooted in the analysis of commerce as an aspect of human behaviour.

136    For present purposes it is important to emphasise that a market is “not a feature of the real world” but rather “an analytical tool devised by economists”: Seven Network Ltd v News Ltd (2009) 182 FCR 160 at 350-351 (Dowsett and Lander JJ). The identification of a market in any given case is therefore generally purposive and directed to the problem or issue at hand. As explained by Allsop J in Australian Competition and Consumer Commission v Liquorland (Australia) Pty Ltd [2006] ATPR 42-123; FCA 826 at [429]:

Market definition is not an exact physical exercise to identify a physical feature of the world; nor is it the enquiry after the nature of some form of essential existence. Rather, it is the recognition and use of an economic tool or instrumental concept related to market power, constraints on power and the competitive process which is best adapted to analyse the asserted anti-competitive conduct.

137    The process of market identification or definition is therefore to be undertaken with a view to assessing whether the substantive criteria for the particular contravention in issue are satisfied, in the commercial context the subject of analysis. Here, in simple terms, the process is undertaken in order to determine if the impugned conduct (the agreement between ANZ and Mortgage Refunds) had the purpose or effect of fixing, controlling or maintaining a rebate, or otherwise had the purpose or effect of substantially lessening competition, in a market in which the parties to the agreement competed in respect of the supply of services.

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 (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.

139    There is, it must be said, an element of artificiality or contrivance in the market definition advanced by the ACCC in this matter.

140    In the particular facts and circumstances of this matter, the identification or definition of the relevant market, if any, in which the impugned anti-competitive conduct occurred, essentially involves three questions.

141    The first question is whether both the ANZ in-house channels (primarily the branches) and the external distribution channels (Mortgage Refunds and other brokers) provide advice and assistance to prospective borrowers or loan applicants of the type alleged in paragraph 7.2 of the ACCC’s pleading and defined as “loan arrangement services”?

142    The primary judge concluded that both the internal and external distribution channels did, on occasion, provide some or all of the advice or assistance said to comprise “loan arrangement services”. His Honour did, however refer to two important qualifications. In particular, his Honour pointed out (at [625]):

It is not clear whether the pleaded “loan arrangement services” comprise a discrete and indivisible bundle or alternatively, a range of services, only some of which may be supplied in a particular case. A potential borrower may have already chosen the product so that no advice will be necessary. He or she may not have satisfied the lending criteria, in which case no application would be prepared or submitted.

143    The evidence relating to the provision of the advice or assistance said to comprise loan arrangement services was extremely general in nature. In the case of ANZ specifically, it was in the form of answers to interrogatories framed in very broad and general terms. There was no evidence that every, let alone most, or even a substantial proportion, of prospective ANZ loan applicants received all, or even any of the advice or assistance said to comprise loan arrangement services. It was, as his Honour observed, possible that some successful loan applicants simply filled out and submitted a loan application without any, or any significant, advice or assistance at all.

144    This qualification and the limited evidence in relation to the actual advice and assistance provided by loan providers, such as ANZ, was one of the matters that led his Honour to doubt that this advice or assistance could properly be said to amount to the provision of a service.

145    The second qualification was that loan providers such as ANZ did not provide advice and assistance in relation to loan products generally. For obvious reasons, they generally only provided it in relation to their own products. Brokers, on the other hand, provided advice and assistance in relation to loan products of many different loan providers. This difference is particularly significant in relation to the second and third questions.

146    Accepting that loan providers, such as ANZ, provided some or all of the advice or assistance said to comprise loan arrangement services, the second question is whether those services were relevantly provided in a market for the supply of such services? The fact that ANZ provided some advice or assistance in relation to its loan products does not mean that it supplied these services in a market for the provision of such services.

147    The critical issue here was, and is, whether the advice or assistance provided by loan providers like ANZ was really no more than conduct ancillary to the sale or distribution of the loan providers’ loan products.

148    Of course, one does not actually sell or distribute a loan. However, the process of having a customer apply for and obtain a loan from ANZ can conveniently, if somewhat colloquially, be described as selling or distributing a loan product. That was essentially how the industry witnesses and ANZ’s internal documents described the process. It could, perhaps more accurately, be called loan origination, loan referral, or loan introduction.

149    In any event, as earlier indicated, his Honour concluded, in effect, that ANZ’s (and it would seem, other loan providers’) provision of advice and assistance to customers, or prospective customers, was conduct that was merely ancillary to, or an adjunct to, the “sale” or distribution of a loan. It did not amount to the supply of a service to customers in a separate and distinct market for the supply of loan arrangement services. Rather, the advice and assistance was provided in the market for loans or loan products.

150    The second question involves a process of characterisation of the advice and assistance alleged to comprise loan arrangement services. The proper characterisation of the relevant product (goods or services) can be critical to the identification and definition of the correct market. The characterisation must be based on, amongst other things, a careful evaluation of the evidence concerning the nature and character of the relevant goods or services, the circumstances in which they were supplied or acquired, and the interactions between, and the perceptions of, the relevant suppliers and acquirers of the product.

151    The question whether it was open on the evidence for the primary judge to reject the ACCC’s characterisation of the advice and assistance provided by bank branches, and to instead characterise it as merely ancillary to the sale or distribution of loan products, is considered in more detail later in the context of the ACCC’s specific appeal grounds. It is sufficient to observe here that it does, in the circumstances, appear to be somewhat contrived and artificial to characterise the provision of advice and assistance by bank officers in relation to loan products as the provision of services in a market separate and distinct from the market for the supply of the loan products themselves.

152    The artificiality of the ACCC’s characterisation can be illustrated by way of an analogy. Whilst by no means a perfect analogy, the characterisation of a simple transaction involving the purchase of a motor vehicle may nevertheless provide some analogous insight into the task involved in this matter.

153    If a prospective customer goes to a car dealer for the purpose of acquiring a car, the likelihood is that the customer will seek and obtain advice from a sales assistant about what particular vehicles are available, which vehicle will suit their needs, and various pricing and perhaps finance options. The sales assistant would also no doubt be in a position to assist the customer to fill out and complete the appropriate contractual documents and other paperwork.

154    If the customer purchases a car, does it follow that the dealership has both supplied to the customer both a product (a car) in the market for motor vehicles and also supplied services in a market for the supply of motor vehicle advice services?

155    At first blush, the ordinary characterisation of this interaction would appear to be that there is a single supply of a vehicle in the motor vehicle supply market. All the more so where, if the customer walked out without buying a car, the dealership would receive nothing, whatever advice and assistance may have been provided. Likewise, if a customer walks in off the street and buys a car without seeking or obtaining any advice or assistance, the dealership would most likely sell the car at the same price it would have sold it at if substantial advice and assistance had been provided.

156    This conclusion would appear to follow even if the dealership organised itself internally so that there was a separate division (or economic unit) that obtained and made the vehicles available for sale (the wholesale division) and a separate division that managed the sales staff who interacted with prospective customers (the retail division). It may be, in these circumstances, that the sales person could be considered to have provided a sales service to the wholesale division. The sales or retail division might even earn a commission from the wholesale division when a successful sale occurs. But it does not necessarily follow that the sales person also provided a separate service (an advisory or assistance service) to the customer in a separate market for the supply of such services.

157    Here, like the car dealership example, the apparent artificiality of the splitting up or disaggregation of the relevant interaction between ANZ branch staff and a prospective customer can be demonstrated by reference to the consideration said to be payable in respect of the alleged provision of loan arrangement services by an ANZ branch. If advice or assistance was provided to a prospective customer, but the customer did not apply for a loan, the customer was not liable to pay anything, irrespective of the level of advice or assistance provided and received. Likewise, a customer who applied for a loan without seeking any advice or assistance paid the same loan application fee as a customer who obtained extensive advice or assistance.

158    The only consideration that could be said to relate to the advice or assistance provided by the branch officer was a notional internal accounting for commission in favour of the ANZ branch when a customer successfully applied for a loan through the branch. But that notional commission was more likely referable to the services provided by the branch to ANZ Mortgage Group, not the services provided to the customer. If anything, it was akin to the commission ANZ paid to an independent broker for services the broker provided to ANZ as a result of successfully referring a customer. It was not remuneration for the provision, by the broker, of services to the customer. Rather, it was remuneration for the services provided to ANZ for successfully selling a loan.

159    Even accepting that the provision of advice or assistance in respect of a loan by an ANZ branch could properly be characterised as the provision of a service in a market for loan arrangement services, there is a third question. That question is whether the service was provided in competition with Mortgage Refunds (and others) in that market. The critical issue here is one of substitutability, in a real-world commercial sense.

160    As has been said earlier in these reasons, the primary judge found, in relation to that question, that there were significant and material differences between the services (if that is what they were) provided by ANZ branches and franchisees, and other loan providers, and the services supplied by Mortgage Refunds, and other mortgage brokers. The main difference was that ANZ only provided advice and assistance in respect of its own products. Likewise with the other loan providers. Mortgage Refunds and other brokers, however, provided advice and assistance in relation to many different loan products from many different loan providers. Indeed, the ability to provide comparative advice was seen by both the brokers and the customers alike as the major drawcard for brokers. This led the primary judge to conclude that the services provided by the bank branches, including ANZ branches, were not substitutable for the services provided by brokers. This in turn led to the conclusion that ANZ and Mortgage Refunds did not compete in the same market for services.

161    Again, the question whether this finding was open to his Honour on the evidence is considered in more detail later in these reasons. For present purposes, it is sufficient to point out another aspect of the apparent artificiality inherent in the ACCC’s market definition that arises at this stage of the inquiry.

162    To deal with the fact that mortgage brokers, unlike bank branches, were able to, and did, provide advice and assistance in relation to loan products from multiple loan providers, the ACCC alleged that the relevant market encompassed advice and assistance in relation to loan products generally. But that created another difficulty or element of artificiality. On the ACCC’s broad market definition (an Australia-wide market for the supply of loan arrangement services to members of the public by loan providers, franchisees and brokers), an ANZ branch, which only supplied services in relation to ANZ loan products, was in competition with branches and franchisees of other banks that only provided services in relation to their own loan products. Thus, an ANZ branch competed with a Westpac or Commonwealth Bank branch in relation to loan arrangement services, even though none of them could offer services in respect of the same loan products. One can readily see how branches of different banks could be seen to be competing in the market for the supply of loan products. It is more difficult to see how it could realistically be said that they competed in relation to loan arrangement services. A consumer who wants advice and assistance about an ANZ loan product is unlikely to, and effectively unable to, go to a Westpac bank branch to obtain such advice and assistance.

163    The artificiality of the ACCC’s characterisation here can be illustrated by returning to the motor vehicle dealership analogy. If the hypothetical motor vehicle dealership was a Toyota dealership, and there is a rival Ford dealership across the road, is it realistic to suppose that the Toyota salesperson (on behalf of his or her dealership) competes with his or her Ford counterpart/dealership in a market for the supply of motor vehicle advisory services, even when they cannot provide advice or assistance in relation to the same cars? Or are they simply competing in the market for the supply of motor vehicles?

164    The ACCC sought to meet this by pleading, in the alternative, that ANZ was in competition with Mortgage Refunds in relation to the supply of loan arrangement services in respect of ANZ loan products (see FASC at paragraph 21). There are two difficulties with this alternative pleading. First, the ACCC did not plead that there was a separate market for the supply of loan arrangement services in respect of ANZ loan products. Second, even if it did, this would perhaps exacerbate the apparent artificiality of the market definition rather than resolve it. If there was a separate market for the supply of loan arrangement services in respect of ANZ loan products, it would seem to follow that there would be separate markets for the supply of loan arrangement services in respect of Westpac loan products, Commonwealth Bank loan products, and so on. If this were the case, Mortgage Refunds would compete in relation to loan arrangement services with branches of banks, but in many different loan arrangement services markets. That would appear to be both unrealistic and unlikely. The existence of a series of loan arrangement service markets in respect of different loan products was not supported by the evidence, including Dr FitzGerald’s evidence.

Challenges to factual findings on appeal

165    An appeal to this Court is an appeal by way of rehearing: Cabal v United Mexican States (2001) 108 FCR 311 at [222] referring to Minister for Immigration and Multicultural Affairs v Jia (2001) 205 CLR 507 at [75]. The Court can have regard to the evidence given in the proceedings out of which the appeal arose and has power to draw inferences of fact: Federal Court of Australia Act 1976 (Cth), s 27. Notwithstanding this, the task of a court on an appeal by way of rehearing is the correction of error: Branir Pty Ltd v Owston Nominees (No 2) Pty Ltd (2001) 117 FCR 424 at [22] and the cases there cited.

166    The relevant principles were summarised in Cabal at [223]-[224] as follows:

The principles which govern the review by a Full Court of a primary judge’s findings of fact are as stated in Warren v Coombes (1979) 142 CLR 531. See also Abalos v Australian Postal Commission (1990) 171 CLR 167; Devries v Australian National Railways Commission (1993) 177 CLR 472; and State Rail Authority of New South Wales v Earthline Constructions Pty Ltd (In liq) (1999) 73 ALJR 306; 160 ALR 588. In general on an appeal by way of rehearing from a judge sitting without a jury an appellate court is in as good a position as the trial judge to decide on the proper inference to be drawn from facts which are undisputed or which, having been disputed, are established by the findings of the trial judge. In deciding what is the proper inference to be drawn, the appellate court will give respect and weight to the conclusion of the trial judge. However, once having reached its own conclusion it will not shrink from giving effect to it.

Notwithstanding the fact that the learned primary judge’s review was conducted on the papers, and without any opportunity to consider the demeanour of the witnesses, the weight to be accorded to the evidence of the experts was primarily a matter for his Honour to determine. This Court can consider whether he fell into appealable error in that regard but it will not approach that evidence as though this were a rehearing de novo in which his Honour’s views count for nought. If, after giving full weight to his Honour’s views, we are persuaded that the conclusions which he reached were erroneous we must set aside his findings of fact. We cannot however simply substitute for his Honour’s findings of fact those findings which we would have made had we been the judges on review who determined this matter at first instance.

See also Branir at [23]-[32].

167    When considering findings of fact, an appeal court must give due regard to the advantages of trial judges. That is, of course, clearest where the findings of the trial judge might have been based on their impression or assessment of the demeanour and credibility of witnesses. That is not a major consideration in this matter. With some relatively minor qualifications or exceptions, the primary judge accepted all of the witnesses as being honest and reliable.

168    However, the advantages of a trial judge are not limited to assessing witness demeanour and credibility. Where inferences or conclusions drawn by the trial judge involve questions of degree, or value judgments or impressions about matters about which there is room for some legitimate difference, there may be some advantages flowing from the fact that the trial judge heard the evidence in its entirety as it unfolded at the hearing: Branir at [24]. Market definition, as the product of detailed and considerable evidence, falls into this category. That is not to say that any such advantages a trial judge may have are unassailable. Rather, it simply serves to demonstrate that error must be demonstrated.

169    In Branir, the proper approach to be taken by an appeal court was put in the following terms at [30]:

The proper approach is not to ask the court to survey all the evidence, directed by the otherwise unassailable findings on credit, and to ask it to arrive at its own conclusions, without “essaying the necessary task of positively demonstrating that the trial judge was wrong”: Williams at [61] per Heydon JA. It is not appropriate to treat the appeal as though it were a new trial on the evidence and constrained merely by the unassailable factual findings. Error must be demonstrated. See also Biogen Inc v Medeva plc at 45 and Williams at [136] and [137] citing Zuvela and Biogen. The views and conclusions of the trial judge ultimately have to be shown to be wrong. They should not be laid to one side and a simple re-argument of the case take place.

170    Whilst the ACCC’s appeal grounds are directed at specific errors said to have been made by the trial judge, there is, to a degree, some merit in ANZ’s submission that many of the appeal grounds could be said to amount to no more than a reargument of its case at trial.

Ground 1 - Did the primary judge err in finding that ANZ did not supply loan arrangement services?

171    The ACCC challenged the conclusion of the primary judge at [631] that he was not satisfied that ANZ branches supplied loan arrangement services to potential borrowers. Rather, his Honour concluded that ANZ supplied sales services to ANZ Mortgage Group and/or ANZ.

172    This finding or conclusion must be read and considered in the context of other findings made by the primary judge and the reasons as a whole. It was not a finding that branches did not from time to time provide some or all of the types of advice or assistance referred to in paragraph 7.2 of the ACCC’s pleading. There was essentially no dispute that, in responding to enquiries from persons seeking to apply for an ANZ home loan product, ANZ employees from time to time advised enquirers as to the features of the respective loan products available from ANZ; advised enquirers which of ANZ’s loan products were available to persons in the enquirer’s circumstances; advised the enquirer which of ANZ’s loan products were best suited to the enquirers needs; and assisted the enquirer to complete and lodge an application for the loan product selected in a manner that met the requirements of ANZ’s Mortgage Division: see ANZ’s answers to the ACCC’s interrogatories at paragraphs 11.1 to 11.4.

173    His Honour’s finding at [631]-[632] concerned the proper characterisation of these activities. His Honour characterised these activities performed by ANZ branch employees as activities done for or on behalf of ANZ Mortgage Group or ANZ as part of the process of “selling” (that is, having a customer sign up for) an ANZ mortgage loan. They were not properly characterised, as the ACCC contended, as also involving the separate provision of services to customers in the Australia-wide market for loan arrangement services.

174    The ACCC submitted that this finding was in error because it was contrary to the expert, lay and documentary evidence. It also attacked aspects of the primary judge’s reasoning that led to this conclusion and submitted, in essence, that the reasoning disclosed error.

175    We are not persuaded that his Honour erred in so characterising the activities carried out by ANZ branch employees. This characterisation was open on the whole of the evidence. It was an inference or conclusion that was reasonably available, and on our reading of the evidence correct. The inference or conclusion arrived at by the primary judge was compelling, having regard to the evidence as a whole.

176    Nor are we satisfied that any of his Honour’s reasoning exposes any material error in arriving at that finding.

177    The first point to make concerning the ACCC’s contentions in relation to this ground is that it is not correct to say that the primary judge’s conclusion concerning the proper characterisation of the activities engaged in by ANZ branches was contrary to the evidence. Putting aside the answers to the interrogatories, there was virtually no direct evidence of what the ANZ branches or their officers actually did when it came to dealing with prospective borrowers. The evidence, such as it was, was limited to very general evidence about what bank branches generally did, or perceptions or conclusions of various people about what bank branches did. It was perhaps touched on by some of the ANZ internal documents, but again those documents spoke in very general terms about the role of the branches and other internal distribution channels. None of the evidence, when closely analysed, compelled or compels conclusions contrary to those arrived at by the primary judge.

178    The problem for the ACCC is that the primary judge considered such evidence as there was that might have supported the ACCC’s characterisation, but ultimately either rejected it, or gave limited weight to it. We are not satisfied that his Honour’s findings in relation to the evidence involved any error.

179    The evidence relied on by the ACCC has been summarised at length earlier in this judgment. It is unnecessary to repeat that summary here. The question is whether, as the ACCC contended, the evidence compelled a characterisation contrary to the one arrived at by the primary judge.

180    The so-called lay evidence referred to by the ACCC in the context of this ground of appeal is the evidence of the Mortgage Refunds customers (Messrs O’Malia and Black) and the “industry witnesses” (Messrs King, Stark, Lahiff and Percy).

181    Neither Mr O’Malia nor Mr Black gave any evidence concerning exactly what ANZ branch officers did or did not do. Their evidence primarily related to the services provided by Mr King when he was at Mortgage Refunds. Those services were described in very general terms as including Mr King discussing and advising them on the available loan options and assisting with the completion of the paperwork. Both witnesses also gave very general evidence that the services provided by Mr King when he was at Mortgage Refunds were “essentially” or “fundamentally” the same as the services he provided earlier when he was with either Suncorp or HSBC.

182    The primary judge ultimately gave little weight to this evidence. This finding is considered in detail later in these reasons in the context of appeal ground 4. Suffice it to say here that it was, at the very least, open to his Honour to give little weight to this evidence. In the context of this ground, it could hardly be concluded that the evidence of Mr O’Malia and Mr Black compelled a different conclusion concerning the characterisation of the activities of ANZ branches. It certainly could not be said that his Honour’s characterisation was contrary to this evidence.

183    Much the same could be said about the evidence of the industry witnesses. Messrs Lahiff, King and Stark had worked at banks before becoming brokers. They also gave evidence effectively equating the services provided by bank branches with the services provided by brokers. That evidence was also in fairly general terms.

184    None of their evidence directly concerned the activities of ANZ branches. Mr Lahiff’s evidence (at [68] of his affidavit) was that Westpac mortgage specialist staff provided the following services to customers in respect of the distribution of Westpac’s mortgage products:

… The services that the Westpac mortgage specialist staff provided to the customer were essentially the same as those services provided by our Mortgage Choice representatives … in that they would:

68.1.    meet with a potential borrower;

68.2.    discuss and record their circumstances, needs and preferences;

68.3.    assess that information to identify the suitable mortgage products;

68.4.    discuss those options with the potential borrower and give advice on the relative merits of each product;

68.5.    if and when the potential borrower wished to apply for a loan, assist them in completing the application and compiling and submitting the necessary documentation to the Westpac mortgages division;

68.6.    liaise between the mortgages division and the applicant if further information was required;

68.7.    inform the borrower if their application was successful;

68.8.    maintain the relationship with the customer, including pursuing opportunities for the sale of further Westpac products; and

68.9.    by at least the early 1990s, travel to customers’ homes if required and offer their services out of office hours.

185    Mr King’s evidence about the services he provided when he was employed at Westpac was in similar terms. He said (at [76] of his affidavit):

The services I provided were all-but identical to those I offered through Mortgage Refunds. For interested customers, I would give information and advice about the Westpac loan products, do a preliminary approval to make sure they were at least potentially qualified for a loan, help them complete the forms and compile the necessary documentation (identification, income verification, etc), submit it to the central loans approval area, liaise between the approvals section and the customer to keep them advise[d] of progress, and then let them know when approval had been given. For a successful application, I would also keep in contact with the customer to make sure they didn’t refinance elsewhere.

186    Relevantly, Mr King also appeared to characterise his activities as a Westpac Home Finance Manager as involving selling Westpac home loan products. In that context he was authorised to offer a prospective borrower a “special deal” through discounted fees and rates. Far from being contrary to the primary judge’s characterisation, this aspect of Mr King’s evidence appeared to support it.

187    Mr Stark did not directly give evidence about the services he provided when he was at Suncorp. His evidence was that the work he did at Mortgage Refunds included “filling in all the requisite application paper work, meeting with clients, advising on the best available mortgage product for each individual situation, collecting all personal details of clients and lodging applications and other documents to the lender” (at [17] of his affidavit). This was “almost identical” to the work he did at Suncorp, the main difference being that when at Suncorp he was only able to offer Suncorp products.

188    As already indicated, each of Messrs Lahiff, King and Stark also gave evidence to the effect that they considered brokers to be in competition with bank branches in relation to the distribution of loan products. This aspect of their evidence has been considered earlier in these reasons.

189    His Honour also gave limited weight to the evidence of these witnesses insofar as they equated the services provided by brokers and bank branches and suggested that there was competition between brokers and bank branches. His Honour’s reasons for so doing are again considered in specifically the context of ground 4. In the context of this ground, however, it can again be said that the evidence of these witnesses did not and does not compel a different characterisation of the activities of ANZ branches to that arrived at by the primary judge. Nor can it be concluded that this evidence was, as contended by the ACCC, contrary to the primary judge’s characterisation.

190    The mere fact that the services provided by the bank’s branches were, on one level, similar to those provided by brokers, does not mean that they necessarily bore the same character. His Honour found, on the evidence, that there were at least two important differences between the services provided by bank branches and brokers that suggested that they were of a different character. The first difference was that brokers were seen to be independent of particular loan providers, including banks. A second and related difference was that brokers could and did give comparative advice concerning many different loan products offered by many different loan providers, including banks. Having regard to those important differences, his Honour had “no difficulty in characterising the brokers’ function as involving the supply of services to potential borrowers”: see [636].

191    The primary judge found, however, that once a broker’s customer had decided to apply for a particular loan, in preparing and submitting the loan application to a particular loan provider, the broker was also performing or providing a service to that particular loan provider in accordance with their accreditation: see [636]-[637].

192    As explained further in the context of ground 4 later in these reasons, the ACCC has not demonstrated error on the part of the primary judge in either finding that there were these differences between the services provided by branches and brokers, or arriving at a different characterisation that in part flowed from these differences. The character of independence of a broker in acting for, and in the interests of, the customer is of such a central importance to the commercial reality of his or her function as to be apt to be a distinguishing, indeed defining, characteristic of his or her role. This is especially so if the comparator for analysis is an employee whose duty is to act in the interests of his or her employer in selling products to the customer.

193    The expert evidence of Dr FitzGerald is dealt with specifically in the context of ground 3 later in these reasons. For present purposes it is sufficient to note that his Honour was not bound to accept Dr FitzGerald’s opinions insofar as they touched on the proper characterisation of the activities of bank branches. Nor does Dr FitzGerald’s opinion necessarily compel a characterisation of the activities of ANZ branches different to that arrived at by his Honour.

194    Importantly, Dr FitzGerald does not directly address or express an opinion in relation to the proper characterisation of the advice and assistance activities carried out by ANZ branches. Whilst it is clearly implicit in his analysis of questions concerning market and competition that he would characterise the relevant services provided by ANZ branches in the same way as he characterised the services provided by brokers, that characterisation depended on two critical assumptions made, or factual inferences drawn, by Dr FitzGerald. Those two assumptions or inferences were: first, that the services provided by branches were the same or substantially the same as those provided by brokers; and second that because the branches and ANZ Mortgage Group could be seen as separate business or economic units within ANZ, the branches could be seen as intermediaries (between ANZ Mortgage Group and the customer) in exactly the same way as could independent mortgage brokers.

195    The importance of the first inference or assumption to Dr FitzGerald’s implicit view concerning the characterisation of the activities of ANZ branches can be seen in his answer to questions relating to whether loan providers competed with brokers in relation to loan arrangement services. Dr FitzGerald saw the answer to this question as turning exclusively on whether the services provided by brokers and other intermediaries “are in the same market if instead offered ‘under the lender’s roof’ by its own employees or equivalent agents”: see [51] of Dr FitzGerald’s first report. He expressed the opinion that “the services in question are the same or close substitutes whether they are provided by bank staff … or by an untied broker operating under a different banner”: see [53] of Dr FitzGerald’s first report.

196    Yet, as has already been outlined, the primary judge rejected the assumption or inference that the services provided by an ANZ branch officer were the same as, or close substitutes for, the services provided by brokers. His Honour found that the assumption or inference was not supported by the evidence. His Honour drew his own inferences or conclusions from the evidence in relation to this essentially factual question. As detailed below in the context of ground 3, it was open to the primary judge to do so. This finding essentially undermined whatever support Dr FitzGerald provided to the ACCC’s case in relation to the proper characterisation of the activities or services provided by ANZ branches.

197    As for the second assumption or inference, whilst the primary judge accepted that in some circumstances it may be appropriate to treat separate business units in a corporation as distinct economic entities, it is readily apparent that his Honour considered that, in the particular circumstances of this matter, the fact that ANZ Mortgage Group and Personal Banking Business unit was of limited, if any, relevance in characterising the services provided by branches to prospective borrowers: see [262]-[263], [381], [520], [632] and [649]. It might follow that, for internal ANZ purposes, the branches might be considered to be providing services to ANZ Mortgage Group. His Honour accepted that this might be so. It did not follow, however, that the activities of the branches should also be characterised as the provision of loan arrangement services to customers.

198    The final category of evidence that the ACCC relied on as being contrary to the primary judge’s characterisation was the evidence of statements in ANZ’s internal documents. The ACCC relied on particular passages from those documents which, in its submission, showed that some officers of ANZ perceived or were of the opinion that the ANZ’s internal distribution channels competed with brokers. On the ACCC’s case, this showed that the services provided by branches should be characterised in the same way as those provided by brokers, or at least that this was how ANZ officers perceived or viewed the relevant activities engaged in by the branches.

199    Again, however, the difficulty for the ACCC is that the primary judge carefully analysed the ANZ documents and drew different inferences or conclusions from those put forward by the ACCC. In particular, and relevant to the question of the characterisation of the conduct of the ANZ branches, the primary judge found that where the documents referred to competition, when read in context, those references related to competition in the lending market, not competition between ANZ and brokers in any market for the supply of loan arrangement services. Whilst the documents showed that ANZ wanted to increase its market share in the lending market and reduce its reliance on brokers as an external channel for the distribution of its loan products, it did not follow that ANZ saw its branches as competing with brokers for the supply of loan arrangement services to customers.

200    The primary judge’s findings concerning the ANZ internal documents is the subject of appeal ground 5. Suffice it to say for the purposes of this ground that the primary judge’s findings concerning the ANZ internal documents are consistent with his characterisation of the advice and assistance given by ANZ branches to prospective borrowers. The documents, in his Honour’s view, were consistent with the proposition that the relevant market in which the branches interacted with customers was the market for the supply of loan products. The documents did not support the proposition that there was a separate market for the supply of loan arrangement services.

201    The ACCC points to three specific findings by the primary judge that it submits were contrary to the evidence. The first is his Honour’s finding (at [625]) that the functions performed by ANZ staff should be characterised as pre-contractual negotiations. That is perhaps a mischaracterisation or overstatement of his Honour’s reasoning in [625]. His Honour did not characterise all of the activities of branch officers as pre-contractual negotiations. After characterising the activities of ANZ staff as “resemble[ing] an interactive procedure between the potential borrower and a representative of the lender, anterior to the making of a loan”, his Honour observed that “[t]o characterise it as the provision of services would lead to the similar characterisation of any pre-contractual negotiations.” Whilst it may not be strictly correct to characterise all of what a branch officer does as “pre-contractual negotiations”, we are unable to find any material error in this minor aspect of his Honour’s reasoning.

202    The second specific finding challenged by the ACCC is the primary judge’s finding (at [631]) that the branches only supplied sales services to ANZ Mortgage Group and/or ANZ. That finding, which is part of his Honour’s broader characterisation of the services provided by the branches, has already been addressed in these reasons.

203    Finally, the ACCC contended that aspects of his Honour’s reasoning in [632] are contrary to the evidence. To consider this submission, it is necessary to set out that paragraph in its entirety. The particular finding challenged by the ACCC is emphasised in bold.

Even if branch staff supplied some sort of service to potential borrowers, I consider that they did so on behalf of ANZ Mortgage Group and/or ANZ. I have pointed out that a potential borrower would go to a branch, expecting to deal with an ANZ staff member for the purpose of obtaining a loan product. As between the branches and ANZ Mortgage Group, the latter had arrangements with the former pursuant to which branch staff would deal with the potential borrower, identifying his or her financial capacity and needs, selecting a product and preparing and submitting a loan application. ANZ Mortgage Group paid the branches for the performance of that function. Obviously, the branch staff were acting on behalf of ANZ Mortgage Group and/or ANZ. The potential borrower would consider everything associated with his or her visit to the branch as being part of the larger transaction which he or she had in mind, which transaction was with ANZ. In those circumstances, there can be no justification for splitting the transaction into parts – loan product and loan arrangement services; or distinguishing between ANZ Mortgage Group and the branches as separate suppliers of those parts. The better view is that if branch staff provided any services to a potential borrower, they were provided by ANZ or ANZ Mortgage Group through the branches.

(Emphasis added.)

204    Whilst it could perhaps be said that there was no direct evidence about the beliefs or perceptions of a potential borrower in relation to his or her interaction with an ANZ branch officer, read fairly and in context this part of his Honour’s reasoning amounts to no more than an inference drawn from the evidence, such as it was, concerning the nature of the interaction between a branch officer and a potential customer. It cannot be said to be contrary to the evidence. Further, whilst the views or perceptions of the customer may not be particularly significant in properly characterising the services provided in a competition law context, this was but one small part of his Honour’s reasoning. It does not bespeak error.

205    It follows that the primary judge’s characterisation of the services provided by the ANZ branches was not, as the ACCC contends, contrary to the expert, lay and documentary evidence having regard to the inferences and conclusions his Honour drew from that evidence.

206    Nor can it be concluded that his Honour’s characterisation was contrary to the evidence considered as a whole. The findings and conclusions of the primary judge, both in relation to individual items or categories of evidence, and in relation to the evidence as a whole, involved elements of fact, degree, opinion or judgment. His Honour also had the advantage of seeing and hearing the evidence in its entirety as it unfolded at the hearing. There is no proper appellate warrant for overturning his Honour’s factual findings and conclusions concerning the characterisation of the relevant services.

207    The conclusion that the primary judge did not err in rejecting the ACCC’s characterisation of the activities of the ANZ branches is perhaps sufficient to dispose of the appeal. Once it is accepted that the ANZ branches did not supply loan arrangement services to prospective customers, but rather performed a sales or distribution service on behalf of ANZ Mortgage Group or ANZ in the market for the provision of loan products, it must follow that ANZ did not relevantly compete with independent mortgage brokers, including Mortgage Refunds. The ACCC’s case, and therefore its appeal, must accordingly fail.

208    In deference to the detailed written and oral submissions of the parties, however, some attention will be given to the remaining grounds. The extent of the consideration of the grounds will depend on the relative importance of the challenged findings to the ultimate findings or conclusions of the primary judge and the extent to which the ACCC’s written and oral submissions directly addressed the particular ground.

Ground 2 - Did the primary judge err in failing to find that Mortgage Refunds provided loan arrangement services?

209    This ground can be dealt with briefly. The relevant “non-finding” was not critical, or even important, to the primary judge’s dismissal of the ACCC’s case. The ACCC did not separately address this ground in its oral submissions.

210    The ACCC’s case was that ANZ was in competition with Mortgage Refunds or “engaged persons” to supply loan arrangement services, or loan arrangement services in respect of ANZ loan products: see FASC at paragraphs 20 to 21. Engaged persons were “persons who were employees or agents of Mortgage Refunds and who were engaged by it to provide loan arrangement services on its behalf”: see FASC at paragraph 16.4.2. To rely on s 45A of the Act, as it sought to do, the ACCC was required to prove that either Mortgage Refunds (as a party to the agreement with ANZ), or its employees or agents, supplied the relevant loan arrangement services.

211    The primary judge found that the individual brokers who operated under the Mortgage Refunds “banner” supplied broking services, including advice, information and assistance to potential borrowers. Accepting, for present purposes, that those services were akin to the ACCC’s pleaded “loan arrangement services”, the question was whether the brokers were employees or agents of Mortgage Refunds. If they were not, it could not be concluded that Mortgage Refunds provided the services. The ACCC bore the burden of proving that they did.

212    The primary judge effectively found that the ACCC failed to discharge its burden of proof in relation to this aspect of the matter.

213    At the beginning of his judgment, the primary judge summarised the evidence in the following terms (at [14]):

As far as the evidence goes, neither AFG nor Mortgage Refunds was directly involved in the submission, by brokers to ANZ, of applications for finance. The brokers dealt directly with ANZ. AFG may have carried out some quality assurance role. Mortgage Refunds acted as a provider of support services to individual brokers. The evidence suggests that they were self-employed, although ACCC suggests otherwise.

214    His Honour returned to this issue towards the end of his judgment under the subheading “Some Other Matters”, after his Honour had already concluded that the proceedings must be dismissed. At [653], his Honour said:

Whether Mortgage Refunds supplied broking or similar services to potential borrowers is also unclear. AFG had a contract with Mortgage Refunds pursuant to which the latter was to “introduce” applicants to AFG, in consideration of which it was to receive a percentage of any commission received by AFG. Both AFG and Mortgage Refunds were content to permit the Mortgage Refunds brokers to deal directly with ANZ in introducing potential borrowers, whilst keeping control over their commissions. Mortgage Refunds, itself was not accredited, at least in respect of ANZ loan products. However Mortgage Refunds advertised the availability of services which were, in practice, provided by the brokers. It provided other support services to the brokers, including office space and a “shop front”. From ANZ’s point of view, AFG and/or the brokers were distributing its products. From AFG’s point of view, Mortgage Refunds was distributing ANZ’s products, using its brokers who were AFG’s authorized officers for the purposes of its originator agreement with ANZ.

215    His Honour also considered that the relationship between Mortgage Refunds and the individual brokers that operate under its banner was unclear: see [654]. There was evidence that suggested that some of the brokers operated as independent contractors, evidence that suggested that Mortgage Refunds had no employees, and evidence that suggested that, for a period, some brokers became franchisees. His Honour’s conclusion was expressed in the following terms (at [656]):

I find it difficult to generalize about Mortgage Refunds’ involvement in the provision of broking services. Some borrowers may have had contractual relationships with it, whilst others may not have had such relationships, depending upon the circumstances in which they approached Mortgage Refunds or the relevant brokers. I doubt whether, on the evidence, it is possible to form a firm view as to whether Mortgage Refunds was supplying broking services. The answer may depend upon the circumstances of each case.

216    The ACCC, in its written submissions, submitted, in effect, that there was “enough” evidence to satisfy the primary judge that the services were provided by Mortgage Refunds. It pointed to some very general assertions in Mr King’s evidence about who provided the services; the fact that under the agreement between AFG and Mortgage Refunds it was Mortgage Refunds that was to receive payment in respect of successful loan applications submitted through accredited brokers; the fact that customers entered into contracts with Mortgage Refunds in relation to the payment of a refund; and the fact that many customers would contact Mortgage Refunds in the first instance before being referred to one of the accredited brokers.

217    A number of the matters relied on by the ACCC were specifically adverted to by the primary judge in his assessment of the evidence. The evidence referred to by the ACCC in its submissions was not, however, the entirety of the evidence in relation to this issue. Nor, as ANZ submitted, is it a fair summary of the evidence. In particular, it ignored much of the evidence given by the Mortgage Refunds witnesses, Messrs King and Stark, about the relationship between Mortgage Refunds and the brokers that operated under that banner.

218    In the circumstances, it is not necessary to set out, in any great detail, all of the evidence. It is sufficient to say that the evidence, considered as a whole, did not compel a conclusion that Mortgage Refunds provided the broking services, as opposed to the individual brokers operating loosely under the Mortgage Refunds banner. Nor did it compel a conclusion that the brokers were all employees or agents of Mortgage Refunds. At the very least, as his Honour found, the position was unclear and may have depended on the particular circumstances of the particular brokers at particular times. It follows that it was open to his Honour to find, in effect, that the ACCC had failed to discharge its burden of proving this aspect of its case.

219    The ACCC has accordingly not made out the alleged error the subject of ground 2.

Ground 3 - Did the primary judge fail to take into account or give appropriate weight to the evidence of Dr FitzGerald?

220    The ACCC submitted that the primary judge rejected, or failed to have regard to, or failed to give appropriate weight to, the economic expert evidence of Dr FitzGerald that ANZ, through its branches, employees, agents and franchisees, and accredited brokers, provided loan arrangement services to customers seeking residential property loans in competition with each other.

221    The ACCC relied, in this context, on the fact that there was no expert evidence contrary to Dr FitzGerald’s evidence. The ACCC also attacked the primary judge’s reasons for not accepting the critical parts of Dr FitzGerald’s evidence. In particular, the ACCC submitted that his Honour’s rejection of Dr FitzGerald’s evidence was based on “propositions” not raised with Dr FitzGerald or not properly analysed by the primary judge.

222    Before addressing these specific challenges to his Honour’s treatment of the evidence of Dr FitzGerald, it is appropriate to make some brief observations concerning the proper way to approach expert evidence of the sort given by Dr FitzGerald; that is, evidence in relation to markets, competition and, in that context, the characterisation of the conduct of persons and entities in a commercial or economic context.

223    Questions relating to the identification and definition of a market or markets, the competition that is said to occur in that market or markets, the characterisation of the conduct of persons said to be participants in that market or markets, and the substitutability of goods or services, are all ultimately questions of fact. There is, however, a proper role or scope for expert evidence in relation to these issues.

224    In Australia Meat Holdings Pty Ltd, Wilcox J explained the role of expert economic evidence in relatively narrow terms. His Honour said (at 316):

… I deprecate the course, taken by AMH, of supplying to economists proofs of the evidence to be given by other witnesses and then eliciting from those economists opinions as to the proper conclusion upon the definition of the market. Economists are able to assist the court in relation to economic principles. But, once the relevant principles are expounded, their application to the facts of the case is a matter for the court. The proper definition of a market is entirely a matter of fact, the determination of which ought not to be made more protracted and expensive by the adduction of unnecessary expert evidence.

225    In Australian Competition and Consumer Commission v Liquorland (Australia) Pty Ltd [2005] FCA 630 at [21]-[36], Allsop J explained the permissible scope or role of expert economic evidence in this context in slightly more expansive terms. Having observed that economics is a social science that involves, amongst other things, the assessment of human behaviour, his Honour said, of the role of an economist (at [29]):

Because it [economics] is a social science and because in one sense and in part it truly is a way of approaching matters and a way of thinking about matters, there is a role, it seems to me, for the economist to assist the court by expressing, in his or her own words, what the human underlying facts reveal to him or her as an economist and what it reflects to him or her about underlying economic theory and its application.

226    The economist may also have a role in ordering, categorising and explaining the primary assumed facts if that is based on the use of the economist’s specialised technique of thinking. In Liquorland [2005] FCA 630, Allsop J said that the economist may (at [32]):

… through his or her skill and background and expertise, be able to identify critical aspects of the underlying material as important to economic theory and consumer behaviour and economic behaviour, illustrating or illuminating the application of simple, or perhaps complex, economic theory. From that, conclusions can be drawn, through a reasoning process, which may be very short, as to what this economist thinks is the appropriate way to order, prioritise, synthesise and explain that underlying material.

227    There are, however, limits to the role in which an economist can properly or usefully play in relation to the primary or underlying facts. It is not for the economist to draw inferences or express conclusions about the primary facts if those inferences or conclusions are not based on his or her skill, background and expertise. Such, “retailing” of the primary evidence (an expression used by Allsop J in Liquorland [2005] FCA 630) is at best of little or no assistance to the Court and, at worst, is inadmissible.

228    It is also ultimately necessary to be able to identify and distinguish the primary facts and assumptions (however they may be categorised, ordered or prioritised by the economist) from the economist’s opinion. That is because, as with all expert opinion evidence, the relevance and weight to be given to the opinion evidence will depend, to a large extent, on the correctness or otherwise of the assumed or proved facts. And, as with all expert evidence, the judge, as tribunal of fact, is not bound by the expert evidence, even if no opposing expert opinion evidence is adduced. The weight to be given to the economist’s expert opinion is a matter for the judge and will depend, to a large extent, on whether the opinion is based on facts or assumptions that are proved to be accurate or correct.

229    The primary judge gave detailed consideration to Dr FitzGerald’s evidence. It could not sensibly be contended that his Honour “failed to take into account” Dr FitzGerald’s evidence, as the ACCC contended in ground 3, if by that expression the ACCC meant that his Honour ignored the evidence.

230    The primary judge expressly accepted Dr FitzGerald’s evidence as to economic principles, but ultimately did not accept Dr FitzGerald’s application of those principles to the facts of the case. In particular, his Honour rejected, or gave no, or very limited, weight to Dr FitzGerald’s opinions or conclusions concerning relevant market definitions and competition. The primary reason for that was that his Honour did not accept the various assumptions or conclusions concerning the underlying facts that were critical to the opinions that Dr FitzGerald expressed concerning markets and competition. In a sense, his Honour was rejecting aspects of Dr FitzGerald’s evidence that amounted to no more than the “retailing” of the primary facts. This undermined Dr FitzGerald’s opinions in relation to the relevant market and competition.

231    The question is whether the primary judge erred in rejecting the critical factual assumptions or inferences relied on by Dr FitzGerald and therefore rejecting his opinions.

232    On a close analysis of all the evidence, and the entirety of the primary judge’s reasons, it cannot be concluded that his Honour erred in rejecting, or giving little weight to, Dr FitzGerald’s opinions. His Honour’s findings were open on the evidence and his reasoning does not reveal any error; they are findings with which we agree, from the examination of the record.

233    The most critical assumption or conclusion by Dr FitzGerald was that the services provided by bank branches (including ANZ branches) and brokers were effectively the same or close substitutes. There could be little doubt that this assumption or finding was fundamental to Dr FitzGerald’s opinions and conclusions concerning the existence of a loan arrangement services market in which branches and brokers competed.

234    It is important to consider how Dr FitzGerald came to make that assumption or arrive at that finding. In his first report, Dr FitzGerald noted that he had been asked to assume, amongst other things, that brokers provide services to persons seeking to acquire loan products from loan providers as pleaded and as defined in paragraph 7 of the ACCC’s pleading as “loan arrangement services” and as pleaded and defined in paragraph 6 of the ANZ’s defence as “broking services”. Both these paragraphs of the pleadings are extracted earlier in these reasons. It is to be noted that Dr FitzGerald was asked to, and made, that assumption despite the fact that ANZ did not admit that brokers provided loan arrangement services as defined in the ACCC’s pleading, and despite the apparent differences between the two paragraphs of the pleadings. In effect, Dr FitzGerald was asked to, and did, assume that there was no difference between loan arrangement services and broking services as respectively defined in the ACCC’s FASC and ANZ’s defence.

235    As well as making this assumption, Dr FitzGerald also effectively found (or expressed an opinion) that there was no difference between the ACCC’s “loan arrangement services” and ANZ’s “broking services”. In [13] of his first report, Dr FitzGerald stated:

In respect of services provided to prospective borrowers I do not see these two definitions as significantly different, except that the SoC explicitly states (para 7.1) that these services may be in respect of the loan products of either a single provider only, or more than one provider. For reasons that I elaborate in my opinions given in answer to the questions I have been asked, set out below, I do consider that these services may be in respect of the loan products of either one loan provider or more than one.

(Emphasis in original.)

236    Thus, Dr FitzGerald saw no significance in the fact that loan arrangement services can relate to loan products of a single loan provider, or the loan products of more than one loan provider, whereas “broking services” could involve independently given comparative advice concerning loan products from different loan providers. At [23] of his report, Dr FitzGerald confirmed again that the two “descriptions … do not differ materially”. And at [53], Dr FitzGerald expressed the opinion that the relevant services “are the same or close substitutes” whether they are provided by bank staff (tied to that bank’s own loan products) or by an “untied broker”.

237    In his second report, which was a report that reviewed a report from an economist served by ANZ but not relied on at trial, Dr FitzGerald addressed this issue again. At [45] and [46] of this report, Dr FitzGerald addressed the question whether there was a difference between the services offered by “tied” channels, such as bank franchisees, and “untied” brokers in the following terms:

Moreover, franchised mobile lenders operate in essentially the same way as brokers, and compete directly with them in a relevant geography, by offering most of the same services, notwithstanding that the choices of loan products they can assist loan seekers with are confined to those of one lender. In a world where a substantial majority of loan seekers begin by obtaining information about a number of competing lenders’ products from the internet before going to an intermediary, this is not in my opinion a significant difference between the bundle of loan arrangement services that the two offer.

Equally, I believe that the bundle of loan arrangement services offered by ANZ in-house channels is not materially different from the bundle offered by brokers, with again the only significant difference being that the in-house channels assist with applications only in respect of choices from the range of products offered by Mortgages.

(Footnote omitted.)

238    Dr FitzGerald then referred to two affidavits (not apparently relied on at trial) that identified, as a relevant difference, the fact that the brokers can offer a greater choice. Dr FitzGerald then said (at [49]-[50]):

As regards the points above that in-house channels offer choices only over ANZ’s range of loan products, and do not have the same level of knowledge as brokers do of alternative lenders’ offerings, I note again that:

•    mortgage brokers are very unlikely to be accredited with all lenders operating in Australia, and may not even be accredited with all major lenders (and may in fact be accredited with only a few), so the extent of ‘choice’ they offer may be limited and is in any event a point of differentiation in what they offer loan seekers;

•    in any event, more than two-thirds of seekers of new loans do their own research on the alternatives offered by lenders before coming to a physical channel. The most important parameters (principally about competing variable and fixed interest rates) are very widely publicised and compared both in print and on the internet; and

•    staff of ANZ in-house channels are provided with access to information (as above) on competing lenders’ products – and so can and do respond to loan seekers’ comments or queries on how ANZ’s products compare.

For these reasons I do not see these as fundamental areas of difference between the bundles of loan arrangement services offered by in-house and independent channels. Notwithstanding the differences, these bundles are in my opinion relatively close substitutes in the economic sense.

(Footnotes omitted.)

239    It is important to note that Dr FitzGerald did not support his beliefs or assertions in this regard by reference to the evidence of the so-called “industry witnesses” or the ANZ internal documents. Indeed, it seems (in particular from [54] of his second report) that his views in that regard appear to be based on little more than his observation that “there is a substantial area of overlap among the bundles of services provided by the in-house or tied channels and by brokers.”

240    The basis of the views expressed in these paragraphs was the subject of detailed cross-examination of Dr FitzGerald. It is unnecessary to discuss that cross-examination here. It is summarised (at considerable length) at [417] to [477] of the judgment of the primary judge. Suffice it to say that weaknesses in aspects of Dr FitzGerald’s reasoning were exposed by this cross-examination, as was the fact that some of his views were no more than a product of Dr FitzGerald’s own fact-finding, or his own factual evaluation, or in at least one respect his personal (rather than professional) experience. Dr FitzGerald agreed in cross-examination that these paragraphs of his second report covered the entirety of the areas he had analysed.

241    Of the reasoning in these paragraphs of Dr FitzGerald’s second report, his Honour made the following observations (at [413]):

I have difficulty with Dr FitzGerald’s dismissal of the wider range of products and lenders offered by brokers, as opposed to in-house and tied channels, and of the associated concept of relatively independent advice as to product suitability. The question of substitutability is, of course, to be determined by reference to behaviour in any relevant market, but the industry evidence seems to suggest that in the broking business, a high premium is placed on choice. One might expect that Dr FitzGerald would have explained his reasons for discounting the matter. I also have difficulty with Dr FitzGerald’s insistence that loan products, themselves, are not significantly differentiated. Again, the industry evidence suggests otherwise. I shall return to these questions. As to Dr FitzGerald’s view that use of the internet may fill any gap in the services offered by in-house and tied channels, I have previously indicated that his understanding of the extent of such use in connection with borrowing involves a misunderstanding of the Bank West survey material.

242    It was open to the primary judge to reject (or give little weight to) Dr FitzGerald’s assumptions, views or findings concerning the equivalence of the services offered by banks and the services offered by brokers based on the deficiencies or weaknesses exposed in cross-examination. To the extent they were findings or opinions, as opposed to assumptions, they were not findings or opinions predominantly based on his economic credentials or experience. They were inferences Dr FitzGerald himself drew from some facts (not all of which were supported by evidence) or his own perceptions or experiences.

243    That was not, however, the only basis for the primary judge’s rejection of this aspect of Dr FitzGerald’s evidence. A more fundamental problem was that his Honour found that the industry evidence did not support Dr FitzGerald’s assumptions and conclusions in this regard. His Honour’s reasoning includes the following (at [535]-[539]):

I turn to Dr FitzGerald’s evidence. The industry evidence also does not support many of his assumptions and conclusions. Dr FitzGerald’s assumptions concerning the product supplied by mortgage brokers, and the differences between that product and the product allegedly supplied by lenders is a particular example. One of his reasons for concluding that the products are substitutable is that the loan products, themselves, are not significantly differentiated, so that competition will focus largely on the accompanying services. The evidence from industry witnesses establishes that the broker’s task is to find a loan product which suits a particular potential borrower’s needs and characteristics, and that such a borrower may meet the lending criteria of one lender but not those of another. The evidence also suggests that lending criteria and product characteristics vary over time.

Dr FitzGerald’s view is also, to some extent, dependent upon his belief that many, perhaps most potential borrowers obtain significant amounts of information from the internet. This perception is based on a survey which was, itself conducted on-line. Further, it discloses only that in 2007 63% of respondents used the internet to “access” information on loans, and that 38.1% used it as their main source. However there is no evidence as to the next step taken by these respondents, in particular, as to whether they went to a broker or a lender. Nor is there evidence as to the nature of the information which they obtained from the internet. Dr FitzGerald’s inclusion of the internet as a major source of information concerning loans suggests that a lender’s internal and tied channels are, by themselves, insufficient to compete with brokers, at least for some sections of the community, perhaps re-inforcing the proposition that brokers’ services are significantly different from those provided by a lender’s in-house and tied channels.

There is no evidence as to whether the sites visited by internet users are lenders’ sites or sites dedicated to comparing available products, although Dr FitzGerald says that the Cannex site (which provides some comparable information) is very popular. Access to lenders’ sites would not provide comparative information. A potential borrower would have to visit numerous sites. A broker is not only a supplier of information. He or she is also a filter. If the internet user goes to a site offering comparisons, there may be questions about the accuracy of the relevant information, currency and general reliability. It may be that information from such a site is not as satisfactory to a potential borrower as information provided by a human being in an established office environment, particularly if the office has been in place for some time and/or the potential borrower has been referred to it by another person. Such questions have not been investigated.

Dr FitzGerald and the ACCC seem also to discount the importance to potential borrowers of a broker’s independence from potential lenders. Again, the industry evidence suggests that this is one of the broker’s strengths. This aspect may not be as prominent in much of the promotional material as one might have expected. However the evidence demonstrates that ASIC has discouraged brokers from describing themselves as being independent of lenders, no doubt because of the universal, or near universal remuneration of brokers by lenders. Nonetheless, it is clear that brokers seek to create an image of ethical conduct which at least implies independence. See, for example, the Mortgage Choice Customer Charter in exhibit 1 tab 1. See also the BankWest survey material (exhibit 2) under the heading “Benefits of Using a Broker”. The benefits surveyed clearly create an impression of expert (and therefore independent) advice as to a wide range of products. Further, ANZ seems to have believed that the brokers occupied an area of trust.

Finally, I have difficulty with Dr FitzGerald’s approach to a potential borrower’s purpose when he or she approaches either a lender or a broker. Clearly, a potential borrower approaches a broker in order to obtain advice and assistance for the purpose of applying for a loan. When a potential borrower approaches a lender directly, he or she will generally be simply seeking a loan from that lender. There may be some people who go to bank branches or other lenders’ internal channels to obtain information as to available and suitable products, intending to visit other lenders for the same purpose, and then to compare the outcomes. There is no suggestion that many people do so. In any event brokers offer an alternative to that time-consuming task. Some information may be acquired “over the counter” in branches. Lenders no doubt have brochures which provide information concerning their products, or some of them, but that is not the service which is alleged in this case. The in-house service, which is said to be offered in competition with that offered by brokers, includes assembling the necessary information and advising as to an appropriate product offered by the lender in question, completing and submitting the application and, possibly, providing follow-up services in connection with the application.

244    There is no error in this reasoning. It was open to his Honour, on the whole of the evidence, to find that these critical assumptions and conclusions that founded Dr FitzGerald’s analysis were not supported by the industry evidence.

245    It should be noted, in this context, that the ACCC complained that his Honour’s reasoning in relation to the different attributes of the services offered by bank branches and independent brokers was erroneous because “on the uncontradicted evidence, including the expert evidence of Dr FitzGerald, those attributes might be regarded by some potential borrowers as making the services of brokers more attractive, but did not differentiate those services to the extent that the services of banks and brokers could not or should not be regarded as close substitutes” (ground 6 of the notice of appeal).

246    That submission is rejected. Dr FitzGerald’s evidence was not “uncontradicted”. To the extent that Dr FitzGerald’s views on this issue comprised opinion evidence based on his training and experience, which is doubtful, it might be correct to say that there was no contrary economic opinion evidence. Nonetheless, the basis and correctness of his opinions in this regard were the subject of direct challenge in cross-examination. More significantly, Dr FitzGerald’s views or opinions about how potential borrowers regarded the differences between the services provided by branches and brokers, when properly analysed, were the product of fact-finding or factual evaluation not solidly grounded in Dr FitzGerald’s expertise as an economist. The primary judge was equally qualified to draw inferences and arrive at conclusions about such matters. The ACCC did not identify any other “uncontradicted evidence” said to compel a conclusion contrary to that arrived at by the primary judge.

247    It was open to the primary judge to conclude that the different attributes of the services offered by brokers were sufficiently significant to mean that they were not close substitutes for the services offered by bank branches. The evidence did not compel a contrary finding.

248    The ACCC contended (paragraph 3.3 of the notice of appeal) that the primary judge’s “criticisms” of the expert evidence of Dr FitzGerald were based on a number of “propositions” which were not “raised” with Dr FitzGerald, either in cross-examination or by his Honour. It also contended, in the same paragraph of the notice of appeal, that his Honour’s “criticisms” were based on his interpretation of parts of documents which were not “raised” with Dr FitzGerald. These contentions were not significantly developed by the ACCC in its submissions. It did not identify the specific “criticisms”, or the “propositions” or “interpretation” of documents not put to Dr FitzGerald. Nor did it identify why fairness, or any other consideration, required the propositions or interpretation to be put to Dr FitzGerald.

249    It is not accurate to characterise the primary judge’s treatment of Dr FitzGerald’s evidence as “criticisms”. His Honour simply did not accept many of the assumptions or inferences of fact that founded Dr FitzGerald’s opinions. The cross-examination of Dr FitzGerald directly or indirectly addressed and challenged the important assumptions or inferences upon which Dr FitzGerald based his opinions. It is not correct that these matters were not “raised” with Dr FitzGerald. Further, to the extent that his Honour’s findings concerning the underlying assumptions or factual findings were based on different interpretations of parts of ANZ’s internal documents, his Honour was not, in all the circumstances, obliged to “raise” the potentially different interpretations with Dr FitzGerald in the course of his evidence.

250    The ACCC raised a number of other complaints concerning the primary judge’s treatment of Dr FitzGerald’s evidence. It is unnecessary to consider all those criticisms. It is enough to say that the additional criticisms have no merit. They were based on a mischaracterisation of parts of the primary judge’s reasoning and the importance of Dr FitzGerald’s evidence to it. One example will suffice.

251    The ACCC complained that his Honour rejected Dr FitzGerald’s “functional analysis” which relied, to an extent, on treating ANZ as relevantly comprising different business units. The separate business or economic units included ANZ Mortgage Group, which was responsible for approval and grant of all loan products, and ANZ’s internal distribution channels, in particular the branches. Dr FitzGerald’s opinion was that the relevant market was a “two-sided” market where ANZ Mortgage Group was in a “vertical relationship” with each of the “downstream channels”, including the ANZ branches, and the ANZ branches were also in a “horizontal relationship”, as competitors, with brokers.

252    The ACCC submitted that the primary judge rejected this aspect of the analysis and that in rejecting it he acted contrary to the admitted facts as to the business structure of ANZ and the agreed view of the competing economists (Dr FitzGerald and ANZ’s economist) and economic theory. That submission is based on a mischaracterisation of his Honour’s reasons. His Honour understood Dr FitzGerald’s distinction between business units and accepted that separate treatment of economic units might be useful and appropriate in certain cases and in certain circumstances: see [262] and [263]. His Honour found, however, on the facts and evidence relating to the advice and assistance provided by ANZ branches to customers, that it was artificial to characterise the activities of the branches as supplying loan arrangement services to customers in a market for the supply of such services. Accordingly, the fact that ANZ may have had separate and distinct internal economic or business units was of no moment.

253    His Honour’s reasons for arriving at this finding have already been dealt with at length earlier in these reasons. The important point is that this finding was not a rejection of any aspect of economic theory involving a functional analysis, separate economic units or vertical and horizontal relationships. It was, rather, a finding of fact concerning the proper characterisation of the relevant activities of the ANZ branches. It was, in other words, a rejection of the ACCC’s case (and Dr FitzGerald’s opinion) that the relevant economic theory applied to the particular facts of this case.

254    The ACCC has failed to demonstrate that the primary judge erred in his treatment of Dr FitzGerald’s evidence. Appeal ground 3 is accordingly not made out.

Ground 4 - Did the primary judge err in rejecting, failing to take into account or failing to give appropriate weight to the evidence of “market participants”?

255    This ground relates to the evidence of the industry witnesses (Messrs King, Stark, Lahiff and Percy) and the Mortgage Refunds customers (Messrs O’Malia and Black). The important parts of the evidence of these witnesses are summarised earlier in these reasons. The ACCC relied on this evidence to show that market participants did not differentiate between the services provided by loan providers, through their branches, on the one hand, and brokers, on the other. This in turn demonstrated, on the ACCC’s case, that the services provided by the branches and the brokers were close substitutes. The ACCC also submitted that this evidence showed that the industry witnesses believed that the bank branches were in competition with brokers.

256    The ACCC submitted that the evidence of these witnesses was consistent and that there was no evidence to the contrary. It also submitted that his Honour was wrong to reject this evidence in circumstances where he otherwise accepted that the witnesses were honest and reliable as to matters of fact, and where the articulated bases for rejecting the evidence were unsound.

257    This ground may be disposed of shortly.

258    His Honour neither ignored nor rejected the evidence of these witnesses. He simply found that, when closely analysed, their views or perceptions concerning the similarity of the services offered by branches and brokers, and their perceptions of competition, offered little support for the ACCC’s case. That view of the evidence was open despite his Honour’s finding that the witnesses were honest and reliable in relation to matters of fact, and despite the absence of contrary evidence. When the evidence of these witnesses is considered in its entirety and together with the balance of the other evidence, it was open to his Honour to conclude that their views and perceptions of these matters carried little weight.

259    It is only necessary to deal with two specific complaints by the ACCC concerning the primary judge’s treatment of this evidence. The first concerns Mr Lahiff’s evidence to the effect that brokers competed with the banks’ “retail branches”. This evidence, which was in fairly general terms, was largely undermined by statements in the 2006 annual report issued by Mr Lahiff’s broking firm, Mortgage Choice. That report specifically referred to Mortgage Choice competing with other brokers (at page 5):

Mortgage Choice believes that the combination of the fundamental components of its business model provide it with significant competitive advantages over other brokers in the marketplace ...

260    By way of contrast, the report referred to loan providers as “partners” (at page 6):

Mortgage Choice recognises the importance of developing and nurturing the relationships between broker and lender. Dedicated specialist staff oversee the operational relationship between the Company and its Franchisees and the lender panel. This team provides lenders with structured access to the Franchise network and promotes operational effectiveness by working with lender partners to improve service and processing efficiencies.

The panel includes Australia’s leading lenders, providing a cross-section of products and lender types that Mortgage Choice considers to be a representative spread of available, quality housing loans.

Mortgage Choice believes the benefits enjoyed by lenders from dealing with credible brokers such as Mortgage Choice include …

261    Mr Lahiff was cross-examined about these statements in the annual report. Some of his relevant answers are referred to earlier in these reasons. Suffice it to say, that he maintained that branches and brokers competed.

262    Nevertheless, the contents of the annual report, and Mr Lahiff’s evidence in relation to it, led his Honour to not accept Mr Lahiff’s evidence in relation to competition between branches and brokers. His Honour concluded as follows (at [203]):

The annual report leads me to infer that Mr Lahiff’s view as to competition with lenders, as expressed in his evidence, is either an ex post facto rationalization or reflects use of the word “competition” in a way which is not relevant for present purposes. It may be that he is using it to describe generally, competition in the supply of loan products in the lending market.

263    The ACCC complained that it was wrong for the primary judge to reject this aspect of Mr Lahiff’s evidence on this basis. It said that his Honour ignored Mr Lahiff’s evidence in cross-examination and ignored another contemporaneous document that included the statement, “Why choose Mortgage Choice over a bank to help you fund your loan?”.

264    Neither of these matters provides a proper basis to conclude that the primary judge erred in relation to this aspect of the evidence. The fact that Mr Lahiff continued, in the face of cross-examination, to maintain that there was competition did not mean that his Honour was bound to accept his evidence in that regard. As for the other document relied on by the ACCC, it also referred to the fact that banks “enthusiastically support the services provided by Mortgage Choice” and pointed to aspects of Mortgage Choice’s services that banks could not replicate. Those aspects included the ability to offer products from many different lenders and the ability to offer unbiased advice because Mortgage Choice would not “plug” a particular product. The latter statement was adverse to the ACCC’s case on substitutability. The former statement at least casts doubt on the existence of competition between Mortgage Choice and banks. It would be unusual for one party to enthusiastically support the services of its competitor. Considered in context and along with all the other evidence, the statement in this document relied on by the ACCC does not undermine the primary judge’s reasons for rejecting this aspect of Mr Lahiff’s evidence.

265    The ACCC’s second complaint concerned the primary judge’s observation, at [526], that when the lay witnesses spoke of brokers “competing” with lenders, it was “not at all clear that they recognised [the] ACCC’s distinction between the supply of loan arrangement services and the supply of loan products.” By this, his Honour meant that when Messrs King, Stark, Lahiff and Percy referred in their evidence to “competition”, they may well have been referring to competition between the various distribution channels in relation to the distribution of loan products. In other words, they were referring to competition between the distribution channels to be the one to lodge successful loan applications on behalf of borrowers. This was competition that occurred in the market for the supply of loan products.

266    His Honour found that Mr Percy was certainly referring to competition in this latter sense. That finding was plainly open on the evidence. When asked in cross-examination what it was that the “branch channel” and the “broker channel” provided that made them in competition with each other, Mr Percy said “loan products.” A fair reading of the balance of his evidence confirms that when he referred to competition, Mr Percy was referring to competition between distributors in the loan products market.

267    His Honour also inferred that the other industry witnesses were also referring to competition in that broad sense. This was not, as the ACCC submitted, a conclusion that was reached by extrapolation from Mr Percy’s evidence. His Honour carefully analysed the evidence of each of the industry witnesses. His Honour had the advantage of hearing all their evidence and considering it together with the evidence as a whole. There is no basis for the ACCC’s contention that the primary judge erred in his assessment of this evidence.

Ground 5 - Did the primary judge err in failing to find that ANZ’s internal documents demonstrated ANZ’s opinion that it competed with brokers in relation to the provision of loan arrangement services?

268    The ACCC submitted that ANZ’s internal documents “consistently” demonstrated ANZ’s opinion that its internal distribution channels, comprising its branches, employees, agents and franchisees, were in competition with brokers to attract customers and, therefore, in competition to provide loan arrangement services for the purposes of attracting customers. The ACCC raised specific complaints in relation to the primary judge’s findings and reasoning concerning the internal documents. In particular, it attacked the primary judge’s observation that the particular parts of the documents relied on by the ACCC needed to be considered in context and that, in most if not all instances, the context was competition in the market for the supply of loan products. The ACCC also submitted that his Honour’s finding that these documents did not assist the ACCC’s case was based on findings or reasons that were not “valid”, “justified” or open on the evidence.

269    The important passages from the documents highlighted by the ACCC in its submissions are extracted earlier in these reasons. His Honour’s key findings in relation to the documents and what inferences or conclusions could be drawn from them are also detailed earlier. It is unnecessary to rehearse these matters here.

270    There were undoubtedly some statements in ANZ’s internal documents that suggested that some senior officers of ANZ considered, in a broad sense, that ANZ’s internal distribution channels competed with brokers and that it was in ANZ’s interests to have its internal channels replicate the services provided by brokers so that ANZ could reduce its level of reliance on brokers. Those statements, however, unquestionably needed to be considered in context. The relevant context included not only the nature and content of the documents in which the particular statements were made, but also the evidence in its entirety.

271    The ACCC complained that his Honour did not identify what the context was. That submission is untenable. The primary judge’s reasons analysed the documents in painstaking detail so as to place the statements relied on by the ACCC in context, both in terms of the particular documents from which the statements were extracted, but also the overall impression conveyed by the totality of the documentary evidence.

272    In the course of his detailed analysis of the documents in the judgment, the primary judge made numerous observations, comments and findings concerning particular statements in particular documents, and the inferences or conclusions that could be drawn from them. Ground 5 in the notice of appeal, and the ACCC’s submissions in relation to it, identify some, albeit only a small fraction, of his Honour’s findings in relation to some of the documents and contends that those findings were not valid, justified or open. This approach to the judgment and findings of the primary judge is neither fruitful nor to the point. Even if it was accepted, which it is not, that some of these individual findings and observations by his Honour were not open on the evidence, that would not undermine in any material way his Honour’s ultimate findings or conclusions on the evidence.

273    The primary judge’s overall conclusions, in relation to the ANZ’s documents are summarised in the judgment at [549]-[551] in the following terms:

My consideration of the ANZ documentation leads me to conclude that a number of matters were under consideration by ANZ Mortgage Group in 2003-2004. They included:

•    increasing market share in the lending market;

•    providing customers with the convenience of mobile lenders, as did other lenders;

•    reducing reliance on the brokers;

•    maintaining the support of the brokers;

•    avoiding retaliation from brokers; and

•    improving customer management and relations.

Despite the occasional references to competing with brokers, little attention was given to the implementation of such “competition”. There was no suggestion that ANZ should abandon the use of brokers, or that future accreditations should be limited. There was no suggestion of an advertising campaign focussing on the advantages of in-house or tied channels over brokers. Although brokers’ methods were to be emulated, there was no suggestion that ANZ could neutralize the advantages of choice and relative independence which brokers enjoyed. There was no suggestion that there be a quota regulating the amount of business to be accepted from brokers.

As I have said, at least one, and perhaps the dominant aim in establishing the franchise model was to remedy a perceived shortcoming in ANZ’s distribution channels as compared to those of other lenders. No real attempt was made to identify potential borrowers who might be tempted away from the brokers. Clearly, the expectation was that the brokers would retain existing levels of custom, even if their overall share of the ANZ “book” were to be reduced. ANZ was aware that its share of the lending market was only 11%. Thus there was plenty of room for expansion, by brokers and by ANZ in-house channels. Indeed, in the early stages, it was thought that ANZ’s plans might result in extra business for the brokers. In any event, ANZ did not want to risk broker support. Any attempt to depart from the principle of channel neutrality was seen as fraught with the risk of retaliation by the brokers. Finally, any expansion of market share seems to have been sought in the lending market rather than in any market for loan arrangement services or broking services. ANZ simply wanted to sell more mortgages in the lending market, preferably through in-house and tied channels.

274    These conclusions were open on the evidence. Indeed, they were supported in many respects by other evidence concerning the nature of the relationship between loan providers, such as ANZ, and mortgage brokers. The ANZ documents did not, as the ACCC effectively contended, compel a conclusion that ANZ competed with brokers in relation to the provision of “loan arrangement services”.

275    The primary judge did not err in his consideration of the documentary evidence or the inferences and conclusions he drew from it.

Ground 6 - Did his Honour err in finding that the services provided by bank branches were not close substitutes for the services provided by brokers?

276    This ground has already been addressed earlier in these reasons in the context of ground 3. The primary judge did not err in concluding that the services provided by the bank branches and brokers were not close substitutes. His Honour’s conclusion was based, in large part, on his finding that the different nature and attributes of the services provided by bank branches and independent brokers were significant. He concluded (at [643]) that the choice consumers would make between the different services would depend on their particular needs at particular times. They were not close substitutes because the choice would not be influenced in any material way by price changes. There is no demonstrated error in that reasoning or the factual findings underlying it.

277    It should also be emphasised that this was an alternative and independent basis for dismissing the ACCC’s case. The main basis for the decision was his Honour’s finding that the “in-house and tied channels” did not provide loan arrangement services to potential borrowers. It follows, that even if this ground was made out, it would not necessarily result in the appeal being allowed.

Ground 7 - Did the primary judge err in characterising the services by reference to rights, duties and liabilities as between bank branch and customer?

278    This ground challenged parts of the primary judge’s reasoning relating to the characterisation of the activities carried out by staff of ANZ branches for the purposes of identifying what market those activities related to. His Honour’s reasons for characterising the activities as he did are outlined earlier in these reasons, including in the context of ground 1.

279    The ACCC’s specific criticisms that are the subject of this ground related to the reasoning in the primary judge’s judgment at [622]-[628], where his Honour analysed the activities by reference to the duties and obligations of the ANZ bank officers. The ACCC’s submissions were primarily directed at [622] of the judgment, which was in the following terms:

Whilst it may accurately be said that ANZ provided banking services, including the supply of loan products at its branches, it does not follow that everything done at a branch, or even each interaction with a potential borrower involved the supply of a service to that person. Branch employees were employees of ANZ. They performed duties as directed by ANZ for the purposes of its banking business. Some branch employees, from time to time, performed duties in connection with the supply of loan products on behalf of ANZ Mortgage Group, including those duties identified in para 7.2 of the statement of claim, or similar duties. They performed such functions at the direction of ANZ and as an aspect of their employment. Such functions were, prima facie, steps taken at the request of, and for the benefit of ANZ Mortgage Group, and at its expense. In a sense the performance of such functions may have benefited potential borrowers, but again, it does not follow that the branch staff provided services to them. In effect, branch staff dealt with potential borrowers on behalf of ANZ Mortgage Group. Neither it, nor any branch staff member undertook a duty to a potential borrower to perform the functions identified in para 7 of the statement of claim. Nor did any obligation arise out of the performance of those functions, subject only to consumer protection and similar legislation, and the law relating to misrepresentation. It may not be necessary, in order that there be a supply of services, that there be accompanying or resulting legal consequences. Nonetheless the presence or possibility of legal rights, duties and liabilities may say much about the nature of a particular transaction or class of transactions.

280    The ACCC contended that this analysis of legal rights, duties and liabilities was a “significant criterion” in respect of the distinction drawn by the primary judge between the services provided by bank staff and the services provided by brokers. It submitted that the reasoning is incorrect for two reasons. First, the ACCC contended that if correct, it would have the result that no vertically integrated firm could compete with a firm operating over only one functional level. Second, the ACCC submitted that there was no evidence that the bank employees’ duty was to advance the interests of ANZ Mortgage Group without regard to the provision of appropriate advice and assistance to the customer.

281    These arguments may be dealt with shortly. In relation to the first argument, the inferences and conclusions drawn by his Honour in relation to the characterisation of the activities of branch officers related solely to the facts and evidence in this matter. His Honour did not suggest that, as a matter of general principle, ANZ, as a vertically integrated firm, cannot compete with non-integrated firms. His Honour did not analyse or characterise the activities or arrangements by reference to economic concepts such as vertical integration. Perhaps his Honour was mindful of the admonition or warning given by the High Court in Visy Paper Pty Ltd v Australian Competition and Consumer Commission (2003) 216 CLR 1 at [23]-[27] concerning the use of such economic terms and analysis in relation to contraventions under the Act in circumstances where those economic terms are not replicated in the relevant provisions of the Act.

282    There may well be cases where a division of a vertically integrated corporation provides goods or services which are close substitutes for goods or services provided by an external supplier. In such a case there may be competition between the vertically integrated firm and the external supplier. But, on the evidence, this matter was not such a case.

283    As for the second argument, there was ample evidence from which the primary judge could infer that branch officers owed a duty, as ANZ employees, to advance the interests of ANZ and to therefore offer advice or assistance only in relation to ANZ’s loan products. In contrast, there was considerable evidence that a particular selling point for brokers was that they were independent of particular lenders. They had no duty to advance the interests of any one lender. It is also wrong to characterise his Honour’s reasoning as including the proposition that branch officers could act without regard to the provision of appropriate advice and assistance to the customer. His Honour recognised, in [622], that branch officers were subject to “consumer protection and similar legislation, and the law relating to misrepresentation.”

284    His Honour’s analysis in terms of rights and duties was also not critical to his characterisation of the activities of the branch employees, though it was plainly an important consideration. His Honour recognised, at [622], that “[i]t may not be necessary, in order that there be a supply of services, that there be accompanying or resulting legal consequences.” His Honour also accepted, at [630], that to describe a particular function as an employee’s “duty” to his or her employer did not necessarily exclude the possibility that services were also supplied. In the particular circumstances of this case, however, his Honour considered that this was a significant point of distinction between the activities or services provided by a branch officer and the services supplied by a broker. There is no error involved in that reasoning.

285    Accordingly, ground 7 has no merit and is rejected.

Ground 8 - Did the primary judge err by relying on his own understanding or own views about the interaction between bank staff and customers?

286    This ground again may be disposed of shortly. It again relates to some of the reasons of the primary judge in relation to the characterisation of the activities of ANZ branch officers. That reasoning concerned the fact that ANZ branches only provided advice and assistance in relation to ANZ loan products, whereas brokers could and did provide comparative advice concerning loan products.

287    At [539], his Honour made the following observations:

Finally, I have difficulty with Dr FitzGerald’s approach to a potential borrower’s purpose when he or she approaches either a lender or a broker. Clearly, a potential borrower approaches a broker in order to obtain advice and assistance for the purpose of applying for a loan. When a potential borrower approaches a lender directly, he or she will generally be simply seeking a loan from that lender. There may be some people who go to bank branches or other lenders’ internal channels to obtain information as to available and suitable products, intending to visit other lenders for the same purpose, and then to compare the outcomes. There is no suggestion that many people do so. In any event brokers offer an alternative to that time-consuming task.

288    The ACCC submitted that this was his Honour’s own view and was unsupported by evidence. That submission is rejected. Two points should be made. First, this is an inference, indeed a fairly obvious inference, that could be drawn from the different features of the activities of the services provided by bank branches and brokers. More significantly, it was supported by evidence adduced from Mr O’Malia and Mr Black in cross-examination. Second, the impugned observations by the primary judge were made in response to some statements made in Dr FitzGerald’s report concerning a potential borrower’s purpose when he or she approaches either a broker or a lender. This was an example of fact-finding or fact evaluation by Dr FitzGerald that was outside his area of professional expertise as an economist. His Honour was in no worse position than Dr FitzGerald to draw such inferences.

289    The primary judge’s findings and reasoning at [622]-[628] relating to the interaction between a bank officer and a customer have been addressed earlier in the context of other grounds of appeal, including ground 7. It is sufficient to say that his Honour’s reasoning was not based on his own understanding or his own views, but was based on inferences able to be drawn from the whole of the evidence.

290    Ground 8 has no merit and is rejected.

Ground 9 - Did the primary judge err in failing to find that ANZ provided loan arrangement services in competition with Mortgage Refunds?

291    This ground has been effectively addressed in the context of other specific appeal grounds. No error in his Honour’s findings or reasoning has been demonstrated by the ACCC.

Ground 10 - Did the primary judge err in failing to find that the evidence of Mr King and Ms Zacka concerning the agreement between ANZ and Mortgage Refunds was consistent with the existence of competition between ANZ branches and brokers?

292    The evidence of Mr King and Ms Zacka concerning their conversation with Mr Carroll of ANZ in relation to the impugned agreement between ANZ and Mortgage Refunds is extracted earlier in this judgment. This evidence was not challenged. Mr Carroll was not called as a witness by ANZ.

293    There is no doubt that, in some circumstances, inferences relevant to market identification and competition can be drawn from the views and actions of those within an industry, including the nature of an agreement between alleged competitors: Rural Press Limited v Australian Competition and Consumer Commission (2003) 216 CLR 53 at [45]; Boral Besser Masonry Ltd v Australian Competition and Consumer Commissioner (2003) 215 CLR 374 at [257]. The ACCC submitted, in effect, that the evidence of the arrangement in this matter, and the terms of the conversation between the senior ANZ officer and the representatives of AFG and Mortgage Refunds, compelled a finding that ANZ competed with Mortgage Refunds in respect of loan arrangement services. The ACCC, in its submissions, asked rhetorically: “what was the need for a level playing field if ANZ management did not have the view that it competed with Mortgage Refunds and other brokers in the provision of customer services?”.

294    The primary judge gave an answer to that rhetorical question. The answer was that a level playing field was needed for the purposes of channel neutrality. In his Honour’s words “Mr Carroll was simply seeking to quell branch dissatisfaction, as his statement in fact suggests” (see [596]). This answer to the rhetorical question was based on the whole of the evidence. His Honour’s reasoning is explained further at [581] in the following terms:

I should make one other point. ACCC’s submission assumes that channel neutrality is inevitably motivated by an anti-competitive purpose. Yet there is logic in the proposition that an up-stream supplier may simply wish to avoid damaging and pointless conflict between distribution channels. In taking action in response to Mortgage Refunds’ conduct, ANZ initially sought to enforce, as against AFG, the terms of the originator agreement. The subsequent “re-instatements” were, in fact, on more favourable terms than those contained in that agreement. ANZ approved the offering of a not-insubstantial refund which, as far as the evidence goes, the branches would not generally match. In order to demonstrate the existence of a competitive relationship by reference to conduct, one must show that a desire to hinder competition is a likely reason for such conduct. For what it is worth, I am simply unconvinced that ANZ’s conduct was so motivated. Having regard to the apparently isolated nature of the conduct and the limited nature of the restraint imposed, I am inclined to the view that ANZ was trying to keep the branches happy rather than seeking to restrain perceived competition. That view is, of course, based on all of the evidence in the case. It does not dispose of the case. However it disposes of one basis upon which ACCC relies in order to establish that ANZ and Mortgage Refunds were providing services in competition with each other.

295    There is no error in this reasoning. His Honour’s refusal to answer the rhetorical question in the way the ACCC would have liked was open on the evidence.

Grounds 11 and 12

296    These grounds are effectively “catch-all” grounds. They assert that the primary judge erred in not finding that ANZ contravened s 45(2)(a)(ii) and s 45(2)(b)(ii) of the Act. To the extent that the ACCC advanced any separate submissions in support of these grounds, they have been dealt with in the context of other specific grounds, or elsewhere in these reasons. His Honour did not err in finding that the ACCC had not made out the contraventions as pleaded on the evidence.

A final observation in relation to the appeal

297    It remains to make one final observation. The resolution of this matter turned on the evidence and the unique facts and circumstances, including the particular way the ACCC framed its case in relation to the alleged market and competition. Ultimately, the evidence did not support the existence of the particular market pleaded by the ACCC and failed to establish competition between ANZ and Mortgage Refunds in the particular defined market. It does not necessarily follow that there can never be a case where a manufacturer (or product originator) which has its own distribution division (or separate economic unit) competes with external distribution channels in the market for the supply of the particular product. No such general principle can or should be extracted from the outcome of this matter, including this appeal. Each case needs to be considered on its own facts and circumstances.

ANZ’S CROSS-APPEAL

298    Given that the ACCC’s appeal has been determined adversely to it, it is strictly unnecessary to consider ANZ’s cross-appeal. Nevertheless, since it raises only a short point concerning the construction of s 45A of the Act and its application to the facts of this case, the better course is to resolve this aspect of the controversy.

299    For the purposes of resolving the cross-appeal it must be assumed, contrary to the primary judge’s findings that have been upheld, that Mortgage Refunds supplied loan arrangement services to borrowers in competition with ANZ.

300    The agreement between ANZ and Mortgage Refunds contained the following term:

The maximum refund that can be provided to the customer in relation to an ANZ Loan Product is to be no greater than the amount of the Loan Approval Fee as determined by the ANZ Bank. The amount of this fee may be altered at anytime and at the Bank’s sole discretion.

301    There is little doubt that this term of the agreement had the effect of fixing, controlling or maintaining the amount of money that Mortgage Refunds could pay, or could offer to pay, its clients if they successfully applied for an ANZ loan through Mortgage Refunds. But could the amount paid by Mortgage Refunds, however it was described by it, be properly characterised as “the price for, or a discount, allowance, rebate or credit in relation to” the loan arrangement services supplied by Mortgage Refunds?

302    The primary judge held that the payment could not be described as a discount, allowance, or credit. It could, however, be described as a rebate. In so concluding, his Honour gave a broad construction to s 45A holding (at [610]) that the use of the words “price”, “discount”, “allowance”, “rebate” and “credit” in s 45A “suggests an intention to cover all payments or reductions in payments due.” The relevant “payment due” was, in his Honour’s view, the amount to be paid to the lender (ANZ) by the borrower as part of the loan transaction. This would include the loan approval fee and/or the first interest instalment payable by the borrower.

303    His Honour also held that the payment by Mortgage Refunds to the borrower was “in relation to” the loan arrangement services supplied by Mortgage Refunds to the borrower. His Honour reasoned as follows (at [612]):

The refund was in relation to such services in the sense that the availability of the refund was used as a “hook” to attract potential borrowers to avail themselves of the whole range of the brokers’ services. I consider that the refund was a rebate in relation to the supply of the brokers’ services. Whether those services were loan arrangement services as pleaded is another matter.

304    ANZ submitted that the Mortgage Refunds payment could not properly be characterised as a “rebate” for the purposes of s 45A for two related reasons. First, however it may be described, the payment was not “in relation to” the loan arrangement services provided by Mortgage Refunds. Because Mortgage Refunds did not charge the customer any fee for the services it provided, in ANZ’s submission there was no nexus between the “refund” and the service. If anything, the refund was “in relation to” the supply of the loan product by ANZ.

305    Second, ANZ submitted that even if the payment was “in relation to” the supply of loan arrangement services by Mortgage Refunds, it could not be characterised as a discount, allowance, rebate or credit. That is because the common feature of all of these words is the notion that the payment was in some way a reduction in, or offset to, the cost to the consumer of the relevant goods or service. Here, it is common ground that the relevant service was free. It therefore makes no sense, as a matter of construction of the ordinary language, to consider the payment as a reduction of a nil amount.

306    There is some merit in ANZ’s submissions. They expose, it must be said, an apparent tension or inconsistency in his Honour’s reasoning in relation to this aspect of the matter. On the one hand, his Honour found that the payment was a rebate because it reduced the amount payable by the borrower to ANZ, being the loan approval fee or an interest payment. That would suggest that the rebate was “in relation to” the supply of the loan product by ANZ. On the other hand, however, his Honour found that the payment was “in relation to” the supply of loan arrangement services by Mortgage Refunds because it was, in effect, offered as an inducement to prospective borrowers to use the services of Mortgage Refunds.

307    The primary judge’s characterisation of the Mortgage Refunds’ payment as a “hook” or inducement to attract prospective customers to utilise the services of Mortgage Refunds is an accurate characterisation of the payment. The offer of a refund was Mortgage Refunds’ key selling point in the market for broking or loan arrangement services. The services offered by Mortgage Refunds included lodging loan applications and effectively liaising between the loan provider and the borrower.

308    In these circumstances, it was correct for the primary judge to conclude that the payment was “in relation to” the loan arrangement services provided by Mortgage Refunds.

309    That conclusion, however, makes it difficult to characterise the payment as a rebate or, for that matter, a discount, allowance or credit. Even accepting the primary judge’s broad construction of these words in s 45A as encompassing any “reductions in payments due”, the payment was not a reduction of any payment due by the borrower to Mortgage Refunds. And, having accepted that the payment related to the services provided by Mortgage Refunds, it is difficult to see how or why it was open to characterise the payment as if it in some way related to the supply of the loan by ANZ, in the sense that it reduced some amount otherwise payable by the customer to ANZ.

310    The proposition that the payment reduced either the loan approval fee or the first interest payment by the borrower is also inconsistent with the characterisation of the payment as an inducement to use the services of Mortgage Refunds. There is no suggestion in the evidence that the payment was made otherwise than straight from Mortgage Refunds to the customer. It was not directed to ANZ to meet any of the borrower’s commitments to ANZ. Nor was the borrower required to use the funds in any particular way. The borrower was not required to use the funds to reduce any amounts otherwise payable to ANZ. It was, in this respect, no different to his Honour’s metaphorical “set of steak knives” (cf. [607]). The customer was obliged to pay the loan application fee and first interest payment to ANZ whether or not Mortgage Refunds made the refund payment and however the customer chose to use that payment.

311    The effect of the ACCC’s submissions on this point was that, so long as any limitation or fixing of the payment could be considered to be anti-competitive because it limited the capacity of Mortgage Refunds to compete on price in the relevant loan arrangement services market, it is unnecessary to give close attention to the words used in s 45A. Those words were designed to “cover the field” and not leave some sort of “incidental gap” through which such competitive behaviour could be squeezed.

312    That submission is rejected. Section 45 is a penalty provision. Section 45A can operate to create a per se contravention by deeming a contract to have the purpose or effect of substantially lessening competition. In the circumstances, there is no warrant for construing s 45A as broadly as the ACCC contended, or in a way which effectively ignores the text.

313    His Honour erred in finding that the Mortgage Refunds payment could be considered to be a “rebate” in relation to any services supplied by Mortgage Refunds in competition with ANZ. Nor, as the ACCC contends, could the payment be characterised as a discount, allowance or credit in relation to those services. This finding alone would have warranted the dismissal of the ACCC’s case.

314    ANZ’s cross-appeal should be allowed.

conclusion and disposition

315    The ACCC has failed to demonstrate any error on the part of the primary judge.

316    The appeal should be dismissed with costs. The cross-appeal should be allowed with costs. There is no need for any further or consequential orders on the cross-appeal.

Associate:

Dated:    31 July 2015