AUSTRALIAN COMPETITION TRIBUNAL
Applications by Australia and New Zealand Banking Group Limited and Suncorp Group Limited [2024] ACompT 1
IN THE AUSTRALIAN COMPETITION TRIBUNAL
ACT 1 of 2023 | |
Re: | Applications by Australia and New Zealand Banking Group Limited and Suncorp Group Limited for review of Australian Competition and Consumer Commission Merger Authorisation Determination MA1000023 |
Applicants: | Australia and New Zealand Banking Group Limited and Suncorp Group Limited |
Intervenor: | Bendigo and Adelaide Bank Limited |
DETERMINATION
TRIBUNAL: | Justice Halley (Deputy President) Dr J Walker (Member) Ms D Eilert (Member) |
DATE: | 20 February 2024 |
WHERE MADE: | Sydney |
THE TRIBUNAL DETERMINES AND DIRECTS THAT:
1. The determination of the Australian Competition and Consumer Commission (ACCC) dated 4 August 2023 be set aside pursuant to s 102(1) of the Competition and Consumer Act 2010 (Cth) (CCA).
2. Australia and New Zealand Banking Group Limited (ANZ) is granted authorisation pursuant to s 88(1) and s 102(1) of the CCA to acquire from Suncorp Group Limited (SGL) 100% of the issued share capital in SBGH Limited, either directly or via a related body corporate of ANZ, and certain real estate and intellectual and other property rights held by other SGL entities to facilitate the operation of Suncorp-Metway Limited, in accordance with a share sale and purchase agreement between ANZ and SGL executed on 18 July 2022.
3. Until further direction of the Tribunal, the reasons of the Tribunal in this proceeding dated today are not to be made available to or published to any person save for:
(a) the ACCC, its staff and any other person assisting the ACCC in relation to the proceeding including the ACCC’s legal advisers; and
(b) the parties’ legal advisers who, by reason of previous directions of the Tribunal, are permitted to have access to the confidential information of each of the parties to the proceeding.
4. Within 10 days of the date hereof, the parties are to file jointly:
(a) a copy of the Tribunal’s reasons that marks, by way of coloured shading, those parts of the reasons that a party or the ACCC seeks to have redacted on the grounds of commercial confidentiality or on the grounds that the information is protected information for the purposes of s 56 of the Australian Prudential Regulation Authority Act 1998 (APRA Act). Different coloured shading is to be used for each party, the ACCC and information that is protected information for the purposes of s 56 of the APRA Act; and
(b) short submissions addressing the basis for the claim of confidentiality on behalf of each party and the extent to which those confidentiality claims are agreed.
5. Nothing in these directions prevents the ACCC from consulting with the Australian Prudential Regulation Authority (APRA) in respect of the reasons of the Tribunal.
THE TRIBUNAL:
1 The applicants, Australian and New Zealand Banking Group Limited (ANZ) and Suncorp Group Limited (SGL), seek a review by the Tribunal of a decision by the Australian Competition and Consumer Commission (ACCC), on 4 August 2023, not to authorise an acquisition by ANZ of Suncorp-Metway Limited (Suncorp Bank) from SGL.
2 The ACCC declined to authorise the proposed acquisition because it was not satisfied that (a) the acquisition would not be likely to have the effect of substantially lessening competition in a national home loans market and in local or regional banking markets in Queensland for agribusiness customers and small to medium enterprises (SME), and (b) that the benefits to the public of the proposed acquisition would outweigh detriments to the public, from the proposed acquisition.
3 The applications by ANZ and SGL are opposed by Bendigo and Adelaide Bank Limited (Bendigo). Bendigo contends that the proposed acquisition of Suncorp Bank by ANZ would be likely to have the effect of substantially lessening competition in a national home loans market and in local or regional banking markets in Queensland for agribusiness customers. It contends that the acquisition of Suncorp Bank by ANZ would further consolidate ANZ’s market position and structural advantages as a major bank, without engaging in competition, to the substantial detriment of competition in the home loans and agribusiness markets.
4 In addition, the State of Queensland sought and was given leave to provide a submission for the sole purpose of clarifying information provided to the ACCC in connection with the making of its determination.
5 For the reasons that follow, the Tribunal is satisfied that the acquisition of Suncorp Bank by ANZ would not be likely to have the effect of substantially lessening competition in each of the national home loans market and local or regional Queensland markets for agribusiness customers and SME. Further, although strictly not necessary to determine given that finding, the Tribunal is also satisfied that the acquisition would be likely to result in a benefit to the public that would outweigh the detriment to the public that would be likely to result from the acquisition.
A.2. The commercial and economic context of the proposed acquisition
6 The Australian banking sector presently includes the four major banks: ANZ, Commonwealth Bank of Australia (CBA), National Australia Bank (NAB) and Westpac Banking Corporation (Westpac) (together, Major Banks), second-tier banks and other authorised deposit taking institutions (ADIs). The second-tier banks include banks at or around the size of Suncorp Bank, Macquarie Bank (Macquarie), ING Bank Australia (ING), Bendigo, Bank of Queensland and HSBC Bank Australia (HSBC). All of the Major Banks and second-tier banks are ADIs.
7 The Tribunal notes that beyond traditional non-bank lenders, since the Global Financial Crisis banking systems around the world have been facing threats of disruption from a range of digital players, particularly “BigTech” firms (large platform-based service providers such as Google, Apple and Amazon). As banking moves from a world of physical service delivery to one of digital service delivery, BigTech firms are expanding into financial services, particularly payment services.1
8 The threat posed by BigTech firms to the banking sector was summarised in the following terms in a [REDACTED] Board Paper from 2021:
[REDACTED]2
9 The Major Banks collectively account for 72% of reported banking system assets in Australia.3 The second-tier banks each have a share of banking system assets greater than 1% and collectively account for close to 14% of reported banking system assets, having increased their share of assets over the past decade. Other ADIs individually hold a share of banking system assets less than 0.7%, which includes 49 foreign bank branches, who primarily target niche areas and 57 credit unions and building societies, who operate under a mutual structure where customers are members and profits are reinvested back into the business.4
10 The share of the Major Banks, second-tier banks and other ADIs, as a proportion of banking system assets as at May 2023, are set out in the following table:5
11 Non-ADI lenders account for around 5% of total financial system assets in Australia.6 Non-ADI lenders include smaller financial technology providers which can be broadly described as “fintechs”.7
12 The Major Banks and the second-tier banks all provide retail and business banking services and have larger retail loan books than business loan books. Banks, however, focus on different customer segments and the relative size of their retail portfolio compared with their business portfolio will vary significantly by bank.8 For second-tier banks such as Bendigo, Bank of Queensland, ING, HSBC, Macquarie and Suncorp Bank, their retail lending (lending to households or individuals) represents between 78% and 89% of their total loan book. Suncorp Bank’s retail lending is 80% of its total loan book.9 It is generally the case that second-tier banks provide a higher proportion of retail lending relative to their business lending, than the Major Banks.
13 Smaller banks also have portfolios focused on particular business customer segments. For example, for [REDACTED], lending to SME customers accounts for between [REDACTED] of total business lending.10 The banks with the largest exposure to agribusiness lending as a proportion of their total business lending are [REDACTED].11
14 On 2 December 2022, ANZ lodged an application with the ACCC seeking authorisation under s 88(1) of the Competition and Consumer Act 2010 (Cth) (CCA) for ANZ to acquire 100% of the shares in SGBH Limited from SGL pursuant to the terms of a share sale and purchase agreement between ANZ and SGL executed on 18 July 2022 (Proposed Acquisition). SGBH Limited is a non-operating holding company that owns 100% of the shares of Suncorp Bank, which together with its subsidiaries, owns and operates SGL’s banking business in Australia.
15 On 4 August 2023, the ACCC made a determination dismissing the application by ANZ for authorisation of the Proposed Acquisition.
16 The ACCC concluded with respect to competitive effects that:
Given the ACCC’s conclusions in relation to the markets for home loans, SME banking, and agribusiness banking, the ACCC is not satisfied in all the circumstances that the Proposed Acquisition is not likely to substantially lessen competition.12
17 The ACCC concluded with respect to public benefits and detriments that:
The ACCC is not satisfied, in all the circumstances, that the Proposed Acquisition would result, or be likely to result, in a benefit to the public that would outweigh the detriment to the public that would result, or be likely to result, from the Proposed Acquisition.13
18 The ACCC records in its determination that it commissioned and had regard to an expert report from Professor Nicolas de Roos and two expert reports from Mary Starks in reaching its conclusions. It also records that it had regard to submissions received from ANZ, SGL and interested parties, including:14
• ANZ’s application in support of the merger authorisation and related annexures including witness statements
• Expert reports provided by the merger parties and interested parties
• 27 submissions from interested parties in response to the ACCC’s initial consultation process including six confidential submissions
• over 20 submissions from interested parties following the ACCC’s statement of preliminary views
• ANZ’s response to submissions from interested parties,89 the ACCC’s statement of preliminary views, and Ms Starks’ reports, including witness statements and expert reports
• Suncorp Group’s response to submissions from interested parties, the ACCC’s statement of preliminary views, and Ms Starks’ reports, including witness statements and expert reports
• submissions from the merger parties
(Footnotes omitted.)
19 The ACCC provided the following summary of the submissions made by interested parties:
Overview of interested party submissions
3.26. Submissions to the ACCC from interested parties raise a variety of views on issues related to the Proposed Acquisition.
3.27. Several interested parties raised competition concerns regarding the Proposed Acquisition, noting:
• it would result in the removal of a significant second-tier competitor
• Suncorp Bank’s position in the supply to agribusiness markets
• there is inertia for customers switching between providers
• the low prospect of successful new entry and expansion in the industry
• increased consolidation of market share among major banks would make the relevant markets more susceptible to coordinated interactions.
3.28. In particular, CowBank and BMAgBiz submit that agribusiness customers have specialist needs and expressed concern that the Proposed Acquisition would reduce competition and availability of finance, particularly for bespoke agribusinesses with more complex needs. Judo Bank and Bank of Queensland submit that markets for supply to agribusiness and small to medium enterprise (SME) customers demonstrate state, regional and local characteristics.
3.29. BEN and Judo Bank submit that there is a realistic likelihood that Suncorp Bank, at least in part, could be acquired by another purchaser. BEN further submits that the counterfactual where Suncorp Bank is acquired by BEN is commercially realistic.
3.30. In addition to competition concerns, interested parties opposing authorisation generally raised concerns that the Proposed Acquisition would lead to public detriments and some questioned the extent to which synergies and benefits claimed by the merger parties will materialise if the Proposed Acquisition proceeds.
3.31. Particularly, several submissions expressed concerns that the Proposed Acquisition would reduce access to and quality of services due to branch closures and a focus away from Queensland and regional areas.
3.32. Similarly, the Consumers’ Federation of Australia, Clyde and Gail Andrews, and Property Rights Australia raise concerns that the Proposed Acquisition would impact on access to services for vulnerable customers including those in regional areas which may have poor internet access, elderly customers and those experiencing vulnerability.
3.33. Other interested parties, including an insurer, supported or expressed neutral views regarding authorisation. Rabobank expressed neutral views.
3.34. Interested parties supporting the Proposed Acquisition generally note that the major banks are subject to existing competitive pressure driving investment in technology, and that the Proposed Acquisition would enable Suncorp Group to be more focused and fill unmet client needs in the insurance industry.
3.35. The ACCC also received a large number of submissions from consumers. These submissions raise a number of concerns regarding the conduct of ANZ and Suncorp Bank and the Proposed Acquisition, including general concerns about competition and regulation in the Australian banking industry, such as applicable dispute resolution obligations, potential branch closures, attrition of Suncorp Bank employees and potential price increases. Several of these submissions also raise concerns about how ANZ and Suncorp Bank have handled individual disputes with their customers.
(Footnotes omitted.)
A.4. The application for review
20 On 25 August 2023, each of ANZ and SGL (collectively, the applicants) filed applications in the Tribunal pursuant to s 101 of the CCA for a review of the ACCC’s determination made on 4 August 2023. On 29 August 2023, the Tribunal made directions for the two applications to be determined together. There is no reason to distinguish between the applications and, in these reasons, the two applications will be referred to as the “application”. The Tribunal also notes that whilst each of the applicants filed separate concise statements of facts, issues and contentions (SOFIC) and submissions, in substance, the applicants advanced their case collectively. Each addressed different aspects of the application and adopted the submissions made by the other. The Tribunal, therefore, has considered it appropriate to address the submissions made by each of ANZ and SGL below, collectively, as the applicants’ submissions.
21 On 29 August 2023, the Tribunal also made a direction pursuant to s 109(2) of the CCA permitting Bendigo to intervene in this proceeding.
22 On 27 November 2023, the Tribunal made a direction that the State of Queensland file and serve the submissions annexed to an affidavit of Michael John Kimmins dated 6 October 2023, and on 15 December 2023, the Tribunal made a direction pursuant to s 102(10)(d) of the CCA formally requesting that the State of Queensland provide those submissions, for the sole purpose of clarifying information provided to the ACCC in connection with the making of its determination.
23 The application for authorisation filed by ANZ is a merger authorisation within the meaning of the CCA.
24 The Proposed Acquisition is subject to three conditions precedent (a) approval by the Federal Treasurer under the Financial Sector (Shareholdings) Act 1988 (Cth), (b) a final determination by the Tribunal to authorise the Proposed Acquisition or a declaration made by the Federal Court of Australia that the Proposed Acquisition would not contravene s 50 of the CCA, subject to there being no lodgements of a relevant application for review of the declaration or a notice of appeal, and (c) the State Financial Institutions and Metway Merger Act 1996 (Qld) (Metway Merger Act) being either repealed or amended such that it does not apply to any holding company of Suncorp Bank or ANZ or its related bodies corporate, with reference to certain agreed amendments and agreed commitments to the Queensland government set out in Schedule 17 of the Share Sale and Purchase Agreement or as otherwise agreed between the parties and the Queensland government.
25 A review by the Tribunal of authorisation determinations made by the ACCC is governed by the provisions of Pt IX of the CCA. As explained in the Tribunal’s decision in Applications by Telstra Corporation Limited and TPG Telecom Limited [2023] ACompT 1 (Telstra/TPG (No 1)) at [9] (O’Bryan J, Dr J Walker and Ms D Eilert), a review of a merger authorisation under Pt IX differs from a review of other authorisations in two material ways:
(a) first, a review of a merger authorisation is required to be completed by the Tribunal within a statutory time period (whereas a review of other authorisations is not subject to any time limit); and
(b) second, a review of a merger authorisation is not a re-hearing of the matter (whereas a review of other authorisations is a re-hearing of the matter) and, correspondingly, restrictions are imposed on the information, documents and evidence to which the Tribunal may have regard in a review of a merger authorisation (whereas no such restrictions are imposed in a review of other authorisations).
26 In the usual course, the statutory time period for the completion of a review of a merger authorisation is within 90 days. However, under s 102(1AD) of the CCA, the Tribunal may determine in writing that the matter cannot be dealt with properly within the initial period, either because of its complexity or because of other special circumstances, and that an extended period applies for the review, which consists of the initial period and a further specified period of not more than 90 days.
27 On 29 August 2023, the Tribunal made a determination in writing to that effect such that the period of the present review is 180 days, which ends on 20 February 2024.
28 With respect to the information, documents and evidence to which the Tribunal may have regard in this review, and in accordance with s 102(9) and s 102(10) of the CCA, the Tribunal has only had regard to (a) information that was referred to in the ACCC’s Reasons for Determination, (b) information furnished, documents produced or evidence given to the ACCC in connection with the making of its determination, (c) supplementary information provided to the Tribunal by the parties that did not exist at the time that the ACCC published its Reasons for Determination, and (d) information provided by the State of Queensland pursuant to the request from the Tribunal, for the sole purpose of clarifying information provided to the ACCC in connection with it making its determination.
29 The information, documents and evidence given to the ACCC in connection with the making of its determination was extensive. The ACCC received more than 50 submissions, 27 witness statements, 12 expert reports, and commissioned a further three expert reports. The ACCC also used its compulsory evidence gathering powers to require the applicants and other third parties to provide information and documents, which culminated in the ACCC receiving more than 200,000 documents. The ACCC conducted compulsory examinations on a number of individuals.15
30 The evidence given to the ACCC included a number of witness statements of executives of ANZ, SGL, Suncorp Bank, and Bendigo and expert reports prepared on behalf of those parties. The Tribunal has found the witness statements to be of assistance in understanding the commercial context in which the Proposed Acquisition arose and the options available to ANZ, SGL, Suncorp Bank, and Bendigo, if the Proposed Acquisition does not proceed. The witnesses who gave statements are summarised below.
31 SGL relied on statements from the following lay witnesses:
(a) Steve Johnston, the CEO of SGL, dated 25 November 2022, 17 May 2023 (two statements), and 13 July 2023;
(b) Clive van Horen, CEO of Suncorp Bank, dated 25 November 2022, 17 May 2023, and 14 July 2023; and
(c) Adam Bennett, CIO of SGL, dated 16 May 2023.
32 ANZ relied on statements from the following lay witnesses:
(a) Adrian Went, the Group Treasurer of ANZ, dated 28 November 2022 and 17 May 2023;
(b) Shayne Elliott, the CEO of ANZ, dated 30 November 2022, 17 May 2023, and 30 June 2023;
(c) Douglas John Campbell, the General Manager, Home Loans Australia, in the Australia Retail Division at ANZ, dated 30 November 2022 and 17 May 2023;
(d) Isaac James Christian Rankin, the Managing Director of Commercial and Private Banking at ANZ, dated 30 November 2022;
(e) Yiken Yang, General Manager, Deposits, ANZ, dated 30 November 2022 and 17 May 2023;
(f) Mark Bennett, Head of Agribusiness, Australia Commercial Division ANZ, dated 1 December 2022, 17 May 2023, and 7 July 2023;
(g) Guy Samuel Mendelson, Managing Director, Business Owners Portfolio, Australia Commercial Division ANZ, dated 1 December 2022;
(h) Peter Dalton, Managing Director Designer and Delivery ANZx, dated 13 December 2022;
(i) Louise Claire Higgins, Managing Director, Suncorp Integration ANZ, dated 17 May 2023 and 17 July 2023; and
(j) James Anthony Lane, State Manager of Business Banking, Queensland, Australia Commercial Division ANZ, dated 17 July 2023.
33 Bendigo relied on a witness statement made by Cameron Telford Stewart, Head of Mergers and Acquisitions at Bendigo, dated 3 March 2023.
34 In the course of its assessment of the application for authorisation, the ACCC also examined a number of those witnesses and other executives of the parties pursuant to its powers under s 155 of the CCA. The persons from ANZ examined by the ACCC were Mr Elliott, Mr Campbell, and Mr Yang. The persons from Suncorp Bank examined were Dr van Horen, Mr Johnston (examined twice), and Dean Cleland, Executive General Manager, Business Banking. The persons from Bendigo examined were Marnie Baker, CEO and Managing Director of Bendigo, (examined twice) and Ryan Brosnahan, Bendigo’s Chief Transformation Officer.
35 SGL relied on the following expert reports:
(a) a report prepared by Dr David Howell, dated 15 May 2023, addressing the likely issuer credit rating for a merged Bendigo/Suncorp Bank; and
(b) reports prepared by Mozammel Ali, dated 17 May 2023 and 23 July 2023, addressing the funding costs and challenges of the Proposed Acquisition, the second report responded to reports prepared by Ms Starks that address advanced internal ratings-based accreditation issues arising out of the Proposed Acquisition.
36 ANZ relied on the following expert reports:
(a) reports prepared by Dr Jeffery Carmichael AO, dated 25 November 2022 and 13 May 2023, addressing the prudential public benefits and detriments arising out of the Proposed Acquisition;
(b) reports prepared by Dr Phillip Williams AM, dated 1 December 2022 and 19 May 2023, addressing the likely competitive effects of the Proposed Acquisition in relation to the supply of banking products in Australia, identification of the relevant product and geographic dimensions of the market or markets for commercial banking products, and whether the Proposed Acquisition is likely to substantially lessen competition in relation to the relevant market or markets; and
(c) reports prepared by Patrick Smith, dated 1 December 2022, 17 May 2023, and 17 July 2023, addressing whether the net cost savings described in the workbook titled ‘Synergies and one-off costs’ would be a public benefit of the Proposed Acquisition and whether the impacts of the funding costs identified in the witness statement of Mr Went would be considered a public benefit.
37 Bendigo relied on expert reports prepared by Professor Stephen King, dated 3 March 2023 and 28 June 2023, addressing whether the Proposed Acquisition would have the effect of substantially lessening competition in any market in Australia and whether public benefits would outweigh detriments as a result of the Proposed Acquisition.
38 The ACCC relied on the following expert reports:
(a) a report prepared by Professor Nicolas de Roos, dated 5 April 2023, addressing the concept of “coordinated effects” as it applies to the competition assessment of mergers and acquisitions in general. The report further sets out a high-level framework for assessing any change in the likelihood, extent, or severity and sustainability of coordinated effects arising out of the Proposed Acquisition compared with a counterfactual in which the Proposed Acquisition did not proceed;
(b) a report prepared by Mary Starks, dated 16 June 2023, in response to the expert reports of Dr Williams, and Dr Carmichael. The report also addresses the appropriate markets or areas of competitive overlap for analysing the competitive effects of the Proposed Acquisition and whether the Proposed Acquisition would have the effect of substantially lessening competition in the relevant areas of competitive overlap; and
(c) a supplementary expert report prepared by Ms Starks, dated 7 July 2023, addressing whether any of her conclusions in her first report are altered considering the additional information provided to her listed at Annexure A of the supplementary report.
39 In accordance with Direction 13 of the directions made on 29 August 2023, as varied by Direction 11 of the directions made on 20 October 2023, the parties have prepared and filed a joint document identifying all findings on factual matters set out in the ACCC’s Reasons for Determination that are not contested by the parties on this review (Agreed Factual Findings).
40 The Tribunal has adopted the Agreed Factual Findings for the purposes of making this determination.
41 The Tribunal has also received and had regard to (a) the SOFICs filed on behalf of each of the parties to this proceeding, (b) written submissions filed on behalf of each of the parties in advance of the hearing, and (c) oral submissions advanced on behalf of each of the parties during the hearing, together with a number of aide memoires and further written submissions provided to the Tribunal during the hearing.
B. STATUTORY FRAMEWORK FOR THE TRIBUNAL’S REVIEW
42 The grant of authorisations by the ACCC under Pt VII of the CCA and the Tribunal’s review of determinations by the ACCC in respect of authorisation applications, was amended significantly by the Competition and Consumer Amendment (Competition Policy Review) Act 2017 (Cth) (2017 Amendment Act): see Telstra/TPG (No 1); Applications by Telstra Corporation Limited and TPG Telecom Limited (No 2) [2023] ACompT 2 (Telstra/TPG (No 2)); see also the Explanatory Memorandum to the enacting bill, the Competition and Consumer Amendment (Competition Policy Review) Bill 2017 (2017 Explanatory Memorandum).
43 The amendments largely implemented the recommendations from the Competition Policy Review chaired by Professor Ian Harper (Harper Review). Consistently with the Harper Review’s two principal recommendations in respect of authorisations, notifications and class exemptions, the 2017 Amendment Act purported to simplify the authorisation regime such that (a) only a single authorisation application was required for a single business arrangement or transaction, and (b) the ACCC was newly empowered to grant authorisation on the basis that the conduct would not be likely to substantially lessen competition.
44 The effect and purpose of the statutory provisions and principles relevant to an authorisation application were largely not in dispute between the parties.
45 The following consideration of the statutory framework and principles governing the Tribunal’s review in this matter was determined in accordance with the opinion of the presidential member presiding: s 42(1) of the CCA.
46 Section 88 of the CCA empowers the ACCC to grant authorisations in respect of conduct that would or might otherwise contravene the provisions prohibiting restrictive trade practices under Pt IV of the CCA and is relevantly expressed in the following terms:
88 Commission may grant authorisations
Granting an authorisation
(1) Subject to this Part, the Commission may, on an application by a person, grant an authorisation to a person to engage in conduct, specified in the authorisation, to which one or more provisions of Part IV specified in the authorisation would or might apply.
Note: For an extended meaning of engaging in conduct, see subsection 4(2).
47 While the authorisation remains in force, the provisions in Pt IV of the CCA will not apply in respect of the conduct that is the subject of the authorisation to the extent it is engaged in by the applicant, any other person named or referred to in the application as a person who is engaged in, or who is proposed to be engaged in, the conduct and any other particular persons or classes of persons, as specified in the authorisation, who become engaged in the conduct: s 88(2) of the CCA.
48 In respect of an authorisation application, the ACCC shall make a determination in writing either granting such authorisation as it considers appropriate or dismissing the application: s 90(1) of the CCA.
49 By s 90(7) of the CCA, the statutory preconditions to the determination of an authorisation application are expressed as follows:
90 Determination of applications of authorisations
…
(7) The Commission must not make a determination granting an authorisation under section 88 in relation to conduct unless:
(a) the Commission is satisfied in all the circumstances that the conduct would not have the effect, or would not be likely to have the effect, of substantially lessening competition; or
(b) the Commission is satisfied in all the circumstances that:
(i) the conduct would result, or be likely to result, in a benefit to the public; and
(ii) the benefit would outweigh the detriment to the public that would result, or be likely to result, from the conduct; or
50 An “authorisation” is defined in s 4 of the CCA to mean an authorisation under Div 1 of Pt VII granted by the ACCC or by the Tribunal on a review of the ACCC’s determination.
51 A “merger authorisation” is defined in s 4 of the CCA in the following terms:
merger authorisation means an authorisation that:
(a) is an authorisation for a person to engage in conduct to which section 50 or 50A would or might apply; but
(b) is not an authorisation for a person to engage in conduct to which any provision of Part IV other than section 50 or 50A would or might apply.
52 The effect of assigning a specific definition to “merger authorisation” is to limit such authorisations to conduct to which only s 50 or s 50A of the CCA would or might apply.
B.3. The review of the ACCC’s determination
53 A person dissatisfied with a determination by the ACCC under Div 1 of Pt VII of the CCA, in relation to an application for an authorisation may apply to the Tribunal for a review of the determination: s 101(1)(a) of the CCA. On a review of the ACCC’s determination in relation to an application for authorisation, the Tribunal may make a determination affirming, setting aside or varying the ACCC’s determination and, for the purposes of the review, may perform all the functions and exercise all the powers of the ACCC: s 102(1)(a) of the CCA. A determination by the Tribunal pursuant to s 102(1) is taken to be a determination of the ACCC: s 102(2) of the CCA.
54 The Tribunal’s task on review is to “make its own findings of fact and reach its own decision as to whether authorisation should be granted or not, and if so, any conditions to which it is to be subject”: Application by Medicines Australia Inc [2007] ACompT 4; (2007) ATPR 42-164 (Medicines Australia) at [135] (French J, Mr GF Latta and Prof C Walsh).
55 The Tribunal’s task in respect of a merger authorisation application does not, however, entail “full merits review” or a “rehearing” because of the time and commercial sensitivities specific to such transactions: 2017 Explanatory Memorandum at [15.49]; see also Telstra/TPG (No 1) at [50]; Telstra/TPG (No 2) at [107]. This rationale is reflected in certain of the requirements for merger authorisations, brought into effect by the 2017 Amendment Act, including that (a) the Tribunal’s review of the ACCC determination must be completed within 90 days and may only be extended in certain circumstances: s 102(1AC) of the CCA, and (b) the Tribunal, in conducting its review, may only have regard to the material that is enumerated in s 102(10) of the CCA. The limitations on the information that the Tribunal may have regard to, in conducting its review, is intended to ensure that applicants to a merger authorisation “do not delay production” of relevant material to the Tribunal. This facilitates, in turn, the Tribunal in conducting its review “expeditiously”: 2017 Explanatory Memorandum at [9.80].
56 The Tribunal’s function is not to consider the correctness of the ACCC’s determination or whether it could have been better formulated: Medicines Australia at [138]; Application by Flexigroup Ltd (No 2) [2020] ACompT 2 at [135] (O’Bryan J, Dr J Walker and Ms D Eilert). The published reasons for the ACCC’s determination can, however, provide a “convenient reference point” for defining the matters in dispute: Telstra/TPG (No 2) at [108]; Re Herald & Weekly Times Ltd (on Behalf of the Media Council of Australia) (1978) 17 ALR 281 at 296 (Deane J, Mr J Shipton and Mr J Walker). Further, to the extent that the parties agree with factual findings made by the ACCC in the course of its determination, the Tribunal, ordinarily, need not examine those facts in detail: Telstra/TPG (No 2) at [108].
B.4. Statutory preconditions to authorisation
57 Section 90(7) of the CCA enumerates two tests for authorisation under s 90(7)(a) and s 90(7)(b). The tests for authorisation are expressed in the alternative, and, accordingly, the Tribunal need only be satisfied that the applicant has discharged its onus in respect one of the authorisation tests.
58 No standard of proof applicable in a civil litigation context, such as the balance of probabilities, applies to decisions bearing an administrative character, as in this case: Telstra/TPG (No 2) at [99] and the authorities cited therein. Hence, the only relevant precondition to granting authorisation is the existence of the Tribunal’s satisfaction of one of the conditions under s 90(7)(a) or s 90(7)(b). Put another way, the ACCC or the Tribunal on review must reach an “affirmative belief” as to one of the matters under s 90(7)(a) or s 90(7)(b): Telstra/TPG (No 2) at [99]. This requisite state of satisfaction is to be formed in good faith, reasonably and upon a correct understanding of the law: Wei v Minister for Immigration and Border Protection (2015) 257 CLR 22; [2015] HCA 51 at [33] (Gageler J, as his Honour then was and Keane J); O’Connor v Construction, Forestry, Mining, Maritime and Energy Union [2023] FCAFC 151 at [34] (Rangiah, Wheelahan and Raper JJ).
59 Both tests for authorisation require the ACCC or the Tribunal on review to compare a future with and without the conduct that is the subject of the authorisation application. In this regard, the tests for authorisation direct primary focus to the effects of the conduct for which authorisation is sought rather than the effects of conduct that is coincident with but not causally related to the conduct the subject of the authorisation application: Telstra/TPG (No 2) at [145]-[147]. For example, in respect of the test under s 90(7)(b), public benefits that are the result of other coincident conduct that is not the subject of the authorisation application may not be taken into account.
B.5. Section 90(7)(a) – the competition test
60 The first test for authorisation requires the ACCC or the Tribunal on review to be satisfied that the conduct would not have the effect, or would not be likely to have the effect, of substantially lessening competition: s 90(7)(a).
61 The application of the test under s 90(7)(a) ultimately requires the relevant decision maker to engage in the following stages of analysis:
(a) identification of the relevant markets, being the “field of actual and potential transactions between buyers and sellers amongst whom there can be strong substitution, at least in the long run, if given a sufficient price incentive”: Re Queensland Co-Operative Milling Association Ltd; Re Defiance Holdings Ltd (1976) 8 ALR 481 (QCMA) at 513 (Woodward J, JAF Shipton Esq and Prof MD Brunt); and
(b) a comparison between the nature and extent of competition in any market potentially affected by the proposed conduct in a future where the proposed conduct occurs and in a future without the proposed conduct: Telstra/TPG (No 2) at [117]; Australian Competition and Consumer Commission v Pacific National Pty Ltd (2020) 277 FCR 49; [2020] FCAFC 77 (Pacific National) at [103] (Middleton and O’Bryan JJ) and the authorities cited therein.
62 Substitutability is the essential concept for market definition. It is perhaps best explained by asking would there be a significant switch in demand or supply in response to a relatively small price increase (all other competitive variables being unchanged).16 In practice, this is reflected in the test known as the hypothetical monopolist, small but significant and non-transitory increase in price (SSNIP) test. It asks whether a small but significant non-transitory increase in price by a hypothetical monopolist would be defeated by demand side substitution (consumers switching to other products) and/or supply side substitution (other firms commencing to supply the product quickly and without significant investment).
63 Further, in addressing market definition it is necessary to distinguish between substitution in supply and new entry. Substitution in supply occurs when a firm is able to make a switch in production that is relatively rapid, using existing assets and without incurring significant expenditure. New entry is likely to take much longer and require significant expenditure on the acquisition of new assets. The likelihood of entry or expansion is taken into account in the second stage of the analysis.
64 The test under s 90(7)(a) is not applicable to authorisations in respect of cartel conduct (Div 1 of Pt IV), secondary boycotts (s 45D to s 45DB), contracts etc. affecting the supply or acquisition of goods or services (s 45E to s 45EA) and resale price maintenance (s 48): s 90(8) of the CCA. The purpose of this limitation was explained in the 2017 Explanatory Memorandum at [9.44]:
This avoids a mismatch between the basis on which the conduct is prohibited, which does not look to whether the purpose, effect or likely effect of the conduct is a substantial lessening of competition, and the basis on which authorisation for that conduct may be granted.
65 Section 90(7)(a) is framed in similar language to s 45, s 46, s 47 and s 50 under Pt IV of the CCA, except for the following key differences.
66 First, s 90(7)(a) is expressed as a negative proposition. Therefore, contrary to the application of the similarly framed provisions in Pt IV of the CCA, s 90(7)(a) requires the Tribunal to be satisfied that the conduct would not have the effect, or would not be likely to have the effect, of substantially lessening competition. It does not necessarily follow that if the Tribunal is not satisfied that the conduct would have the effect, or would be likely to have the effect, of substantially lessening competition that the Tribunal must be satisfied that the conduct would not have the effect, or likely effect, of substantially lessening competition. In some cases, the Tribunal may be satisfied that it can make a negative finding that it was not satisfied the conduct would be likely to substantially lessen competition but might have insufficient evidence to make a positive finding that it was satisfied the conduct would not be likely to substantially lessen competition.
67 Second, proof of contravention of similarly framed provisions in Pt IV of the CCA requires proof on the balance of probabilities. The evidentiary burden of proof applicable in civil litigation, is not applicable to an administrative decision of the ACCC or the Tribunal. The Tribunal’s task in respect of s 90(7)(a) is to reach a state of satisfaction or an “affirmative belief” that the conduct the subject of the authorisation application would not have the effect, or would not be likely to have the effect, of substantially lessening competition.
68 Notwithstanding these differences, the principles developed in respect of the meaning of the “likely effect of substantially lessening competition” as it appears in the language of s 45, s 46, s 47 and s 50 of the CCA, are directly instructive to the application of s 90(7)(a) of the CCA. The meaning of the “likely effect of substantially lessening competition” has been canvassed in a large body of case law and was most recently considered by the Full Court of the Federal Court in Pacific National and the Tribunal in Telstra/TPG (No 2).
69 In Telstra/TPG (No 2), the Tribunal summarised the meaning of “competition” at [112] to [114]. The Tribunal relevantly stated at [114], by reference to O’Bryan J’s decision in Australian Competition and Consumer Commission v BlueScope Steel Limited (No 5) [2022] FCA 1475 at [124]-[127], that:
[C]ompetition is best described by reference to its aim, mechanism and effect:
(a) The basic aim of business competition is to win sales – competitors strive to replace each other in the supply of products (whether goods or services) sought by customers.
(b) The key mechanism of competition is through substitution – to supply products to customers in place of another competitor’s supply. Substitution occurs on the demand side, whereby customers substitute one product or source of supply for another, and on the supply side, whereby suppliers adjust their production mix to substitute one product for another or one area of supply for another. Competitors strive to bring about substitution in a number of ways: through lowering their costs of production to enable them to profitably lower their prices; through improving the quality of their product and thereby increasing the value of the product to customers; and through inventing new products to meet the needs and wants of customers in new or better ways.
(c) As to effect, competition enhances the welfare of Australians by creating incentives and pressure for suppliers to reduce their costs of production and their prices (which, in the language of economics, is referred to as an improvement in productive efficiency), to commit resources to the production of goods and services most wanted by customers and to improve the quality of those products (which, in the language of economics, is referred to as an improvement in allocative efficiency) and to invest in innovation with the object of inventing new products to meet the needs and wants of customers (which, in the language of economics, is referred to as an improvement in dynamic efficiency).
70 Section 4G of the CCA provides further colour to the concept of lessening competition, provides that:
For the purposes of this Act, references to the lessening of competition shall be read as including references to preventing or hindering competition.
71 As to the meaning of “substantially”, Middleton and O’Bryan JJ stated in Pacific National at [104] that:
[T]he word does not connote a large or weighty lessening of competition, but one that is “real or of substance” and thereby meaningful and relevant to the competitive process.
(Emphasis added. Citations omitted.)
72 Critically, the meaning of “likely”, in the context of its application to an assessment of a substantial lessening of competition, has been the subject of significant scrutiny, judicially and before the Tribunal.
73 In Pacific National at [246] (Middleton and O’Bryan JJ), their Honours adopted the approach of French J in Australian Gas Light Co v Australian Competition and Consumer Commission (No 3) (2003) 137 FCR 317; [2003] FCA 1525 at [348] and stated that “likely” means a “real commercial likelihood”. The Full Court stated at [246] that the determination of whether an acquisition or merger is likely to have the effect of substantially lessening competition requires the application of the following approach:
(a) the application of s 50 requires a single evaluative judgment;
(b) it is a distraction (and, we would add, wrong) to ask what standards of proof apply to the primary facts which will involve predictions about the future;
(c) however, the degree of likelihood of any particular future fact existing or arising will be relevant to the assessment of the likely effect on competition of the acquisition.
(References to primary judgment omitted.)
74 This approach was recently affirmed by the Tribunal in Telstra/TPG (No 2) at [117].
B.6. Section 90(7)(b) – the net public benefits test
75 The second test for authorisation requires the ACCC or the Tribunal on review to be satisfied that the conduct would result, or be likely to result, in a benefit to the public and the benefit would outweigh the detriment to the public that would result, or be likely to result, from the conduct: s 90(7)(b). The form of this second test is consistent with tests previously contained in s 90: 2017 Explanatory Memorandum at [9.41]; Telstra/TPG (No 2) at [120]. The principles developed in respect of previous iterations of s 90(7)(b), therefore, are directly applicable to the present iteration of the test. The relevant principles are summarised as follows.
76 The relevant “public” to which the test is directed, is the Australian public: QCMA at 507-8; Medicines Australia at [107].
77 A “benefit to the public” has a wide import and will include “anything of value to the community generally, any contribution to the aims pursued by the society including as one of its principal elements (in the context of trade practices legislation) the achievement of the economic goals of efficiency and progress”: QCMA at 507-508; Telstra/TPG (No 2) at [121]. A “benefit to the public” can take into account both group interests and individual interests, to the extent that such individual interests are, in accordance with general societal views, worthy of inclusion and measurement: Re Qantas Airways Ltd [2004] ACompT 9 (Qantas Airways) at [187] (Goldberg J, Mr GF Latta and Prof DK Round). In this regard, cost savings or productive efficiency gains achieved by a private firm can constitute a “benefit to the public” in circumstances where it ultimately leads to “public” outcomes including a reduction in prices to final consumers or dividends to a range of shareholders. The weight to be given to such costs savings, however, depends on the extent to which they are passed through to consumers: Qantas Airways at [189]-[190].
78 A “detriment to the public” can include “any impairment to the community generally, any harm or damage to the aims pursued by the society including as one of its principal elements the achievement of the goal of economic efficiency”: Re 7-Eleven Stores Pty Ltd (1994) 16 ATPR 41-357 at 42,683; Medicines Australia at [108]. In many cases, the most important potential detriments will flow from the anti-competitive effect, which will result, or is likely to result, from the conduct the subject of the authorisation application: Medicines Australia at [108]. It may be the case that a purported benefit when “viewed in terms of its contribution to a socially useful competitive process” is in fact, to be judged as a detriment to the public: QCMA at 510.
79 The test under s 90(7)(b) requires the ACCC or the Tribunal on review to undertake a balancing exercise in order to determine whether benefits to the public ultimately outweigh detriments to the public: see Australian Competition and Consumer Commission v Australian Competition Tribunal [2017] FCAFC 150 (ACCC v ACT) at [7] (Besanko, Perram and Robertson JJ); QCMA at 506. Such a balancing exercise will involve what has been described as an “instinctive synthesis”, not least because many of the purported benefits and detriment will be “incommensurable and possibly unmeasurable”: ACCC v ACT at [7].
80 The weight to be given to a particular “benefit” requires an assessment of its nature, characterisation and the identity of the beneficiaries to it: Qantas Airways at [188]. This includes discerning “who takes advantage [of the benefit to the public] and the time period over which the benefits are received”: Qantas Airways at [189]; Telstra/TPG (No 2) at [122].
81 An applicant for authorisation is required to show that there is a factual basis for concluding that the purported benefits are likely to result, rather than to quantify in precise terms, the benefits: Qantas Airways at [201]; Telstra/TPG (No 2) at [125]. Relevantly, in Telstra/TPG (No 2), the Tribunal adopted and provided the following summary of further principles governing benefit analysis, which were set out in Qantas Airways at [203]-[209]:
(a) an accurate, objective quantification of public benefits is difficult, in part because benefits have to be estimated for some period in the future and so their magnitude becomes a matter not only of empirical estimation based on assumptions but also one of statistical likelihood;
(b) the nature of public benefits should be defined with some precision, a degree of precision which lies somewhere between quantification in numerical terms at one end of the spectrum and general statements about possible or likely benefits at the other end of the spectrum;
(c) any estimates involved in benefit analysis should be robust and commercially realistic, in the sense of being both significant and tangible;
(d) appropriate weighting will be given to future benefits not achievable in any other less anti‑competitive way, and so the options for achieving the claimed benefits should be explored and presented;
(e) the Tribunal is not assisted by fanciful and speculative modelling of benefits where the underlying assumptions are not clearly spelled out, where the estimates have not been subject to rigorous sensitivity analysis, and where the estimating process is not wholly transparent;
(f) while detailed quantification of benefits is the best option, quantification is not required by the CCA and benefits should be quantified only to the extent that the exercise enlightens the Tribunal more than the alternative of qualitative explanation; and
(g) where benefits cannot be quantified in monetary terms, they can still be claimed in qualitative terms.
82 The power conferred on the ACCC or the Tribunal on review to authorise conduct is discretionary. The statutory criteria under s 90(7) are only necessary conditions and, therefore, satisfaction of the criteria does not oblige the ACCC or the Tribunal on review to grant authorisation: Telstra/TPG (No 2) at [127]; Medicines Australia at [106]. Given the large variety of circumstances to which the discretion may be applied, it is not necessary nor desirable to define its outer limits, other than to say that it is not narrowly confined and will be informed by the objectives of the CCA: Medicines Australia at [126]; Water Conservation and Irrigation Commission (NSW) v Browning (1947) 74 CLR 492 at 505; Oshlack v Richmond River Council (1998) 193 CLR 72 at 84.
83 Typically, however, authorisation will be granted if the ACCC or the Tribunal on review are satisfied that the conduct the subject of the application, is likely to result in a net benefit to the public.
84 In circumstances where the ACCC or the Tribunal on review are not satisfied that the conduct would lead to a net public benefit, the Tribunal relevantly stated in Medicines Australia at [128] (prior to the 2017 Amendment Act):
Similarly, where the anti-competitive detriment is low to non-existent the ACCC may be entitled to say, as a matter of discretion, that it would only authorise the conduct if the public benefit to be derived from it, beyond that necessary to outweigh the anti-competitive detriment, or satisfy the per se conduct test is substantial. That is to say that the ACCC can require, in the proper exercise of its discretion, that the conduct yields some substantial measure of public benefit if it is to attract the ACCC’s official sanction. The Tribunal is in a similar position.
85 As set out above at [40], the Tribunal adopts the factual matters set out in the Agreed Factual Findings. At the time of the filing of the Agreed Factual Findings, Bendigo was not in a position to express a view on the correctness of certain of the factual matters, which were not contested by the applicants and are identified in Table 2 of the Agreed Factual Findings.
86 The Tribunal has examined the findings of fact made by the ACCC, having regard to evidence on which they were based, the submissions made by the parties about them, and the Agreed Factual Findings.
87 The following section sets out the relevant background and industry information in respect of the Australian banking industry, having particular regard to the positions of ANZ, Suncorp Bank and Bendigo.
88 Suppliers of services in the Australian banking industry are subject to significant regulatory and prudential oversight, reflecting the important role that the banking sector plays in the Australian economy. This section sets out the relevant background information in respect of that regulatory and prudential framework.
89 The Australian Prudential Regulation Authority (APRA), the prudential regulator of the Australian financial services industry, prescribes strict regulatory requirements in relation to obtaining and maintaining an ADI licence. Without an ADI licence, banks are unable to offer deposit products to customers.17 This is significant because deposits can be a comparatively cheap source of funding for lending activities compared to other funding sources.18
90 APRA provides two routes through which an entity can obtain an ADI licence: the “direct route” and the “restricted route”. Both routes are depicted at [4.28] of the ACCC’s Reasons for Determination, in the following figure referred to as “Figure 2”:19
91 An entity that holds an ADI licence, obtained through the direct route, can conduct the full range of banking activities, including deposit-taking, from the time it obtains the licence. To obtain a licence through the direct route, an entity is typically required to hold substantial capital resources at the point of application, or at least, a very clear avenue for access to such resources. APRA’s prudential requirements will also commence from the time the point of the licence.20
92 Alternatively, an entity that holds a restricted ADI licence, will be subject to “phased-in” regulatory obligations and in turn, only be permitted to conduct a restricted range of activities. An entity can apply for a restricted ADI licence before they are ready to be a fully licenced ADI and can do so through APRA’s Restricted ADI Licensing Framework.21
C.2.2. Calculating ongoing ADI capital requirements
93 An entity that is a fully licenced ADI must comply with all applicable prudential standards including requirements for financial resilience (such as minimum bank capital and liquidity requirements), governance, risk management, recovery and resolution, and reporting. The prudential standards establish minimum expectations for regulated entities, having regard to APRA’s purpose of ensuring that the financial interests of Australians are protected and the financial system is stable, competitive and efficient.21F22
94 APRA requires ADIs to hold a prudent minimum level of capital relative to a risk-adjusted measure of their assets, to ensure that banks are more likely to remain solvent during periods of financial adversity. Holding such a financial buffer increases the probability that ADIs have the capability to absorb unexpected losses, such as higher than usual defaults by borrowers on their loans.23
95 APRA permits two approaches for determining banks’ credit risk capital requirements: the internal ratings-based (IRB) approach and the standardised approach.24 Currently, each of the Major Banks, Macquarie and ING use the IRB approach. All other banks, including Suncorp Bank and Bendigo, use the standardised approach.25
96 Banks that are approved for the IRB approach can determine their capital requirements for credit risk using internal models that have been approved by APRA. To use the IRB approach, banks are required to hold extensive historical data, a sophisticated risk measurement framework, develop advanced internal modelling capabilities and undergo a rigorous accreditation process. Banks that use the IRB approach are also subject to more stringent regulatory requirements and more intensive ongoing supervision than banks using the standardised approach.26
97 In contrast, banks that employ the standardised approach apply risk weights, prescribed by APRA, for different types of lending. Standardised risk weights are intentionally simple and conservative. Consequently, banks that use the standardised approach may need to hold more capital against a similar exposure compared to banks that use the IRB approach.27
98 The capital framework is calibrated such that IRB capital requirements are, on average, lower than standardised capital requirements.28
99 The Tribunal notes that as capital is generally more expensive than other forms of funding, banks that use the IRB approach can typically provide lower cost offerings than banks that use the standardised approach. This gap has narrowed somewhat since (a) APRA implemented higher minimum capital requirements for IRB banks in 2017 (150 basis point increase, compared to a 50 basis point increase for standardised banks) in response to recommendations in the Financial System Inquiry Final Report in 2014 and (b) the implementation of APRA’s new capital framework on 1 January 2023.
100 The quality of banks’ capital has also improved. Common Equity Tier 1 (CET1) capital, the highest quality form of capital that is more expensive than other sources of funding and has a directly negative effect on banks’ return on equity (ROE), accounts for most of the rise in total capital, since it was introduced as a minimum requirement in 2013.29
C.2.3. Significant financial institutions
101 The Major Banks are labelled by APRA as domestic systemically important banks (D-SIBs). This is due to factors such as their size, interconnectedness, substitutability and complexity and that their distress or failure would cause more significant dislocation in the domestic financial system and economy than less systemically important institutions. Banks designated as D-SIBs must hold more CET1 capital to meet higher loss absorbency (HLA) requirements, which increases their ability to absorb losses and, therefore, reduces the probability of failure.30
102 The designation of entities as significant financial institutions (SFIs) was recently also introduced by APRA, which includes ADIs with assets above $20 billion and those determined by APRA due to factors such as complexity of operations and group membership. This definition allows APRA to differentiate consistently between prudential requirements for larger and smaller entities.31
103 Further, in 2017, the “Major Bank Levy” of 0.015%, paid each quarter on the balance of a bank’s liabilities (funding sources), was introduced for banks with over $100 billion in total liabilities (Major Bank Levy). Currently, the Major Banks and Macquarie are subject to the Major Bank Levy, which recognises that large leveraged banks are a source of systemic risk in the financial system and the wider economy. The Major Bank Levy is also intended to level the playing field for smaller ADIs and non-ADI competitors, relative to the Major Banks, “whose size and market dominance affords them significant funding cost advantages and pricing power at the expense of their customers”.32
C.2.4. Other regulatory requirements
104 ADIs are also subject to other regulatory requirements including anti-money laundering and counter-terrorism financing, and conduct regulations. The government agencies responsible for setting and enforcing regulations include the Australian Transaction Reports and Analysis Centre (AUSTRAC) and the Australian Securities and Investments Commission (ASIC).33 Obligations to which ADIs and financial service providers will be subject include, the requirement that to conduct a financial services business, an entity must hold an Australian Financial Services Licence and meet the conditions which attach to the holding of the licence.34 These additional regulatory obligations will impact each ADI’s compliance costs differently and relevantly, larger banks will have greater capability than smaller players to absorb fixed compliance costs.35
105 Non-bank lenders are required, in most cases, to hold an Australian Credit Licence (or operate as an authorised representative of an Australian Credit Licensee) and are regulated by ASIC. Non-bank lenders are also required to meet Anti-Money Laundering and Counter-Terrorism Financing requirements prescribed by AUSTRAC and entities that are registered financial corporations must report periodic data to APRA.36
106 There are several measures of a bank’s profitability. The most common measure is a bank’s ROE, which shows how efficiently the bank is using its equity capital to generate income.37 The Tribunal notes that net interest margin (NIM) is also a measure of a bank’s profitability, calculated by reference to its lending activities.
107 Banks generally accrue profits by lending at interest rates that are higher than the interest rates of their funding.38
108 Profitability, however, can be influenced by numerous factors, including the domestic interest rate environment, broader macroeconomic conditions, structural factors and the level of competition in each jurisdiction. The size, operating efficiency and lending profitability specific to individual banks will also impact on a bank’s ROE.39
109 To improve ROE, participants have an incentive to lower their cost-to-income (CI) ratio, which refers to the ratio of total operating costs (excluding bad and doubtful debt charges) to total income (the sum of net interest and non-interest income) and is a proxy for operational efficiency in banking.40
C.4. Price and non-price competition
110 Competition between banks can occur on price and non-price dimensions. The extent to which banks compete on both dimensions will depend on factors such as the relevant market in which they compete and the strategies that each bank employs.41
111 Price competition will occur across a range of pricing mechanisms and levers, depending on the specific product, the type of customer and what they value. For example, for loan products, which include home loans and business loans, pricing competition can occur across pricing levers such as headline interest rates, fees and charges, sign-on and cash-back bonuses and advertised and discretionary discounts.42
112 Non-price competition can occur where a business seeks to gain an advantage over competitors by differentiating the goods, services, and terms they offer to make them more attractive to buyers.43 There are also certain non-price dimensions that are particularly important to specific products and services. These include the relative speed of approval in home loans, customers’ access to and development of close relationships with specialist bankers that possess institutional and industry knowledge for agribusiness and to a lesser extent, SME banking and willingness to lend, particularly for bespoke SME business customer segments.44
C.5. Structural barriers to competition
113 Scale is important in banking because the provision of banking services involves significant fixed costs. These fixed costs include operating a branch network and maintaining head office functions, meeting regulatory compliance requirements and investing in technology to process transactions and otherwise serve customers.45
114 The Tribunal notes that in its view, as stated by [REDACTED] in an email to Mr Elliott, scale determines how these fixed costs are to be leveraged and absorbed over operational and customer bases.46 Scale is critical for banks to fund necessary investments in technology to enable them to compete effectively in the supply of banking services. The increased scale of the Major Banks compared with the regional banks enables them to spend significantly more on technology and provides them with a material competitive advantage in both price and non-price dimensions of competition.
115 Due to their larger scale, on average, the Major Banks benefit from [REDACTED].47 The Tribunal notes that, for the year ended December 2021, the Major Bank’s CI ratio was 51.9%,48 and the profit margin (net profit after tax as a proportion of total operating income) was 36.4%,49 compared with the other domestic banks’ CI ratio of 64.150 and profit margin of 25.8%.51
116 Deposits from households and businesses are the largest funding source for ADIs, accounting for approximately two-thirds of the Major Banks’ non-equity funding. Most ADIs’ deposit funding is “at-call”, meaning the depositor may withdraw funds in the short term and a substantial portion is available to be withdrawn at any time. A rising interest rate environment is generally associated with increased deposit funding costs for ADIs.52
117 ADIs also raise funds through issuing debt and securitisation in the wholesale debt market. Banks can issue short-term and long-term debt in the form of bonds and hybrid securities into the wholesale market.53
118 Smaller industry participants such as non-ADI lenders and smaller ADIs, typically source funding by using the securitisation warehouses from larger banks. As non-ADI lenders cannot raise funding from deposits, they are typically more reliant on securitisation to raise funds. Further, ADIs with a lower credit rating may have more limited access to unsecured wholesale funding markets compared with the Major Banks, and similarly may rely more on securitisation alongside their deposit funding.54
119 The Tribunal notes that deposits generally provide banks with the lowest costs of funding. A new entrant would be at a cost disadvantage if they had not first established a deposit base.
120 Further, the Major Banks and Macquarie benefit from higher credit ratings, compared with second-tier banks such as Suncorp Bank and Bendigo and other ADIs, which allows them to raise funds from wholesale debt markets at lower cost.55 Credit ratings are issued by credit rating agencies and are an important driver of a bank’s cost of and access to wholesale funding.56
121 Suncorp Bank currently receives the following long term credit ratings from the three main credit rating agencies:57
(a) S&P Global: A+ = strong chance that a borrower will meet their financial obligation, (positive outlook);
(b) Moody’s: A1 = low level of credit risk (positive outlook).
(c) Fitch: A = low risk of default, but slightly vulnerable to economic conditions on review for upgrade.
These ratings are higher than the ratings of other second-tier and regional banks because of a three-notch uplift received due to the support of SGL.58
122 The Tribunal notes that the level of switching in the Australian market for banking services will, to an extent, depend on product specific factors. In home loans, switching lenders occurs when a borrower repays their home loan with one lender (the previous lender) using the proceeds of a new home loan obtained from a different lender (the new lender), known as external refinancing.59 In retail deposits, it is common for consumers to hold multiple transaction or savings accounts across different banks, which is known as “multi-banking”. A consumer’s “main financial institution” (MFI), however, is often considered by banks to be the institution where the consumer holds their main transaction account and transacts with most frequently.60 Generally, the more price sensitive consumers are, the more likely they are to look for better deals and be motivated to switch for a better deal.61
123 The Tribunal notes the assessment by the Productivity Commission that the reasons why the banking sector has historically had a relatively low propensity for switching has been due to factors including:
(a) the costs and time involved in switching to a new provider;62
(b) varying degrees of financial literacy, levels of engagement, product complexity and information asymmetry that inhibit ready access to product information, an ability to assess the information and act on the information to make a decision to purchase or switch to a product;63
(c) cognitive and behavioural biases, causing customers to exhibit a preference for the status quo and undervalue the potential benefits of switching to a superior alternative offering.64
124 The Tribunal notes that it is important also to have regard to the distinction between “front book” pricing and “back book” pricing. Front book pricing refers to pricing of loans for new customers who benefit more from discretionary discounts, cash backs and introductory rates than existing customers who generally have higher back book pricing. The back book customers typically pay closer to the standard variable rate of interest on their home loans, which in turn is a reflection of consumer inertia and switching costs. Recently, there appears, however, to be an increased focus on back book loans. Mr Campbell, the General Manager of Home Loans in the Australian Retail Division of ANZ, gave evidence that brokers are regularly looking at their back books (customers for whom they have already arranged a loan) and where appropriate “instigate repricing activities on behalf of their customers”.65 He also gave evidence that as competition for back book customers has intensified, [REDACTED].66
C.5.5. Lack of brand recognition
125 Brand and trust are factors considered by customers when choosing who to deposit funds with.67 The Major Banks are more likely to benefit from brand recognition and have a larger share of MFI customers, who are considered valuable because they are likely to acquire more products.68
126 The nature of distribution channels dictates how customers engage with banks, and how competition takes place across the banking industry in practice.69 Banks provide products and services through a range of channels including: physical networks (such as branches, ATMs, and Bank@Post), digital channels (such as through websites and mobile apps) and intermediaries (such as brokers and relationship managers).70
127 The importance of access to branches and other physical networks as a facilitator of competition in the banking sector has declined since the expansion of online banking in the last two decades and other factors such as the increased penetration of brokers in some segments.71 Services traditionally offered by physical branches, such as opening or closing a bank account, submitting a home loan application, or customer service assistance are now increasingly offered by banks online through either web-based banking services, or through apps. A strong digital proposition is also considered increasingly important to business and SME customers. The use of digital payment methods is also increasing and notably, the Reserve Bank of Australia’s (RBA) 2022 Consumer Payments Survey found mobile payments were used by nearly two-thirds of Australians aged between 18 and 29, an increase from less than 20% in 2019.72
128 Branch access is, however, still required for certain banking transactions and activities and in person engagement is still highly valued by certain customer segments. For example, consumers who are not technologically literate, or able to use technology due to personal circumstances will continue to have a need to rely on access to branches.73
129 Nevertheless, due to the long-term decline in the importance of physical branches, financial institutions have reduced the number of bank branches and sought to develop alternative ways to deliver bank services by having several different points of presence.74 APRA data relevantly shows that the number of bank branches in Australia has reduced from 5,694 to 4,014 in the five years to June 2022.75
130 The increasingly important role for online banking as a distribution channel brings benefits to both consumers and the banks. For example, the cost of providing in-branch and call centre services can be multiples higher than for digital self-service for some customers’ banking needs.76 The benefits for consumers include faster loan application processing times, the reduced need to attend a branch for customer service assistance, reductions in search costs and an improved ability to compare products, thus facilitating switching.77
131 The increasing prominence of digital channels in the banking sector can prove challenging for banks with multiple legacy systems because undertaking digital transformation involves considerable financial investment and time. For example, ANZ has been undertaking a digital transformation program to improve its retail banking product offering and the underlying technology systems supporting these products. This includes ANZ launching an “ANZ Plus” product in March 2022, which currently includes a transaction account product and a savings account product on a mobile banking app.78
132 Further, smaller lenders, digital-only and recent entrant lenders without the requisite physical networks rely heavily on broker distribution networks and aggregator panels to gain new customers and increase scale.79
133 Brokers act as an intermediary by matching borrowers to lenders (and their loan products), assisting and advising borrowers on the loan application process and negotiating interest rates on loans.80 The Tribunal notes that brokers are under regulatory obligations to put the interests of their clients first.81
134 Aggregators, in turn, act as intermediaries between brokers and loan providers and provide the “panel” of lenders (and their associated products) that affiliated brokers choose from. Aggregators can determine which banks feature on their lending panels (typically Major Banks are strongly represented) and may provide a range of services to brokers including licensing, white-labelling services for loans, training and administrative support. Some brokers operate under an aggregator’s licence as representatives, others can only write loans from the panel lenders of the aggregator they are associated with.82 Broker distribution networks, therefore, provide lenders with access to a wider range of consumers than the direct lender channel.83
D. COMPETITION IN THE FUTURE WITHOUT THE PROPOSED ACQUISITION
135 The ACCC submits that the competition analysis in the present case requires consideration of two counterfactuals. It described the first as a No Sale counterfactual in which Suncorp Bank remains under the ownership of SGL. It described the second as a Bendigo Merger counterfactual in which there is a merger between Bendigo and Suncorp Bank.
136 The applicants submit that the only credible counterfactual is the No Sale counterfactual. They submit that the Bendigo Merger counterfactual is not a realistic possibility.
137 Both the ACCC and Bendigo contend that the Bendigo Merger counterfactual is a commercially realistic possibility in the future if the Proposed Acquisition does not proceed. They submit that (a) both Bendigo and Suncorp Bank have strong incentives to merge, (b) the merger would attract substantial scale, and efficiency benefits, (c) SGL wants to divest Suncorp Bank since it sees itself as a “pureplay” insurer, (d) a merger between Bendigo and Suncorp Bank would be value accretive to Bendigo and SGL shareholders, and (e) Bendigo has the capacity to make a compelling offer.
138 The ACCC and Bendigo submit that the likelihood of the Bendigo Merger counterfactual being a realistic commercial possibility is strengthened by the extent to which both Bendigo and SGL have invested significant resources into modelling such a merger scenario and Bendigo has actively pursued engagement with SGL in relation to a merger of Bendigo and Suncorp Bank.
139 All parties before the Tribunal accepted that, in the future without the Proposed Acquisition, the No Sale counterfactual, in which SGL retains Suncorp Bank, was commercially realistic.84
140 The ACCC, however, submits that the Tribunal should not consider the No Sale counterfactual on the basis that it is limited to a scenario, in which SGL retains Suncorp Bank, in its current form.
D.2.2. The applicants’ submissions
141 The applicants submit that while there may be a “prospect” that SGL would [REDACTED], the ACCC is wrong to suggest that the required transformational investment in home loans and SME would follow.85 The applicants submit that Dr van Horen’s evidence makes clear that [REDACTED] would only permit SGL to undertake the transformational technology investments if it is “earning the right with investors”.86
142 The applicants submit, that in any event (a) any [REDACTED] strategy is highly speculative because [REDACTED] is fraught with risk and involves several execution risks,87 (b) the evidence does not support any suggestion that [REDACTED] would confer any new found competitive strength on Suncorp Bank, (c) it is wholly speculative to presume that any [REDACTED], and (d) the interested party submission cited by the ACCC goes no further than suggesting that [REDACTED] may be a plausible counterfactual.88
143 The ACCC submits that Suncorp Bank’s current organic plans for Suncorp Bank are not exhaustive of the ways in which Suncorp Bank may be operated in the No Sale counterfactual. It submits that while Suncorp Bank may continue to execute its current organic plans, the material before the Tribunal suggests other commercially realistic possibilities. It submits that the Tribunal should consider, evaluate, and weigh those other possibilities when considering the No Sale counterfactual.89
144 The ACCC submits that one of these commercially realistic possibilities, is that SGL may:
[REDACTED]. 90
145 It submits that the above commercially realistic possibility was captured in a strategy document prepared by Boston Consulting Group titled “Suncorp Strategy 2025+: Horizons 2+3” (BCG paper). The BCG paper noted that SGL may be able to fund investment in processes and capabilities in Suncorp Bank’s home lending, if SGL [REDACTED].91
146 Further, the ACCC submits that [REDACTED] become more competitive [REDACTED]. The ACCC cites a submission from an interested party,92 which referred to the possibility of a counterfactual in which certain of Suncorp Bank’s portfolios might be acquired by a range of “second-tier competitors” that might then assist them to enhance their competitive advantages, deliver material public benefits and/or counter the concentration of the Major Banks in the relevant markets.93
147 The ACCC submits that if Suncorp Bank were to divest segments and use the proceeds from those sales to fund the “transformational” change it has long contemplated, there is a real prospect that Suncorp Bank would be a re-invigorated competitor in home lending and SME banking in the medium term. It submits that the [REDACTED] might likewise be more competitive in the hands of their new owners.94
148 The ACCC acknowledges that the competitive effects of these postulated [REDACTED] strategies cannot be assessed with any certainty but submits that:
In these circumstances, the nature of the test in s 90(7)(a) is particularly significant. While the Tribunal is likely not in a position to draw detailed conclusions about all the commercially realistic possibilities that arise in the No-Sale Counterfactual, the Tribunal can properly refuse authorisation if, when undertaking the single evaluative judgment, it is not affirmatively satisfied that the transaction will not likely substantially lessen competition in a market. 95
149 The Tribunal accepts that for the purposes of undertaking the required “single evaluative judgment”, it is necessary to have regard to all commercially realistic counterfactuals advanced by a party, but necessarily the weight that can be given to a counterfactual must depend on the Tribunal’s assessment of the degree of likelihood of it occurring.
150 The Tribunal is not persuaded that the possibility that SGL might improve its competitive position in the home loans and SME markets by using the proceeds of an assumed successful divestment of one or more of its portfolios, is sufficiently probable that it should be given any material weight for the purposes of considering the competition test in s 90(7)(a) of the CCA. Further, such a divestment would reduce the scale of an already small bank.
151 It would be an error of principle for the Tribunal to conclude that it could not be affirmatively satisfied that an acquisition would not have the effect, or would not be likely to have the effect of, substantially lessening competition on the basis that it was not able to assess, with any certainty, the competitive effects of a postulated but almost entirely theoretical counterfactual divorced from any probative evidentiary foundation. The evidentiary support advanced by the ACCC in support of the alternative counterfactuals does not rise above a single submission made to the ACCC, from another bank, which suggested that they might acquire any divested portfolios of Suncorp Bank, and a possible future strategy to be pursued by SGL, advanced in the BCG paper.
D.3. Bendigo Merger counterfactual
152 In order to determine the commercial likelihood of the Bendigo Merger counterfactual, it is necessary to consider the ability and incentives of Bendigo to make an offer for Suncorp Bank that SGL would be likely to accept. Those issues require an assessment of whether any likely offer would be value accretive for both Bendigo and SGL. The ability of Bendigo to make an offer that was sufficiently compelling for SGL to accept, largely turns on an assessment of the quantum and timing of the likely synergies that Bendigo might be able to realise from a merger with Suncorp Bank.
153 It was common ground that [REDACTED].96 [REDACTED].
154 The parties relied on conflicting contemporaneous assessments of the likely quantum and more significantly, the likely timing of the receipt of synergies that might by realised by a merged Bendigo/Suncorp Bank.
155 The assessments of likely synergies relied upon by the parties included, most relevantly:
(a) two assessments undertaken by [REDACTED] for Bendigo [REDACTED]97 and [REDACTED]98 [REDACTED];
(b) an assessment undertaken by Barrenjoey for SGL dated April 2022 (Barrenjoey 2022 Analysis);99 and
(c) a revised assessment undertaken by Barrenjoey for SGL dated May 2023 (Revised Barrenjoey Analysis).100
156 At the outset, the Tribunal notes the submission advanced by Bendigo and the ACCC, that the Tribunal should place limited weight on the Revised Barrenjoey Analysis.
157 Bendigo submits that the Revised Barrenjoey Analysis was prepared for the purpose of making a submission to the ACCC after it had issued its Statement of Preliminary Views (SOPV) and was created for a “self-serving” and “regulatory advocacy purpose”.101 Further, due to the confidential nature of the document, Bendigo’s legal representatives submit that they have been unable to seek instructions on it and the underlying modelling or analysis has not been made available.102
158 Similarly, the ACCC submits that the Tribunal should approach the Revised Barrenjoey Analysis with caution, given it was prepared in response to the SOPV, which had stated that a merger with a second-tier bank, such as Bendigo, was a commercially realistic prospect if the Proposed Acquisition did not proceed.103
159 The Tribunal accepts that generally business records prepared independently of the authorisation application to the ACCC, should be given greater weight. It is, however, always necessary to consider the extent to which those documents might have since been overtaken by other events, or may be coloured by specific objectives sought to be advanced by the authors of the documents.
160 The Tribunal accepts that it can readily be inferred, given its timing and its provision to the ACCC by SGL, that the Revised Barrenjoey Analysis was procured for the purpose of seeking to persuade the ACCC that the Bendigo Merger counterfactual would be inferior to the No Sale counterfactual. At the same time, the assumptions on which the Revised Barrenjoey Analysis was prepared were clearly exposed and the methodology used by Barrenjoey was, in substance, the same that Barrenjoey had employed for the purpose of preparing the Barrenjoey 2022 Analysis. The weight that the Tribunal can ultimately give to the Revised Barrenjoey Analysis, will turn upon the Tribunal’s assessment of the validity of the assumptions on which the report was based.
161 It is convenient to address first, the historical consideration by SGL of a potential divestiture of Suncorp Bank, and by Bendigo of a merger with Suncorp Bank, before considering the principal contentions of the parties, and then to address specific issues that would impact on the commercial likelihood of a merger between Bendigo and Suncorp Bank.
D.3.2. Historical consideration by Bendigo and SGL of potential merger/divestiture proposals
D.3.2.1. SGL consideration of divestiture of Suncorp Bank
162 [REDACTED].104 [REDACTED].105 [REDACTED].106
163 Internal documents, [REDACTED], show that SGL [REDACTED].107 [REDACTED].108 [REDACTED].109
164 [REDACTED], [REDACTED], [REDACTED], [REDACTED], [REDACTED]. [REDACTED], [REDACTED], [REDACTED], [REDACTED]. [REDACTED].110 [REDACTED], [REDACTED], [REDACTED], [REDACTED].111
165 [REDACTED], [REDACTED].112 [REDACTED], [REDACTED], [REDACTED], [REDACTED], [REDACTED], [REDACTED].113 [REDACTED], [REDACTED].114 [REDACTED], [REDACTED]. [REDACTED].115
166 [REDACTED], [REDACTED],116 [REDACTED].117
167 [REDACTED], [REDACTED],118 [REDACTED], [REDACTED], [REDACTED].119 [REDACTED], [REDACTED], [REDACTED]. [REDACTED], [REDACTED], [REDACTED]. [REDACTED], [REDACTED].120 [REDACTED].121
168 [REDACTED], [REDACTED].122
169 [REDACTED], [REDACTED], [REDACTED], [REDACTED].123
170 [REDACTED], [REDACTED].124 In a paper prepared for a SGL “Board Insights Session” in February 2022 (SGL February 2022 Analysis) it was stated that:
[REDACTED], [REDACTED]. [REDACTED]. [REDACTED], [REDACTED], [REDACTED].125
171 The SGL February 2022 Analysis concluded that on geography and diversification [REDACTED], [REDACTED], [REDACTED], [REDACTED]. [REDACTED], [REDACTED], [REDACTED].126
172 It is also to be noted that SGL’s February 2022 Analysis, [REDACTED].127
173 In April 2022, an internal SGL board paper provided [REDACTED]. [REDACTED](SGL April 2022 Analysis).128
174 The SGL April 2022 Analysis concluded that [REDACTED].129 The differential was $[REDACTED] to $[REDACTED] million,130 and was based on a [REDACTED].131 The SGL April 2022 Analysis noted that [REDACTED]. The SGL April 2022 Analysis also noted that [REDACTED].132 In stark contrast to the Revised Barrenjoey Analysis, SGL’s April 2022 analysis identified that [REDACTED].133
175 On 13 May 2022, SGL received a non-binding indicative offer from ANZ providing a headline purchase price for Suncorp Bank of $[REDACTED].134 Before entering into any engagement with ANZ, the SGL board had considered and then determined that SGL was not [REDACTED] and had, therefore, decided to enter into “[REDACTED]”.135
176 The valuation analysis in a 30 June 2022 SGL board paper (SGL June 2022 board paper) noted that (a) the ANZ offer [REDACTED], and would deliver total value to shareholders of $[REDACTED] billion, and (b) continued to show that a [REDACTED], achieving overall value in the order of $[REDACTED] billion, while the organic option to retain Suncorp Bank would [REDACTED] achieve a risk-adjusted fundamental value of $[REDACTED] (taken at the midpoint).136
177 The valuation analysis in the SGL June 2022 board paper was prepared by Barrenjoey (Barrenjoey Valuation Assessment). It concluded that there were the following alternative options to accepting a non-binding indicative offer of $4.9 billion from ANZ: (a) “[s]eek another cash buyer, [i]f not Copernicus [ANZ], could we sell the Bank to another major with superior price and terms?” and, more relevantly, (b) [REDACTED], [i]f not a cash sale, how does a [REDACTED] compare versus Copernicus’ offer or retaining the bank?”.137
178 The Barrenjoey Valuation Assessment provided the following analysis on the “current offer versus the alternatives” from a “shareholder valuation perspective”:138
Regional merger
A merger with a regional bank [REDACTED] will deliver synergies over time and also unlock the conglomerate discount unwind. However, any merger would involve dealing with the same threshold issues as we are facing under the cash sale (e.g. warranties and indemnities/brand licensing/non-compete).
Organic plan / retain bank
Within a range, the market is unlikely to give credit to Suncorp Bank materially divergent to regional bank peers (BEN/BOQ). This option forgoes the value of the conglomerate discount unwind.
179 The Barrenjoey Valuation Assessment calculated that the net value from a sale of Suncorp Bank, which represents the net present value of synergies realised over time (after deducting bank levy and integration costs), would be $[REDACTED] billion for [REDACTED], compared with $[REDACTED] billion for [REDACTED].139 In comparison to the net value figures presented in the SGL April 2022 analysis, it can be inferred that these figures are in the middle of the range indicated in the earlier SGL April 2022 Analysis, and that SGL had concluded that a regional merger would likely [REDACTED].
D.3.2.2. Bendigo consideration of a potential merger with Suncorp Bank
180 [REDACTED].140 [REDACTED].141
181 [REDACTED].142
182 [REDACTED].143 [REDACTED].144
183 [REDACTED].145 [REDACTED].146
184 [REDACTED].147 [REDACTED].148
185 [REDACTED] indicated that its plans for Suncorp Bank was [REDACTED].149
186 On 28 June 2022, Ms Hey sent an email to Ms McLoughlin, attaching a letter from Bendigo (28 June 2022 letter). [REDACTED].150 [REDACTED]. [REDACTED].151
187 It was also stated in the 28 June 2022 letter that Bendigo [REDACTED].152 [REDACTED].153
188 On 11 July 2022, Ms Hey sent an email to Ms McLoughlin [REDACTED].154 [REDACTED]:
[REDACTED].155
189 Following the announcement of the Proposed Acquisition, a meeting of Bendigo’s relevant sub-committee was held on 20 July 2022, [REDACTED].[REDACTED].156
190 On 9 August 2022, Bendigo sent a further letter to SGL expressing interest in acquiring Suncorp Bank (9 August 2022 letter). [REDACTED]. [REDACTED].157 [REDACTED], Ms Baker, CEO and Managing Director of Bendigo, [REDACTED].158
191 [REDACTED].159 [REDACTED]:
[REDACTED]. [REDACTED].160
D.3.3. Principal contentions of the parties
D.3.3.1. The applicants’ submissions
192 The applicants contend that it is not commercially realistic to conclude that Bendigo would acquire Suncorp Bank in a future without the Proposed Acquisition. Principally, they contend that this is because it is not credible that Bendigo would have the ability or incentive to make a reasonable offer for Suncorp Bank and it is not credible that SGL would be prepared to accept an offer from Bendigo.161
193 The applicants also rely on several additional matters, which they contend substantially diminish the prospect that a merged Bendigo/Suncorp Bank would be a successful entity, and, therefore, reinforce that a merger would be unlikely.162
D.3.3.1.1. Prospects of Bendigo making a credible offer for Suncorp Bank
194 The applicants rely on the following matters to support their contention that it is not credible that Bendigo could make a credible offer for Suncorp Bank.
195 First, if Bendigo made an offer for Suncorp Bank that reflected its value, the offer would be materially earnings per share (EPS) dilutive for Bendigo shareholders in the first three to five years.163 The applicants submit that there is no realistic prospect that Bendigo would make an offer to acquire Suncorp Bank if it was for a consideration that was materially EPS dilutive for Bendigo shareholders in the [REDACTED] years.164
196 The applicants submit that the Revised Barrenjoey Analysis, the most recent value analysis, establishes from a valuation perspective, that a sale of Suncorp Bank to Bendigo would be inferior to SGL retaining Suncorp Bank.165 The applicants submit that the Revised Barrenjoey Analysis was correct to assume delayed costs synergies,166 funding dis-synergies,167 the inclusion of the Major Bank Levy,168 capital dis-synergy, difficulty securing IRB accreditation for a merged Bendigo/Suncorp Bank169 and stranded costs.170 Each of these considerations and the relevant submissions by the parties are addressed further below.
197 Further, the applicants submit that using Bendigo’s relevant price-earnings ratio (P/E ratio), which implies that Suncorp Bank is valued at $[REDACTED] billion and assuming that synergies would be phased in during years [REDACTED], a scrip acquisition by Bendigo would not be EPS accretive for Bendigo shareholders for [REDACTED] years. They submit that the Revised Barrenjoey Analysis demonstrates that, even at a reduced price of $[REDACTED] billion, a merger would be EPS dilutive for the first [REDACTED] years.171
198 Second, the requirement for Bendigo to undertake a $[REDACTED] million equity capital raising to cover integration costs and to ensure sufficient capital in the combined Bendigo/Suncorp Bank entity would be a further material obstacle to any merger.172 The applicants submit that it is also likely, in circumstances where acquiring Suncorp Bank would be EPS dilutive for Bendigo shareholders for a number of years, that shares in the capital raising would be offered at a very substantial discount to the current market value, otherwise investors would be unlikely to participate.173
199 Third, given any potential offer by Bendigo to acquire Suncorp Bank [REDACTED], existing Bendigo shareholders would be required to approve the acquisition under ASX Listing Rule 7.1.174 The applicants submit that it is likely that an independent expert report would be provided to shareholders to assist them in assessing the merger175 and it is likely that in those circumstances, the report would present an unfavourable view of the merger by highlighting that the merger would be EPS dilutive. They submit that this further diminishes the prospect that Bendigo’s board would sanction an offer for Suncorp Bank and if it were to, it is unlikely to receive shareholder or market support.
200 Fourth, without due diligence, the prospect of any reasonable offer being made by Bendigo for Suncorp Bank is “entirely speculative”,176 a fact which Ms Baker accepted in her s 155 examination.177 The applicants submit that without having properly investigated and considered matters such as execution risks and Queensland legislative requirements, Bendigo cannot know whether it would be rational to pursue a merger with Suncorp Bank.178
201 Further, the applicants submit that the “speculative” nature of any merger between Suncorp Bank and Bendigo [REDACTED].179 [REDACTED] letter was “merely an invitation to treat”.180
D.3.3.1.2. Prospects of SGL accepting an offer from Bendigo for Suncorp Bank
202 The applicants advance the following reasons to support their contention that SGL is unlikely to accept an offer from Bendigo to acquire Suncorp Bank.
203 First, any material benefit that might be obtained from being a “pureplay” insurer from divesting Suncorp Bank and unwinding any conglomerate discount would need to be taken into account by the SGL board but ultimately any sale of Suncorp Bank would have to be value accretive.181 The applicants submit that similarly, any views expressed [REDACTED], must give way to whether any offer was value accretive for SGL shareholders.182
204 The applicants submit, put simply, that any offer for Suncorp Bank would need to stand on its own feet and exceed the value of Suncorp Bank under SGL’s organic plans for the bank, on the assumption that there was no divestment.183 They submit that [REDACTED]184 and that it [REDACTED].185
205 Second, any references to potential divestments of Suncorp Bank in internal SGL documents [REDACTED], should be read in the context of the critical qualification that [REDACTED].186 The applicants submit that this qualification is significant because it reinforces that the question of [REDACTED] value. They submit that this was explained by Mr Johnston, the CEO of SGL, in his first examination, during which he also stated that the market had changed since [REDACTED].187
D.3.3.1.3. Additional hurdles to acceptance of a Bendigo offer for Suncorp Bank
206 The applicants submit that while the valuation analysis is sufficient to conclude that SGL would be unlikely to accept any offer from Bendigo for Suncorp Bank, there are five additional matters that further diminish the prospect that SGL rationally would accept such an offer.188
207 First, if the Proposed Acquisition were not to proceed, SGL would need to [REDACTED].189 The applicants submit that this would be essential to ensure that Suncorp Bank can [REDACTED]. 190 They submit that it would be difficult for SGL, a listed corporation, [REDACTED].
208 Second, the fact that Mr Johnston, as CEO of SGL, has given non-confidential evidence that he is not satisfied that a merger of Suncorp Bank with Bendigo delivers value to SGL shareholders,191 would also make it difficult for SGL to reverse this position in the short to medium term.192
209 Third, it would not be feasible for SGL to put its staff through another uncertain merger process within the next five years because that would likely lead to significant staff attrition and adversely impact Suncorp Bank’s competitiveness.193
210 Fourth, there would be substantial execution and technology integration risks with a merger of Suncorp Bank and Bendigo that would be a further factor against agreeing to a sale.194
211 Fifth, Bendigo’s price to book ratio is ~0.7x relative to Suncorp Bank’s ~1.2x and has been lower than Suncorp Bank’s multiple for 15 years. The applicants submit that it would be irrational to merge an entity valued at 0.7x (and less then 1x) with an entity valued at 1.2x and hope that the combined book value would exceed 1.2x such that shareholders are no worse off.195
212 The applicants submit that the foregoing five matters, which essentially are certain, would weigh heavily against the merger’s uncertain hypothesised synergies.196
D.3.3.2. Bendigo’s submissions
213 Bendigo submits that in a future without the Proposed Acquisition, it would be commercially realistic for Bendigo to merge with Suncorp Bank. It points to three “undeniable” commercial propositions in support of this submission, in summary, (a) the merger would deliver substantial scale, (b) the merger would deliver substantial synergies, and (c) Bendigo and Suncorp Bank are “very complementary” businesses.197
214 Bendigo advances the following principal submissions in support of its contention that a merger with Suncorp Bank is commercially realistic.
215 First, a merger with Suncorp Bank was shown in a presentation to Bendigo’s board on 24 to 25 January 2022 [REDACTED].198
216 Second, it could acquire Suncorp Bank [REDACTED].199 [REDACTED].200
217 Third, if it [REDACTED].201 Bendigo submits that it is a well-managed bank with a record of strong financial performance and [REDACTED].202
218 Fourth, the history [REDACTED].203
219 Fifth, it remains committed to pursuing a merger with Suncorp Bank and it has developed further analysis, since the announcement of the Proposed Acquisition, which demonstrates that [REDACTED].204 Bendigo submits that it has also written to SGL [REDACTED].205
220 Sixth, [REDACTED].206 Bendigo submits that any unwinding of the discount would be additional value that could be extracted by SGL from a merger, but not essential to establishing a positive business case.207 It submits that notwithstanding any historical volatility of the conglomerate discount, [REDACTED]. It submits that [REDACTED].208
D.3.3.3. The ACCC’s submissions
221 The ACCC contends that the material before the Tribunal supports a conclusion that there is a realistic possibility that Suncorp Bank would be acquired by, or merged with Bendigo in the future, if the Proposed Acquisition does not proceed.209
222 The ACCC relies on the following matters in addition to those otherwise advanced by Bendigo in support of that contention.
223 First, the evidence indicates that should the Proposed Acquisition not proceed, the SGL board would evaluate any offer by Bendigo, or by anyone else, having regard to factors including the benefits of SGL focusing upon insurance and the potential unwinding of the conglomerate discount.210 The ACCC submits that one of SGL’s key rationales for selling Suncorp Bank is to allow for a “singular focus on its insurance business”,211 which is a rationale that continues to exist, independent of the identity of the purchaser.212
224 Second, as recently as April 2022, it appeared that SGL [REDACTED], should a cash sale to ANZ not proceed.213
225 Third, it can be accepted that SGL “are not sellers at any price”214 but, at the same time, SGL’s internal documents [REDACTED].215 The ACCC submits that there is, [REDACTED], a real prospect that SGL would do so, even if the price is less than what could be realised from the Proposed Acquisition.216
226 Fourth, the Tribunal should give more weight to the internal documents of both SGL and Bendigo when considering the likelihood of a merger between Suncorp Bank and Bendigo, than to evidence given since the Proposed Acquisition was finalised.217 The ACCC points to the observations in Telstra/TPG (No 2), that such documents are “a more reliable guide to the relevant commercial and economic considerations that will influence commercial decision-making in the future” than witness statements given by the applicants’ (or the intervenor’s) executives, whose views are “likely to be coloured, whether consciously or unconsciously, by the interest that their respective company has in the application for authorisation”: Telstra/TPG (No 2) at [482].
227 Fifth, the uncertainty around the quantum, timing and realisation of costs synergies is inherent in any merger and there is reason to believe that the net cost synergies of a merger between Suncorp Bank and Bendigo would be meaningful.218 Similarly, the ACCC submits that the evidence does not support a conclusion that the costs and timing risks of integration with Bendigo are prohibitive or materially greater than those that attend any integration of this nature.219
D.3.4. Realisation of cost synergies
228 The Revised Barrenjoey Analysis assumed that a merged Bendigo/Suncorp Bank could realise full run-rate synergies of $[REDACTED] million220 in contrast to assumed figures of between $[REDACTED] million to $[REDACTED] million in the Barrenjoey 2022 Analysis and $[REDACTED]221 [REDACTED].
229 The Revised Barrenjoey Analysis was prepared on the assumption that the synergies would be phased in over years [REDACTED].222 In contrast, the Barrenjoey 2022 Analysis proceeded on the basis that [REDACTED]% of the synergies would be achieved by year [REDACTED] and [REDACTED]% of the synergies by year [REDACTED].223 The graphs in the [REDACTED], appeared to proceed on the assumption that the synergies would be achieved within [REDACTED].224
230 As submitted by the applicants above at [197], the Revised Barrenjoey Analysis was prepared on the basis that Bendigo’s P/E ratio at that time implied that Suncorp Bank had a value of $[REDACTED] billion.225 The Bendigo ratio was used because Suncorp Bank did not have a P/E ratio, given it is part of the wider SGL group.226
D.3.4.2. The applicants’ submissions
231 The applicants submit, for the following reasons, that realisation of cost synergies in a merger between Suncorp Bank and Bendigo would be significantly delayed.
232 First, the phasing assumption in the Revised Barrenjoey Analysis is far more reasonable than the phasing assumption in the Barrenjoey 2022 Analysis, because events since 2022 show that it is unlikely any synergies would be achieved within [REDACTED] years.227 The applicants submit that the need for Bendigo to negotiate the repeal or amendment of the Metway Merger Act, as it applies to Suncorp Bank, would likely mean that cost synergies would commence in FY[REDACTED] and would not be fully realised until FY[REDACTED].228
233 Section 64 of the Metway Merger Act presently requires Suncorp Bank to be headquartered and locate its key banking functions in Queensland. The applicants submit following the announcement of the Proposed Acquisition in mid-2022, negotiations between ANZ and the Queensland government to repeal or amend the Metway Merger Act, as it applies to Suncorp Bank, took almost [REDACTED]. They submit that ANZ and SGL agreed to a significant package of financial investments and ANZ guaranteed no net job loses to Queensland across ANZ and Suncorp Bank for three years.229
234 Further, the applicants submit almost half of the synergies anticipated by [REDACTED] in the [REDACTED], were derived from approximately [REDACTED].230
235 Second, Bendigo’s view that SGL would offer the same commitments to the State of Queensland, as it has under the Proposed Acquisition, is “misconceived”.231 The applicants submit that a merger with Bendigo, particularly one [REDACTED], would not be one which would justify SGL committing to provide the same package of financial investments to the State of Queensland.232 They submit that it is, therefore, likely that Bendigo would need to offer even more significant commitments than ANZ because the funding of the commitments was only possible because of the value arising to SGL and its shareholders from the Proposed Acquisition.233
236 Third, Bendigo’s view that it could integrate Suncorp Bank into Bendigo within [REDACTED] years is “speculative and ambitious”, and, at best, likely to be wrong, which further supports the view that realisation of cost synergies would be delayed.234 The applicants point in this regard, to the evidence of Mr Brosnahan, Bendigo’s Chief Transformation Officer, in which he agreed that any estimates about integration timing was based on assumptions around the complexity of the integration and would not be known until due diligence occurs and Bendigo obtains access to Suncorp Bank’s systems.235 They note that Dr van Horen, the CEO of Suncorp Bank, has also given unchallenged evidence that he believes Bendigo will likely experience difficulty and delay integrating Suncorp Bank, which they submit, should carry significant weight because of his extensive experience in major bank integrations.236
237 The applicants also submit that Bendigo’s view that it could integrate Suncorp Bank within [REDACTED] years is also inconsistent with its past experience integrating Adelaide Bank following its acquisition in 2007, Delphi following its acquisition in 2012 and Rural Bank before 2018.237 They submit that Bendigo did not begin planning the integration of those banks until 2019,238 integration did not commence until 2020239 and integration will not be completed until [REDACTED].240
238 In addition, the applicants submit that the legacy technology issues that presently affect both Bendigo and Suncorp Bank, provide further support for their contention that it is unlikely Bendigo could integrate Suncorp Bank within [REDACTED] years. In particular, they submit, Suncorp Bank has [REDACTED] to transition a legacy smartphone banking application to the current iteration of the smartphone application.240F241
239 Fourth, if the Tribunal accepts that cost synergies would not be realised within [REDACTED] years, the applicants point to Ms Baker’s acceptance that taking [REDACTED] years to become value accretive, generally would [REDACTED].241F242 The applicants submit that Ms Baker’s evidence that she would be [REDACTED], should be given little weight.242F243 They submit that (a) there is no contemporaneous documentary evidence that Bendigo’s management would recommend, or its Board would accept, making an offer to SGL in those circumstances, (b) [REDACTED],243F244 and (c) it is rational to expect that [REDACTED], Bendigo would not have viewed acquiring Suncorp Bank favourably.
240 Further, and relatedly, the applicants submit that Ms Baker has said clearly and unequivocally that [REDACTED],244F245 [REDACTED].245F246
D.3.4.3. Bendigo’s submissions
241 Bendigo submits that the assumption that no cost synergies would be realised until year [REDACTED] after a merger, as relied upon in the Revised Barrenjoey Analysis, cannot be accepted for the following reasons.
242 First, it took ANZ and Suncorp Bank “just 7 months” to reach an agreement with the State of Queensland in relation to the application of the Metway Merger Act to Suncorp Bank.246F247
243 Second, s 64 of the Metway Merger Act only imposes requirements on the domicile of the head office, managing director and the offices of certain company personnel and principal company services. They do not apply to branch offices and general branch staff.247F248
244 Third, Ms Baker accepted that if Bendigo and SGL were required to offer commitments to the Queensland government that were similar to those offered by ANZ and SGL, [REDACTED], [REDACTED].248F249
245 Fourth, SGL makes the speculative assertion that Bendigo would need to offer even more financially significant commitments to Queensland than ANZ because SGL’s commitments were possible only because of the value to SGL arising from the Proposed Acquisition but fails to acknowledge that at least some of its Queensland commitments are pre-existing promises, including its financial commitments to the Disaster Response Centre.249F250 In any event, Bendigo submits that there is no logical or established connection between the extent of any commitments that Bendigo might have to give to the Queensland government and the timing for realising cost synergies.250F251
246 Fifth, SGL overstates the impact of technology integration on the realisation of synergies. Bendigo submits that Mr Brosnahan’s evidence is to the effect that integration between Bendigo and Suncorp Bank’s technology platforms would take [REDACTED] years and [REDACTED].252 It submits that Mr Brosnahan’s view is that Bendigo has a team that is “match fit” to undertake integration of Suncorp Bank because it has been undertaking similar integrations for past acquisitions over the last two years. It submits that it is also motivated to complete the integration in [REDACTED] years, [REDACTED].253 It also submits that even if completion of Suncorp Bank integration did not occur until [REDACTED],254 this would not mean that synergies could not begin to be realised earlier.
247 Sixth, Bendigo submits that it has completed consolidation of four of its seven banking platforms and is progressing with its transition to cloud-based applications.255 Bendigo submits that its remaining three banking platforms will be consolidated into the core banking platform in the next 18 months to two years, and its modernisation program is being completed in parallel.256 It submits also that (a) any integration with Suncorp Bank would not need to be [REDACTED], (b) it does not need to undertake any significant changes to its core systems to accommodate Suncorp Bank’s customer base, particularly because [REDACTED],257 and (c) therefore, it is unlikely to be any significant delay in the realisation of technology-based synergies, as otherwise suggested by the applicants.
248 Seventh, the applicants’ suggestion that Bendigo would have to inject $[REDACTED] million of CET1 capital to maintain the same level of excess capital that it currently maintains on the standardised approach is misconceived. Bendigo refers to what it understood to be an observation by Ms Starks in her supplementary report that SGL has been informed by APRA that [REDACTED].258 It submits that when Suncorp Bank [REDACTED].259
249 The Tribunal interpolates at this point to note that Ms Starks was referring to the concept of “excessive procyclicality”. A probability of default model that exhibits excessive procyclicality is a model that is sensitive, not insensitive, to the economic cycle because it produces probability of default estimates based on a “Point in Time” model. By way of contrast, a probability of default model that uses a “Through The Cycle” model is insensitive to the economic cycle.260
250 APRA recommends that ADIs should not use a probability of default model with excessive procyclicality and rather should use a model that strikes an appropriate balance between excessive procyclicality and risk sensitivity.261 In this case, APRA’s feedback to SGL was that it should [REDACTED].262
251 The Revised Barrenjoey Analysis predicts the net present value of total dis-synergies from increased funding costs if Suncorp Bank were to merge with Bendigo would be $[REDACTED] million263 (based on an initial annual figure of $[REDACTED] million increasing to $[REDACTED] million per annum). By contrast, the Barrenjoey 2022 Analysis assumed a net present value of cost dis-synergies of only $[REDACTED] million.264 The reason for the difference is that it was assumed in the Revised Barrenjoey Analysis that Suncorp Bank’s credit rating would [REDACTED] (without SGL group support), whereas the Barrenjoey 2022 Analysis assumed a rating [REDACTED] for the merged Bendigo/Suncorp Bank to [REDACTED].265
252 [REDACTED].266
253 [REDACTED].267
D.3.5.2. The applicants’ submissions
254 The applicants submit that the assumption as to Suncorp Bank’s credit rating used in the Revised Barrenjoey Analysis is correct whilst the corresponding assumptions in the earlier Barrenjoey 2022 Analysis and [REDACTED] are incorrect, for the following reasons.
255 First, there is unchallenged evidence from S&P Global itself that Suncorp Bank under Bendigo ownership would [REDACTED].268
256 Second, there is unchallenged evidence from Dr Howell, who has over 20 years’ expertise in credit ratings, that a merged Bendigo/Suncorp Bank would have a S&P Global rating of [REDACTED] ([REDACTED]), a Fitch rating of [REDACTED] and a Moody’s rating of [REDACTED].269 The applicants submit that Ms Baker’s evidence that [REDACTED] should be given little weight because it is “triple hearsay” and lacks any documentary support.270
257 Third, according to the expert opinion of Mr Ali, who has over 28 years’ experience in debt, equity, hybrid and other structured capital raising, the debt pricing impact of [REDACTED], would be marginal.271 The applicants submit that Ms Baker’s conflicting evidence that a [REDACTED], should not be given any weight,272 particularly because she is not a debt capital raising expert and, therefore, not qualified to express such an opinion.
258 Fourth, the assumption in the Revised Barrenjoey Analysis that a merged Bendigo/Suncorp Bank would incur increased annual funding costs of $[REDACTED] million in year 1, increasing to $[REDACTED] million by year 5, is consistent with and supported by Mr Ali’s evidence on this issue. Mr Ali gave evidence that based on Suncorp Bank’s forecast balance sheet for FY25 the pre-tax incremental cost of wholesale funding, relative to Suncorp Bank under SGL ownership would range between $[REDACTED] million and $[REDACTED] million per year.273
259 Relatedly, the applicants submit by reference to a report SGL obtained from S&P Global, that a merged Bendigo/Suncorp Bank [REDACTED].274
260 Fifth, moving Suncorp Bank to Bendigo ownership would add a further impact on valuation in the range of $[REDACTED] million to $[REDACTED] million, based on Mr Ali’s estimates. The applicants submit that Mr Ali’s estimates are consistent with the Revised Barrenjoey Analysis which assumed the valuation impact of the funding cost dis-synergies to be in the sum of $[REDACTED] million.275
261 Sixth, the Revised Barrenjoey Analysis assumed the Major Bank Levy would have to be paid, but the [REDACTED] did not. The Major Bank Levy of 0.015% is required to be paid quarterly on the balance on the bank’s liabilities for banks with liabilities over $100 billion.276 The applicants submit that there is little doubt that a merged Bendigo/Suncorp Bank would be required to pay this so Barrenjoey were correct to factor it into their analysis.277
D.3.5.3. Bendigo’s submissions
262 Bendigo submits that the Tribunal should not accept the applicants’ contention that the [REDACTED] were deficient because they failed to assume [REDACTED], for the following reasons.
263 First, [REDACTED].278 Bendigo submits that the only analysis which in fact did not account for the rating [REDACTED] for a merged Bendigo/Suncorp Bank, was the Barrenjoey 2022 Analysis.279
264 Second, and relatedly, the Revised Barrenjoey Analysis, which incorporates the dis-synergy associated with a ratings downgrade, is an inflated valuation. Bendigo points to the evidence of SGL’s own expert, Mr Ali, who suggested that the annualised pre-tax incremental funding cost increase in a merged Bendigo/Suncorp Bank, compared with a scenario whereby SGL retains Suncorp Bank, is in the range of $[REDACTED] million to $[REDACTED] million per annum, over [REDACTED] months.280 Bendigo submits that on Mr Ali’s calculations, the total dis-synergy, would be approximately $[REDACTED] million. It submits that the total dis-synergy from increased funding costs of $[REDACTED] million, that includes a dis-synergy rising to $[REDACTED] million in year five, assumed in the Revised Barrenjoey Analysis,281 is well beyond anything supported by Mr Ali’s evidence.
265 Third, contrary to the applicants’ submissions, the [REDACTED].282 [REDACTED]283 but if it had been included, Bendigo submits that it would have nevertheless shown that a merged Bendigo/Suncorp Bank was value accretive. It submits that [REDACTED].284
266 Fourth, it is reasonable not to include dis-synergies arising from both the Major Bank Levy and a 3-notch discount on a merged Bendigo/Suncorp Bank’s credit rating. Bendigo submits that this is because banks subject to the Major Bank Levy, including Macquarie, receive a 2-notch upgrade on their credit ratings due to implicit government support.285 It also submits that [REDACTED] return on tangible equity (RoTE).286
D.3.6. Capital dis-synergy and IRB status
267 The analyses prepared for SGL consistently assumed, even before the Proposed Acquisition, that a merged Bendigo/Suncorp Bank would incur a small capital dis-synergy.287 By contrast, [REDACTED].288
268 APRA accepts that IRB accreditation allows banks to maintain headline risk-weights 14% lower than in the standardised approach and, as Ms Baker notes, this allows IRB banks to price more competitively and win more business within existing markets.289
D.3.6.2. The applicants’ submissions
269 The applicants submit that the assumption that Bendigo would achieve IRB status and an associated capital benefit by acquiring Suncorp Bank is unreasonable, and should be rejected, for the following reasons.
270 First, the possibility that Bendigo would achieve IRB status is speculative, and if it does happen, it is unlikely to be quick. In December 2021, Bendigo was informed by APRA that [REDACTED].290 Mr Ali has also expressed the view that based on his experience advising APRA regulated clients, Bendigo could only achieve IRB accreditation, at best, in two to three years. 291
271 Second, [REDACTED],292 [REDACTED]. The applicants point by way of example to a [REDACTED].293
272 The applicants submit that [REDACTED].294 The applicants submit that no weight should be placed on Ms Starks’ suggestion that if SGL were to incorporate more procyclicality, it may obtain a day-one capital benefit,295 particularly because Ms Starks has no relevant banking expertise. In any event, they submit that Mr Ali observes that increased procyclicality in IRB credit risk models increases volatility under stress conditions which banks the size of Suncorp Bank or a combined Bendigo/Suncorp Bank have less ability to absorb.296
273 Third, Bendigo’s acquisition of Suncorp Bank would not make achieving IRB status any more likely and it is, therefore, inappropriate to include it as a benefit of the merger in the valuation analysis.297 They point to Mr Ali’s evidence that there is no minimum size requirement for achieving IRB accreditation and Bendigo’s acquisition of Suncorp Bank may even make it more difficult to obtain298 and Dr van Horen’s evidence that increased scale would not increase the likelihood of obtaining IRB status, and on the contrary, the increased complexity of the combined bank would reduce its likelihood.299
D.3.6.3. Bendigo’s submissions
274 Bendigo submits that the prospect of a merged Bendigo/Suncorp Bank obtaining IRB status [REDACTED], but it would be likely to occur, for the following reasons.
275 First, SGL is wrong to suggest that [REDACTED].300 [REDACTED],301 [REDACTED].302 It submits that although scale alone does not trigger IRB accreditation, as Ms Starks observes, “[g]iven the costs and data requirements for developing an IRB model, smaller banks tend to be [standardised approach] banks and larger banks tend to be IRB banks.”303 It notes that aside from the Major Banks, only Macquarie and ING currently use IRB models.304
276 Second, a merged Bendigo/Suncorp Bank would overtake Macquarie in scale, suggesting that the IRB accreditation would be suitable and more likely. As the ACCC concluded, “a merged BEN/Suncorp Bank entity would have the benefit of increased scale, and could potentially achieve greater benefits from having a larger, more diverse portfolio of loans, allowing better, more granular modelling of risk”.305
277 Third, [REDACTED] Interest Rate Risk in the Banking Book (IRRBB) [REDACTED].306 The Tribunal notes that it is necessary to have both IRRBB and IRB (credit risk) to achieve IRB accreditation, [REDACTED].307 However, Bendigo submits that [REDACTED],308 [REDACTED].309 Bendigo also points to Ms Baker’s evidence that [REDACTED].310
D.3.7. Capital raising and integration costs
278 The Revised Barrenjoey Analysis suggests that a Bendigo/Suncorp Bank merger would not be value accretive for Bendigo shareholders, in part, because Bendigo would need to undertake a $[REDACTED] million capital raising to cover integration costs and increased capital requirements.311 This would be made up of $[REDACTED] million in integration costs, and $[REDACTED] million in equity to increase the merged entity’s CET1 capital ratio to [REDACTED]%.312
279 Mr Johnston stated that in his view, the combined entity’s CET1 capital ratio would fall below [REDACTED]% in [REDACTED] and that [REDACTED].313
D.3.7.2. The applicants’ submissions
280 The applicants submit that Mr Johnston’s view that Bendigo would need to undertake a $[REDACTED] million equity capital raising to cover integration costs and to ensure sufficient capital in the combined entity [REDACTED].314 They submit that the other integration cost component to be raised would be approximately $[REDACTED] million.315 They submit that [REDACTED].316
281 The applicants submit that there does not appear to be much dispute that integration costs would be around this level. They refer to SGL’s “insights” analysis in April 2022, prior to any offer from ANZ, that estimated pre-tax integration costs would exceed $[REDACTED] million,317 and [REDACTED].318 In addition, they submit that if Bendigo were to include a cash component in any offer, this amount would also need to be raised in equity capital markets.319
282 Further, the applicants submit that consistently with market practice an independent expert report would be commissioned and provided to shareholders.320 They contend that the report would highlight that the proposed offer for Bendigo Bank would be earnings dilutive and therefore Bendigo shareholder approval would be unlikely to be obtained.321
D.3.7.3. Bendigo’s submissions
283 Bendigo submits that the integration costs of a merger with Suncorp Bank are likely to be much lower than the $[REDACTED] million assumed in the Revised Barrenjoey Analysis322 and SGL’s analysis as late as April 2022 suggested that SGL would share the integration costs with Bendigo.323
284 Bendigo does not dispute that the post-merger CET1 ratio would likely be [REDACTED]%,324 but submits [REDACTED].325
285 Bendigo also submits that it is presently in a strong financial position. It submits that in the half-year to December 2022, its cash earnings were up 22.9% to $294.7 million, its customer volume was up by 5%, its CET1 capital was up 45 basis points to 10.13% and its NIM was up 19 basis points to 1.88%.326 It submits that it has strong funding and liquidity, with customer deposits comprising 73.9% of its total funding base,327 and a household deposit/loan ratio that significantly exceeds industry levels.328
286 Each of the applicants, the ACCC and Bendigo framed their consideration of the Bendigo Merger counterfactual by first addressing whether it would be a realistic commercial possibility. Before considering the parties contentions on this issue, it is necessary to say something further about the correct approach to the evaluation of postulated counterfactuals for the purposes of s 90(7)(a) of the CCA. In order to undertake the necessary forward looking “with and without” analysis, it is necessary to identify at least one counterfactual. In many cases the most likely counterfactual in a merger/acquisition context would be a “no sale” counterfactual. In other cases, there may be varying alternatives including both a “no sale” counterfactual and “an alternative merger counterfactual”, with one or more alternative entities.
287 Prior to any application to the Tribunal for the review of a decision of the ACCC not to grant authorisation for a proposed acquisition, the ACCC, the applicants and any intervenor would necessarily have had to identify and evaluate the likelihood of counterfactuals. In the usual course, the likelihood of those counterfactuals would have been the subject of extensive lay evidence and expert reports. Both the lay and expert evidence would be directed to the specific counterfactuals postulated by the parties.
288 It is in that context, that the Tribunal must then identify a counterfactual, or counterfactuals, for the purposes of undertaking its task pursuant to s 90(7)(a) of the CCA. By reason of s 102(10) of the CCA, the Tribunal, subject to the limited exceptions in s 102(9) of the CCA, is limited to the evidence before the ACCC. Hence, except in perhaps the most exceptional circumstances, the Tribunal’s identification of counterfactuals is framed by the parties and in particular, the counterfactuals that the merger parties, the ACCC and any intervenor advance to the Tribunal.
289 The identification of counterfactuals is an important and necessary step to be taken in the context of the requisite single evaluative judgment of competitive effects of a proposed acquisition. As Yates J relevantly stated in Australian Competition and Consumer Commission v Metcash Trading Ltd (2011) 198 FCR 297; [2011] FCAFC 151 at [228], in addressing what his Honour described as a “calculus” to detect changes in the state of competition in a market:
As I have noted, a counterfactual is no more than an element of that calculus. Conceptually, it has no separate existence or purpose in the present context, other than as an aid to detect the existence and extent of change in the process of competition.
290 It follows that it would be an error of principle to seek to erect some degree of satisfaction or likelihood as a precondition to the Tribunal considering a counterfactual advanced by a party. It is in that context that the concept of a “realistic commercial possibility” must be understood. An assessment that a counterfactual may be a realistic commercial possibility does not carry with it any implication that it is more probable than not to occur, or that it is likely in the sense of a real chance. Rather, it provides some assistance, given the breadth of the concept of a “realistic commercial possibility”, in assessing the “degree of likelihood” of that particular fact existing or arising, as explained by Middleton and O’Bryan JJ in Pacific National at [246(c)], and by Beach J at first instance in Australian Competition and Consumer Commission v Pacific National Pty Ltd (No 2) [2019] FCA 669, in which his Honour stated at [1279]:
Now I accept that the ACCC does not necessarily need to prove its counterfactual on the balance of probabilities. But the magnitude of any real chance that it demonstrates in respect of the alleged future states will practically and ultimately affect the magnitude of the real chance that it is able to demonstrate in respect of the alleged effects on competition and whether that rises to the requisite level of a likely effect of substantially lessening competition. Considering both the counterfactual and the alleged competitive effects together, the evaluation required compositely is whether a real commercial likelihood of a substantial lessening of competition has been shown.
(Emphasis added.)
291 Ultimately, it is the task of the Tribunal to determine the degree of likelihood for any postulated counterfactual as a constituent element of its single evaluative judgment.
292 The Tribunal accepts, on balance, that in determining the degree of likelihood of the Bendigo Merger counterfactual occurring, it does fall within the broad concept of a realistic commercial possibility, notwithstanding some significant challenges to its potential execution, for the following principal reasons.
293 First, both Bendigo and SGL, objectively, would have compelling incentives to pursue a merger if the Proposed Acquisition did not proceed. The large increase in scale and the doubling of its current market share in home loans, would provide Bendigo with a greater asset base over which to fund fixed costs. A merged Bendigo/Suncorp Bank would also facilitate expenditure on improvements in information technology necessary to address legacy issues for both banks and improve competitiveness in processing loan applications. At the same time, SGL would benefit from the additional value that could be achieved by an unwinding of the conglomerate discount and the ability to operate as a “pureplay” insurer.
294 The Tribunal accepts that there has been historical volatility in the perceived conglomerate discount and, therefore, its significance to any consideration by the SGL board of the future of Suncorp Bank, could vary. On balance, however, the Tribunal is persuaded that it was recognised in SGL strategy papers to be a potential material benefit of any divestiture of Suncorp Bank.
295 In an update to the SGL board in April 2021 on “organic options” and “inorganic options” for Suncorp Bank, [REDACTED].329 The update also noted that the uplift in market valuation for SGL [REDACTED]was estimated to be $[REDACTED] million.330
296 A strategy paper provided to the SGL board in April 2022 for scenario planning purposes for Suncorp Bank included the following observation about the [REDACTED]:331
[REDACTED].
297 One of SGL’s principal rationales for selling Suncorp Bank was to simplify its business, and refocus its capital investments by allowing for a “singular focus on its insurance business”.332 By selling Suncorp Bank, it “could achieve a rebalancing and rerating with the consequential enhancement of value for shareholders”, and increase its capacity to “attract capital to support the Suncorp general insurance operations at a higher valuation”.333
298 At the same time, the Tribunal accepts that SGL was not a seller of Suncorp Bank “at any price”.334 The strategy papers presented to the SGL board consistently compared “organic options” (that is, retaining Suncorp Bank), and “inorganic options” (that is, divestitures of Suncorp Bank by way of sale or merger with other banks). The Tribunal accepts SGL’s position that its willingness to divest Suncorp Bank would depend on the value that could be realised from a proposed transaction. Relevantly, however, an SGL board submission in June 2022, following SGL’s receipt of ANZ’s non-binding indicative offer, stated that [REDACTED].335
299 The June 2022 board submission identifies the following “two inorganic options”:336
1. [REDACTED].
2. [REDACTED].
(Emphasis in original.)
300 These contemporaneous documents presented to the SGL board illustrate the extent of the incentives and explain the perceived rationale [REDACTED].
301 Second, the possibility of a merger between Bendigo and Suncorp Bank [REDACTED] in the period leading up to the announcement of the Proposed Acquisition. [REDACTED] Notwithstanding the announcement of the Proposed Acquisition, it is readily apparent that Bendigo is still pursuing a merger with Suncorp Bank, not least because of the extensive written and oral submissions that it has made to the Tribunal in the course of the hearing of this application.
302 Third, prior to the Revised Barrenjoey Analysis, [REDACTED]. The Tribunal accepts that the assumptions as to [REDACTED] in the Barrenjoey 2022 Analysis and the [REDACTED] were likely overstated, and Bendigo would have to negotiate amendments to the Metway Merger Act, as it applies to Suncorp Bank, with the Queensland Government, in order for any merger to proceed.
303 At the same time, the Tribunal is not persuaded by the assumption in the Revised Barrenjoey Analysis that cost synergies could not even begin to be realised until year [REDACTED]. The Tribunal is also not persuaded that realisation of any cost synergies is dependent on branch closures or reductions in the number of employees in Queensland, the timing, scope and extent of which might well be dependent on the Metway Merger Act negotiations with the Queensland government.
304 The synergies that had been identified in earlier analyses by SGL and Bendigo were not limited to synergies in Queensland. The SGL April 2022 analysis summarised the likely synergies that would be realised by a merged Bendigo/Suncorp Bank as:
– [REDACTED].
– [REDACTED].
– [REDACTED]
– [REDACTED].337
305 There was also evidence before the Tribunal that Bendigo had successfully integrated technology systems following recent acquisitions of other entities, in a relatively efficient manner.338 The acquisitions to which Bendigo referred included First Australian Building Society in 2000, Adelaide Bank in 2007, Bank of Cyprus Australia in 2011, Rural Finance Corporation of Victoria in 2014 and ANZ Investment Lending Portfolio in 2022.339 The Tribunal also accepts that Bendigo is in the process of consolidating its banking platforms and transitioning to cloud-based applications, in a relatively efficient manner.340
306 As to Bendigo’s delay in integrating past acquisitions including Adelaide Bank, Delphi and Rural Bank,341 Mr Brosnahan gave evidence to the effect that such a delay was [REDACTED]. Mr Brosnahan then came into his role in 2019. Based on this timeline, Mr Brosnahan expressed the view that Bendigo was “pushing through” one integration “every six months”.342
307 By contrast, Adam Bennett, the CIO of SGL, expresses the view that Bendigo’s examples of previous acquisitions that were effectively integrated cannot be relied on to establish that Bendigo has the capacity to successfully integrate Suncorp Bank.343 Relevantly, Adam Bennett states that Suncorp Bank would need to be “de-integrate[d] and re-host[ed]” to be integrated into Bendigo, which is the most complex form of integration,344 and none of the successful integration examples provided by Bendigo required this approach. 345 In this regard, the Tribunal accepts that the integration of Suncorp Bank would be more complex and on a materially different scale to past successful integrations undertaken by Bendigo.346
308 There was also evidence before the Tribunal as to what ANZ considered to be the level of difficulty it would encounter during the integration of Suncorp Bank and ANZ’s consideration of the potential timeline for the integration after the acquisition. [REDACTED].347 [REDACTED],348 [REDACTED],349 [REDACTED].350 Although this evidence demonstrates that [REDACTED], it is specific to ANZ and its internal development of ANZ Plus.
309 To the extent that Bendigo would encounter increased difficulty integrating Suncorp Bank, the Tribunal accepts that Bendigo would be strongly motivated to overcome such difficulties. In particular, because of Mr Brosnahan’s evidence that a scenario where Bendigo was [REDACTED].351
310 Fourth, the necessity to undertake [REDACTED] may present challenges to Bendigo, but the Tribunal is not persuaded that they would be insurmountable. In this regard, there was a considerable difference in the Barrenjoey and [REDACTED] estimates of integration costs. [REDACTED].352 [REDACTED].353
311 The assessment of net shareholder value realised for SGL on a divestiture of Suncorp Bank in the SGL April 2022 Analysis included estimates of integration costs to be incurred by SGL. It estimated integration costs of $[REDACTED] million on a low case (giving rise to net cost synergies of $[REDACTED] million), and integration costs of $[REDACTED] million on a high case (giving rise to net cost synergies realised of $[REDACTED]billion).354 These estimates are substantially lower than the integration costs assumed in the Revised Barrenjoey Analysis of $[REDACTED] million.355
312 The Tribunal has placed more weight on the integration costs assumed in the Revised Barrenjoey Analysis. Those assumed integration costs are more consistent with the [REDACTED] average multiple of integration costs to run rate synergies incurred in larger bank mergers and acquisitions in Australia, in the period between 1995 and 2020.356 The integration costs of $[REDACTED] million, assumed in the Revised Barrenjoey Analysis, were a [REDACTED] multiple of the $[REDACTED] million (pre-tax) forecast of run rate synergies that could be achieved by a merged Bendigo/Suncorp Bank.357 In contrast, the integration costs of [REDACTED].358 The Tribunal notes that if both figures were on a consistent tax basis, the ratio would be [REDACTED].
313 Further, the integration costs assumed in the Revised Barrenjoey Analysis are more consistent with ANZ’s estimate of integration costs of between $[REDACTED] and $[REDACTED], calculated with the benefit of extensive due diligence, in May 2023.359
314 SGL’s earlier analyses had suggested that it would [REDACTED]. [REDACTED]:
[REDACTED].360
315 Fifth, on balance, the Tribunal gives more weight to the expert evidence of Dr Howell on the likely credit ratings of a merged Bendigo/Suncorp Bank, and Mr Ali on the prospects of the merged entity obtaining IRB accreditation, in preference to Ms Starks’ opinions. The Tribunal considers that Ms Starks lacks the specialist banking expertise of Dr Howell and Mr Ali on those subjects.
316 The Tribunal accepts Dr Howell’s evidence as to the unlikelihood, at least in the short term, of any improvement in the credit ratings for a merged Bendigo/Suncorp Bank, from Bendigo’s current credit ratings, other than a one notch improvement in the Moody’s rating.361 Dr Howell’s evidence is consistent with [REDACTED].362 [REDACTED].363
317 The Tribunal accepts the evidence of Mr Ali that a [REDACTED] year time frame is realistic to achieve IRB accreditation, and that Bendigo’s prospects of achieving IRB accreditation would not be materially enhanced by a merger with Suncorp Bank.364 Dr van Horen gave evidence to a similar effect, and even went further, stating that in his view, the increased complexity of a merged Bendigo/Suncorp Bank may well reduce its likelihood of obtaining IRB accreditation.365
318 It is also necessary to take into account the cost of developing appropriate capabilities to achieve IRB accreditation. Mr Ali gave evidence of the substantial cost associated with the development of appropriate risk management and modelling capability in order to achieve IRB accreditation. He notes that Suncorp Bank’s estimated costs of $[REDACTED] million to pursue IRB accreditation is “broadly consistent with my expectation of [an] up to $[REDACTED] million investment requirement (depending on the size and complexity of the individual bank’s circumstances)”.366
319 The Tribunal notes that Bendigo has been pursuing IRB accreditation for [REDACTED] independently of any merger with Suncorp Bank or any other regional bank. The Tribunal is not satisfied that a merged Bendigo/Suncorp Bank would have any greater chance of obtaining IRB accreditation, given Mr Ali’s evidence that there is no minimum size requirement for a bank to seek IRB accreditation.367 Further, a merger may well require Bendigo to develop new or revised models that could further delay achieving IRB accreditation and, therefore, as Mr Ali concludes, it may be more complex and time consuming than Bendigo continuing to pursue IRB accreditation independently of any merger.368
320 The absence of any significant improvement in credit ratings or unlikelihood of IRB accreditation, at least in the short term, would likely make it more difficult for the merged Bendigo/Suncorp Bank to raise capital and reduce its marginal cost of lending. This, in turn, would make it more difficult for the merged entity to compete and undermine any coordinated conduct by the Major Banks. The absence of any likely significant improvement in credit ratings, at least in the short term, however, would only diminish rather than extinguish the prospects of a successful Bendigo/Suncorp Bank merger.
321 Sixth, the further considerations raised by the applicants and summarised at [206]-[212] above, highlight the inherent difficulties and limitations of any counterfactual analysis. They focus on the inevitable difficulties in assessing how a firm faced with an unfavourable regulatory decision would have to present what could be described as a “second-best” alternative to shareholders, and the market more generally.
322 On balance, while the Tribunal accepts that the further considerations may present some practical issues that would need to be addressed, it has not placed significant weight on them in assessing the likelihood of the Bendigo Merger counterfactual. In the absence of the formulation of a fully costed and specific proposal following the completion of necessary due diligence, it is not clear what weight could be given to these matters. Previous statements made by senior executives about the viability of a merger, particularly those made after the announcement of the Proposed Acquisition, could well become superseded in an environment where the Proposed Acquisition was no longer possible. One could readily infer that statements made after the announcement of the Proposed Acquisition and in the context of the authorisation application before the ACCC by senior executives of the applicants, would have an understandable predisposition to focus on potential downsides and conservative assumptions of potential benefits. Conversely, it could also be expected that statements made by Bendigo’s senior executives, have an understandable predisposition to emphasise potential upsides and make aggressive assumptions of potential benefits.
323 A necessary preliminary step that must be undertaken in considering any authorisation application, is to identify the areas of competition that may be affected by the proposed acquisition.
324 The ACCC identified the following markets for the purpose of identifying the nature of the competitive environment in which the Proposed Acquisition would occur, in order to assess likely competition effects, benefits and detriments, and to identify other factors that might constrain such effects:
(a) a national home loans market;
(b) a national retail deposits market, including transactions and savings accounts, and term deposits;
(c) local or regional markets for the supply of banking products to agribusiness customers in Queensland; and
(d) local or regional markets for the supply of banking products to SME customers in Queensland.369
325 The ACCC identified these markets for consideration based on their assessment that these were the products and services, and geographic areas in which ANZ and Suncorp Bank compete most closely.370
326 The ACCC concluded that the Proposed Acquisition was likely to lead to some, but not a substantial, lessening of competition in the national market for retail deposits.371 The ACCC did not resile from that position before the Tribunal and the applicants did not address this market in their written submissions. The ACCC, however, submits that if the Tribunal came to a similar conclusion that there had been some, but not a substantial, lessening of competition in a market, the Tribunal must take that into account in deciding whether there is a net public benefit under s 90(7)(b) if the Proposed Acquisition were to proceed.372 That submission is addressed in the Tribunal’s consideration of net public benefits below at [907].
327 For present purposes, the Tribunal is satisfied that it is sufficient to consider the likely competition effects, benefits and detriments of the Proposed Acquisition in the context of the national home loans market and the local or regional markets for agribusiness banking and SME banking in Queensland. These markets capture the products and geographic areas that have the highest degree of overlap in the products and services offered by ANZ and Suncorp Bank, and the most likely areas in which competitive harm could occur.
E.2. National home loans market
328 It was common ground that there is a national market for home loans.373
329 The Tribunal agrees that a national market for home loans is a relevant market for the purpose of analysing the competitive effects of the Proposed Acquisition.
330 Home loans comprise loans and related services provided by banks to customers to enable them to finance the construction or purchase of residential properties for owner-occupation or investment. They include loans that are provided to customers to enable them to refinance their home loans with other banks and to switch to a new product with the bank that has provided their current home loan.374
E.3. Agribusiness and SME Banking markets
331 The applicants do not accept that there are discrete markets for the supply of banking products to either agribusiness or SME customers. Rather, they submit that banking products are supplied to those customers in a national business banking market.375
332 The ACCC maintains that there are distinct local or regional markets for the supply of banking products to agribusiness and SME customers in Queensland. It submits that if the Tribunal were not satisfied that there were distinct local or regional markets, the Tribunal could nevertheless conclude that each of the agribusiness and SME sectors in Queensland was a significant section of the “relevant market”.376
E.3.2. The applicants’ submissions
333 The applicants submit that agribusiness and SME customers cannot readily or coherently be defined separately from broader business banking customers. They submit that (a) there is no distinct cohort or accepted dividing line between SME, agribusiness, and other business customers, (b) many banks treat SME, agribusiness, and other business customers as a single customer cohort, or otherwise distinguish segments within the broader business customer cohort idiosyncratically, and (c) even where banks distinguish, for instance, SME customers as a particular segment, those customers are not consistently defined.377
334 They submit that products and services supplied by banks to agribusiness and SME customers are generally substitutable and provided in a national business banking market for the following reasons.378
335 First, the products and services demanded by and supplied to SME, agribusiness and other business customers are generally the same. The applicants submit those products and services are typically provided through one or more transaction accounts, business loans, commercial credit cards and merchant services. They submit that the only material exception to the products and services demanded by and supplied to SME, agribusiness and other business customers are Farm Management Deposit accounts that operate pursuant to a government tax concession scheme to eligible agribusiness customers.379
336 Second, the relationship management services provided to business customers, including to agribusiness, SME and other business customers, are typically the same and, regardless of customer segment, business customers value relationships with banks that understand their business.380
337 Third, the products and services supplied to business customers are supplied nationally, the same credit policy and pricing frameworks typically apply to customers nationally, and pricing decisions are often made nationally.381
338 Fourth, customer risk profile is relevant to individual pricing decisions for business loans. The applicants submit, by reference to Dr Williams’ report, that risk profile in agribusiness is informed by geographic location and product sub-sector and, therefore, the need for geographic diversity to balance risk in agribusiness is an important feature indicating competition occurs at a national level.382
339 Fifth, business customers, including agribusiness and SME customers, do not necessarily require a local branch presence. The applicants submit that the importance of local branch presence for all business banking customers has been reduced by structural changes, including digitisation, the decline of cash, alternative cash-handling options, and the use of brokers.383 They submit that the contention that there is no substitutability in relationship management for agribusiness and SME customers is not correct. They also submit that agribusiness customers, in particular, do not require a local branch presence, and whilst they value local knowledge and commitment, this can be achieved without a local branch presence.384
340 Sixth, the applicants submit that many agribusiness and SME customers are not served by dedicated relationship managers and are instead portfolio managed, or are served by non-specialist relationship managers or managers serving both agribusiness and SME customers.385 In any event, they submit that (a) relationship managers can develop a specialty in agribusiness in the medium term,386(b) many business customers are also portfolio managed or remotely managed nationally, rather than locally or regionally, and primarily use self-service banking,387 (c) for customers who are relationship-managed, relationship managers typically visit their business premises at least annually,388 and this can be, and is, achieved by basing relationship managers in key regional centres or capital cities from which they can travel (often) long distances.389
341 The ACCC submits that by applying the orthodox approach to market definition of focusing on substitutability the following factual propositions emerge from the evidence with respect to the product and geographic dimensions of the market or markets for the supply of banking products and services to agribusiness customers. 390
342 First, agribusiness customers have demands that differentiate them from other customers. Those demands include the need for flexible repayment terms given the impact of seasonality on agribusiness cashflows.391
343 Second, banks offer tailored banking products to agribusiness customers in response to the specific needs and characteristics of those customers who are generally, “asset-rich and cash flow poor”.392
344 Third, agribusiness customers typically require and demand bankers with specialised knowledge in agribusiness.393
345 Fourth, specialised knowledge in agribusiness is to some extent, sector specific, given agribusiness customers want bankers who understand their particular line of agribusiness, such as beef, wheat or cotton.394
346 Fifth, agribusiness customers typically require relationship management, including on-site farm visits from their relationship managers.395
347 The ACCC submits that there is evidence before the Tribunal that weighs in favour of a conclusion that there is a discrete market for the supply of SME banking services that does not include agribusiness banking customers or corporate customers.396
348 The ACCC submits that this evidence includes evidence that (a) there are banking products that are specifically tailored to SME customers’ needs, (b) some banks have specific customer portfolios or strategies for their SME customers, and (c) SME customers are unlikely to switch to other types of business banking products in response to small but significant non-transitory price increases.397
349 The ACCC submits that SME markets may be local or regional rather than national given the potential for cash handling to be important for SME customers and the importance of personal knowledge and local knowledge for lending decisions for SME customers.398
350 Alternatively, the ACCC submits that the Tribunal might otherwise conclude, if it were not satisfied that there are distinct local or regional SME banking markets in Queensland that “the SME sector” in Queensland was a significant section of the “relevant market”.399
351 It is well established that identifying a market and defining its product and geographic dimensions is a focusing process that seeks to determine the clearest picture of the relevant competitive process in the light of commercial reality and the postulated specific theory of harm that has been advanced: Australian Competition and Consumer Commission v Flight Centre Travel Group (2016) 261 CLR 203; [2016] HCA 49 at [69] (Kiefel and Gageler JJ). Further, if an acquisition or conduct is found not to have the likely effect of substantially lessening competition in narrow markets it would generally be even less likely to raise competition concerns in more broadly defined markets.
352 The Tribunal is satisfied that it is appropriate to consider the competitive effects of the Proposed Acquisition in agribusiness and SME local/regional banking markets in Queensland. Other than with respect to home loans, the overlap between ANZ and Suncorp Bank in the supply of banking services is closest for agribusiness and SME customers in Queensland. In assessing the competitive effects of the Proposed Acquisition, any likely substantial lessening of competition could be expected to be most evident in the area of greatest overlap in supply.
353 Further, contrary to the submissions advanced by the applicants, the use of national lending standards and pricing frameworks is not determinative in defining the geographic boundaries of a market. A uniform framework does not necessarily result in uniform prices and uniform prices may simply reflect similar levels of competition in different geographic markets. The critical question is whether a local or regional presence is necessary in order to be considered a close substitute in demand or supply. 400
354 Similarly to SME banking customers, agribusiness banking customers typically require a bundle of banking products including term debt, overdrafts, equipment finance, transactional banking, credit cards, residential investment loans and term deposits. There were some suggestions in the evidence, however, that agribusiness customers may be more willing to unbundle loan products. This included evidence that Suncorp Bank is largely a secondary lender to agribusiness customers rather than their MFI and explains why it has a larger proportion of agribusiness loans than deposits.
355 Agribusiness products are generally very similar in design to general business banking products but are distinguished by serviceability and seasonality of agribusiness cash flows. Different and unique considerations also arise with respect to security. As explained by Bendigo in a response to an ACCC information request, (a) loans to agribusiness customers are typically for longer terms than for other customers as they form a basis of the capital reserve that a farmer holds, (b) discrete issues arise for key assets such as water, suitable usage of land and inputs costs related to specific locations and access to key markets, and (c) it is necessary to determine whether the security provided is “intergenerational” or ultimately to be used for clearance of a debt.401
356 In addition, there are some banking products that are only offered to agribusiness customers. The only commonly used specific agribusiness products are Farm Management Deposits accounts. These products permit farmers to smooth the cash flow cycle for their agribusinesses by enabling them in a tax efficient manner to invest profits to meet commitments in years when cash flow is much weaker.402 Other products that were identified by banks in responses to the ACCC that they provided to agribusiness customers in addition to generic “business products” and Farm Management Deposit accounts were described by (a) NAB, as a “farm management transaction account overdraft”, “farm management transaction account”, “Livestock finance” and “Agri green Loan”,403 (b) Bendigo, as “Stock and crop loans”,404 and (c) CBA, as an “AgriGreen Loan”.405
357 Moreover, agribusiness banking services are generally supplied by banks through relationship managers or with the support of specialist staff. All parties agreed that an agribusiness customer’s relationship with their relationship manager is a “particularly important factor for agribusiness customers in choosing a financial institution”.406 The importance of relationship managers was demonstrated by Suncorp Bank’s loss of agribusiness customers when it reduced services for smaller agribusiness customers and then was forced to reinstate relationship managers for them in order to arrest the decline in its small agribusiness customer base.407 All 660 agribusiness customers of Bendigo in Queensland are relationship managed directly rather than through a pool management model.408
358 The Tribunal is satisfied that the existence of Farm Management Deposit accounts and the specialist support services provided by specialist agribusiness managers would preclude sufficient demand side substitution to make a SSNIP by a hypothetical monopolist of agribusiness banking services unprofitable. In addition, supply side substitution would be unlikely in the absence of specialist agribusiness resources. Recruiting specialist agribusiness managers or otherwise investing to acquire specialist resources would be in the nature of new entry, not supply side substitution that could defeat a SSNIP.
359 Other matters that support the existence of agribusiness markets, include competition between banks for specialist agribusiness bankers,409 and the existence of Rabobank as a specialist agribusiness banker.
360 The geographic dimension of agribusiness banking markets are more likely to be local or regional rather than national, given the requirement for on-site visits to agribusiness customers and the necessary knowledge to provide specialist agribusiness services is likely to be at least regional in nature given the broad range of agricultural activities and climatic conditions in Australia.
361 The long distances that relationship managers travel to visit agribusiness customers, as Ms Starks concluded in her supplementary expert report, suggest the geographic dimension of agribusiness markets in Queensland are more regional than local. A local branch may improve a bank’s competitiveness but is not essential as agribusiness customers typically do not need to make cash deposits and the more critical issue is visiting a customer’s farming operations rather than the customer travelling into a local branch of a bank. Bendigo’s relationship managers visit approximately 90% of Bendigo’s Queensland agribusiness customers in Queensland on their farms.410 For the purposes of this determination by the Tribunal, it is not necessary to draw any bright lines between “local” and “regional” markets.
362 The lack of any uniform definition by banks of what constitutes an SME customer is not determinative. The relevant question is whether there is an area of product and geographic space (albeit often without precise boundaries) that can be distinguished from other products on the basis of substitution. This is typically determined by employing the hypothetical monopolist test, characterised by asking whether a SSNIP would be profit maximising or would be defeated by demand or supply side substitution.411
363 In assessing demand and supply side substitution it is important to recognise that SME customers require a cluster of banking services and a majority according to market research conducted by Cameron Research Pty Ltd (Cameron Research) in 2022 have a preference for acquiring those services from a single bank for reasons of convenience.412
364 Other characteristics of SME banking identified by Cameron Research also need to be considered in assessing substitution possibilities. These include the existence of a relationship with a dedicated business banker who understands their business and access to a local branch for cash handling and relationship management purposes.413 These give rise to distinct service element to SME banking beyond the products offered and the need for a degree of local business knowledge by banks, not only to provide the service element effectively but also to assess credit risk.
365 Ultimately demand side substitution for SME customers will turn on their willingness, in response to a SSNIP, to unbundle their banking services and acquire different unspecialised banking services from other banks or to acquire general banking services. Supply side substitution in response to a SSNIP will depend on any unbundling costs in supply and the extent to which banking resources can meet the requirements of SME customers as distinct from other banking customers.
366 On balance, the Tribunal is satisfied that the requirements of SME banking customers are sufficiently differentiated, in particular on service dimensions, to make it unlikely that a SSNIP by a hypothetical monopolist of SME banking services would be defeated by demand side or supply side substitution.
367 The Tribunal is also satisfied, notwithstanding that the importance of branches is declining, that the geographic dimension of the relevant SME markets in Queensland is best characterised as local/regional given the importance to SME customers of a relationship with a dedicated business banker who understands their business and access to a local branch for cash handling and relationship management purposes.
F. COMPETITIVE EFFECTS OF THE PROPOSED ACQUISITION IN THE NATIONAL HOME LOANS MARKET
368 None of the expert economists consider that the Proposed Acquisition would have unilateral effects that would have the effect, or would be likely to have the effect of, substantially lessening competition in the home loans market, compared with either counterfactual. Nor, has any party sought to contend that the unilateral effects of the Proposed Acquisition would have the effect, or would be likely to have the effect of, substantially lessen competition in the home loans market compared with either counterfactual.
369 The ACCC concluded at [6.172] of its Reasons for Determination that:
[W]hile there is likely to be some lessening of competition arising from the unilateral effects of the Proposed Acquisition, the ACCC considers it is unlikely to be substantial.414
370 The ACCC submits, however, that the requirement to make a single evaluative judgment requires the Tribunal to take into account the unilateral effects of the Proposed Acquisition in the home loans market, together with coordinated effects.415
371 The applicants submit that if a contended likely lessening of competition was accepted not to be substantial, it could not be relevant or meaningful to the competitive process, and should not be taken into account for the purposes of the competition test in s 90(7)(a).416 They submit that insubstantial effects in a market cannot be combined to give rise to a substantial effect in the market.417
372 The requirement to make a single evaluative judgment for the purposes of the competition test in s 90(7)(a) does not carry with it any obligation or entitlement to have regard to mutually exclusive theories of harm, in a cumulative manner, in assessing any contended “non-substantial lessenings” of competition in markets.
373 Generally, it is not possible to have coordinated and unilateral anticompetitive effects as a result of eliminating competition between the same parties.418 While there would always be some unilateral effects from eliminating competition between parties, once one introduces coordinated effects into a competition inquiry, the focus turns to an effect that is generally larger than unilateral effects and effectively subsumes them. If the unilateral effects were larger than the coordinated effects, for example, because of the dominant position of the merged firm, the merged firm would not be incentivised to engage in coordinated conduct.
374 In this case, the ACCC may be contending that on an ex ante, pre-merger, perspective looking forward, that you should weigh the probability of unilateral effects (which may be small or large) between the merger parties and coordinated effects between the Major Banks. Such a probabilistic approach would not lead to any double counting of anti-competitive effects, but it would not make much difference to the likelihood of a substantial lessening of competition if, as is the case here, the chance of significant unilateral effects is very small.
F.2. Economic framework to analyse coordinated effects
F.2.1. What are coordinated effects?
375 In merger analysis, the distinguishing feature of coordinated effects, when compared with unilateral effects, is that the effect depends on the accommodating actions of suppliers, other than the merger parties, to increase prices or reduce output and other non-price attributes. They reflect the exercise of the combined market power of firms (or a group of firms) in the market. Coordination may be tacit or explicit and will not necessarily involve any conduct that otherwise breaches the CCA. Through repeated interaction in the market, firms may simply come to understand that lowering their prices will be quickly matched by rivals, they will likely gain no market share, but will lose margin.
376 Whether a proposed acquisition would not have the effect, or would not be likely to have the effect, of substantially lessening competition, on the basis of coordinated effects, will depend on the objective characteristics of the market and, importantly, whether and how the merger may make the market more vulnerable to coordination. Coordination does not need to be perfect: indeed, if coordination was delivering monopoly pricing pre-acquisition, there may be no coordinated effects following the proposed acquisition. Nor does a coordinated effects theory of harm require any particular prediction about pricing or other outcomes. Rather, the proposed acquisition needs to change the conditions for coordination, such that the Tribunal is not satisfied that it would not have the effect, or would not be likely to have the effect, of substantially lessening competition.
377 Unilateral effects, by contrast, include both the result of eliminating competition between the parties to the proposed acquisition, and the responses of other suppliers in the market, which may enhance or dilute the anti-competitive effect of eliminating competition between the merger parties. In practice, however, the distinction between unilateral and coordinated effects is not as clear cut. In oligopoly markets with a small number of suppliers, each supplier will rationally take account of both the actions, and the expected reactions of their rivals when setting their own price and output (including product range, attributes, quality, etc). Before cutting prices or improving their products, suppliers will consider whether, and how quickly, rivals are likely to copy them, diluting or eliminating any market share and profit gains they might otherwise make from these competitive actions. This type of conduct is sometimes classified as unilateral conduct and sometimes considered as accommodating behaviour under coordinated effects.
378 A merger in an oligopoly market could lead to any of the following:
(a) increased risk of an explicit cartel agreement (in terms of establishment, stability, coverage and duration);
(b) increased risk of a tacit agreement or understanding that includes some broader signalling, communication, facilitating practice, or plus factors; and/or
(c) raising the oligopoly outcome purely as a result of each player now setting prices on the basis of the expected reactions of a smaller group of rivals, without any form of communication.
379 The literature on coordinated effects is not consistent regarding inclusion of the third category, which is where the line between coordinated and unilateral effects starts to blur.
380 “Live and let live” coordination, which is the concern raised by Ms Starks and the ACCC,419 could be categorised under either the second and/or third category, depending on whether the “understanding” emerges organically from repeated interactions or whether there are “plus factors” which could simply be things like advanced announcements of price increases.
381 Coordination is always possible in oligopolistic markets with repeated interaction, as long as firms are “sufficiently patient”, that is, they value higher future prices and profits over increased current sales at lower prices. The outcome can be difficult to predict: the “folk theorem” of oligopoly suggests that “anything can happen”, ranging from competitive to monopoly prices. It ultimately depends on the characteristics of the market, and the conduct of market participants. In the context of a merger and coordinated effects, what matters is how the merger changes the conditions for coordination, and specifically, how they might be enhanced to the detriment of consumers.
382 Economists (following Stigler),420 generally agree that three conditions need to be satisfied for coordination to be sustained, firms need to be able to (a) reach a consensus (either explicitly, tacitly or organically) on price or quantity, (b) detect deviations from that consensus, and (c) punish those deviations. The Tribunal notes that the coordinated consensus may simply involve prices being sustained above the competitive level as a result of repeated interaction in the market, and “punishment” of cheating may simply involve reverting to competitive pricing. While it will generally be in the joint interests of suppliers to coordinate to raise prices, it will often be in their individual interest to cheat and undercut rivals. Raising prices above the competitive level will leave unmet demand, and the individual firm can benefit from this demand by shaving prices. Sustaining supra-competitive pricing depends on each firm in the coordinating group valuing the benefit of higher returns in the future, over increased sales, in the present. If cheating cannot be detected and punished, sustaining a coordinated outcome becomes more difficult.
383 In addition to these “internal” factors, the coordinated outcome is unlikely to be sustained if threatened by “external” factors such as the reactions of actual and potential competitors outside the coordinating group or of customers able to mobilise alternatives.
384 A number of market characteristics are recognised as potentially making a market more or less vulnerable to coordination, that is, they make it more or less likely that suppliers are able to reach and sustain a coordinated outcome. Market concentration, cross ownership, multi-market contact, product homogeneity, firm symmetry, price transparency, the size and frequency of purchases, customer stickiness, the presence of “facilitating practices” and barriers to entry and expansion, are relevant factors considered by Professor de Roos and Ms Starks in their reports and/or in merger guidelines around the world.421
F.2.2. The key to coordinated effects is what changes
385 It is important that the various factors considered relevant to coordination do not lead to a “tick box” approach to assessing coordinated effects. It is necessary to consider the combination of these factors, and how they interact with each other in the particular market of concern. Not all factors need to be present for coordination to be successful.
386 The key question on a merger authorisation application is, how does the merger change the likelihood of successful coordination, if at all? That question necessarily involves consideration of whether the proposed merger could make the market more vulnerable to coordination, or how coordination might become more complete or more sustainable post-merger. Evidence of actual coordination, in isolation, will not answer this question. Indeed, in some cases, it might suggest that the proposed merger would make little difference.
387 The most obvious change in a market following a merger, is concentration. With fewer firms in the market, it generally becomes easier to reach consensus, and to detect deviations from that consensus. Similarly, an increase in symmetry may make it easier to reach a consensus and reduce the incentive to cheat where the interests of the remaining firms in the market become more closely aligned.
388 Coordinated effects are most likely to be significant where coordination has, to date, been imperfect, or has broken down, such that the proposed merger could restore coordination or make it more stable or enduring, and less vulnerable to cheating and breakdown. The acquisition of a “maverick”, or, more broadly, a firm with the ability and incentive to undermine attempts at coordination, tends to be particularly troubling. For example, a firm with a small market share, has more to gain and less to lose from lowering prices and expanding output, assuming it has sufficient capacity to meet demand.
389 For similar reasons, coordinated effects are also particularly relevant where there is a counterfactual involving an alternative merger scenario that would create a rival with increased ability and incentive to undermine coordination in the future without the proposed acquisition that is under consideration.
390 This is the core of the coordinated effects concerns of the ACCC and Ms Starks, referred to at [380] above, and Professor King, more generally.422 The theory of harm is that “live and let live” coordination on prices has been undermined in recent years, and that the Proposed Acquisition poses the risk of restoring effective coordination between the Major Banks (the “coordinating group” according to the ACCC theory of harm).423 This would be so, because it increases their symmetry, both in terms of their relative size, domestic market focus and source of funds, which it is argued makes ANZ less likely to deviate from coordinated outcomes in a future with the Proposed Acquisition. The “coordinating group” would also be less vulnerable to competition from rivals with divergent interests outside the coordinating group, which is either Suncorp Bank in the No Sale counterfactual, or particularly, a merged Bendigo/Suncorp Bank in the Bendigo Merger counterfactual.
F.3. The principal contentions of the parties
F.3.1. The applicants’ submissions
391 The applicants submit that the incorporation of Suncorp Bank, the ninth largest bank by banking system assets, into ANZ, would not alter the competitive dynamics, or substantially lessen competition in any of the relevant markets in the Australian banking industry, including in the home loans market.424 They submit that this proposition is apparent if the Tribunal compares the future with and without the Proposed Acquisition.
392 In the future with the Proposed Acquisition, ANZ continues to be the smallest of the Major Banks by banking system assets, it retains incentives to compete against, and is constrained by, existing competitors.425 They submit that ANZ will also be constrained by the credible threat that other banks will enter or expand in the home loans market, including by potential mergers involving Bendigo, Bank of Queensland and/or ING.
393 In the future without the Proposed Acquisition, the only commercially realistic counterfactual is the status quo, which would have competitive dynamics that are materially the same as those in a world with the Proposed Acquisition.426 They submit that even if a merger were to occur between Bendigo and Suncorp Bank, there is no real likelihood that the merged entity would become a materially more effective competitor or impose a competitive constraint greater than that exerted by either bank alone.
394 The applicants submit that the evidence does not establish that, in the future with the Proposed Acquisition, compared with the future on any commercially realistic counterfactual, that coordinated conduct would be more likely to be initiated or sustained, let alone to a degree that might give rise to a substantial lessening of competition in the home loans market.427 They submit that the evidence establishes that the home loans market is not conducive to coordination and the Proposed Acquisition would not increase the likelihood of coordination in the market.
395 Bendigo submits that as a consequence of the market concentration and structural advantage of the Major Banks, prices for banking products, including home loans, tend to cluster.428 Bendigo submits that the Major Banks aim to maintain a mostly uniform position rather than compete vigorously on price. It submits, citing the conclusions reached by the Productivity Commission in its report of 29 June 2018,429 that smaller institutions will follow Major Banks’ pricing decisions, which, at best, results in prices that reflect the cost incurred by the least efficient Major Bank, rather than the most efficient Major Bank.
396 Bendigo refers to the statements in the ACCC’s 2018 Residential Mortgage Price Inquiry that it had found evidence of an “accommodative and synchronised approach to pricing” between the Major Banks, enabled by their oligopoly market structure.430 They also cite the ACCC’s analysis of ANZ’s interest rate increases, which were stated to reflect its strategic interdependence with the other Major Banks and that ANZ was “unlikely to have chosen to increase its interest rates without the expectation that its competitors would follow its lead”.431
397 Bendigo also submits that the recent evidence of price competition in the home loans market should be given no or little weight because (a) it appears to be short term, (b) the home loans market is structurally conducive to coordination, and (c) it is not sustainable for smaller banks to compete on pricing because the Major Banks can sustain lower prices, including below the cost of capital, for longer than smaller banks.432
398 The ACCC submits that if the Tribunal concludes that the Proposed Acquisition would increase the likelihood, severity, or sustainability of coordinated behaviour among the Major Banks, then the Tribunal may not be satisfied that the Proposed Acquisition would not be likely to have the effect of substantially lessening competition in the home loans market.433
399 The ACCC submits that in the future with the Proposed Acquisition, ANZ has more to lose and less to gain by deviating from coordinated conduct among the Major Banks.434 They submit that the Proposed Acquisition leads to an increase in ANZ’s market share of 2.3%, which was described by its CEO, Mr Elliott, as “the equivalent of many years of organic system growth”.435
400 The ACCC submits that the Proposed Acquisition would also increase the symmetry between ANZ and the other Major Banks in other respects, including that it would (a) achieve a material 22% increase in retail deposits,436 making it more like the other Major Banks in circumstances where Mr Elliott had described its current funding base as “very different” to the other Major Banks437 and (b) increase ANZ’s domestic focus in relation to both loans and funding.438
401 The ACCC submits that the likely impact of the Proposed Acquisition is more pronounced when compared with the Bendigo Merger counterfactual than the No Sale counterfactual because the likelihood of coordination would be reduced by the greater scale of a merged Bendigo/Suncorp Bank.439
F.4. Is the National Home Loans market conducive to coordination?
402 Although the expert economic evidence advanced before the Tribunal identified many factors that might feed into an analysis of coordinated effects, the parties focused on factors that they considered to be particularly relevant to assessing the likelihood of coordination, and coordinated effects resulting from the Proposed Acquisition, by the Major Banks, in the home loans market. The Tribunal has, therefore, also focused on those factors.
403 Concentration affects both the ability and incentive of suppliers in a market to reach and maintain a coordinated outcome. With fewer suppliers in the market, it should be easier to achieve a consensus on the level of prices or output, and to detect deviations from that outcome.
F.4.2.2. The applicants’ submissions
404 The applicants submit that the home loans market is not concentrated. They submit that the post-merger Herfindahl-Hirschman Index (HHI) does not exceed the market concentration thresholds in the ACCC merger guidelines and neither of the other two factors that are stated in the guidelines to give rise to potential competition concerns, namely, the parties are particularly close competitors and the target has grown market share quickly or driven innovation, are present here.440 ANZ and Suncorp Bank are not close competitors and Suncorp Bank has not grown its market share quickly or driven innovation.441
405 The applicants also submit that current market shares reflect historic loans and do not reflect current competitive dynamics in the home loans market. They submit that, relevantly, Macquarie’s share of new loans is materially larger than its share of existing loans442 and its ambition is to be significantly larger.443
F.4.2.3. Bendigo’s submissions
406 Bendigo submits, by reference to Professor King’s report, that if the Proposed Acquisition proceeds, it will “stabilise” the asymmetry between the Major Banks by increasing ANZ’s market share closer to the average market shares of the other Major Banks and thus reducing the current disparity in the Major Banks’ market shares that has developed over the past decade.444
407 Bendigo also submits, by reference to Professor King’s report, that the Proposed Acquisition will allow ANZ to “buy” its market share in the home loans market, which will substantially reduce ANZ’s incentives to compete.445
F.4.2.4. The ACCC’s submissions
408 The ACCC submits that the home loans market is moderately concentrated in that the Major Banks comprise approximately 75% of all home loans nationally and only eight other ADIs have a market share of more than 1%.446 It submits that the structure of the market is relatively static and could not be characterised as dynamic.
409 The ACCC submits that there is also evidence of accommodative pricing strategies, whereby the Major Banks determine price by reference to what the other Major Banks are doing, while other market participants are “price takers” who cannot sustainably compete on price.447
410 The ACCC submits that the advantages of scale make it difficult for new entrants and smaller existing banks to provide effective competition that meaningfully constrains the Major Banks. It submits that Macquarie’s experience demonstrates that it takes significant time and investment for even a well-funded and recognised brand to successfully enter and gain market share in the home loans market. It also submits that the proposition that non-bank lenders are effective competitors to the Major Banks is not borne out in the evidence.448
411 The Tribunal is firmly of the view, for the following reasons, that the structure of the home loans market can fairly be characterised as moderately concentrated. The pre-merger and post-merger HHI measures for the market may not exceed the market concentration thresholds in the ACCC merger guidelines, and neither of the other two factors that are stated in the guidelines to give rise to potential competition concerns below the thresholds may be present, but the HHI measures are relatively high, and the Major Banks have, for many decades now, been the same four banks with dominant market shares.
412 A limitation of the HHI measure is that it is very sensitive to asymmetry in market shares, in that asymmetric market shares will lead to a much higher HHI than symmetric market shares. By way of example, if the market shares of four firms were 70%, 10%, 10%, 10%, the HHI being the sum of the squares of each market participants’ shares, would be 5,200 (4,900 + 100 + 100 + 100). By contrast, if the firms each had a market share of 25%, the HHI would be 2,500 (625 + 625 + 625 + 625). In turn, the use of a squaring calculation makes the HHI very sensitive to measurement error in the market shares of the largest firms in a market.
413 In assessing the likelihood of coordinated conduct by an oligopoly, however, symmetric market shares of dominant firms are likely to increase, rather than decrease, the likelihood of coordination. Firms are more likely to reach a consensus and would have similar incentives not to compete away profits by engaging in price competition. Indeed, the increased symmetry of ANZ with the other Major Banks is one of the two central contributors to the ACCC’s coordinated effects concerns.
414 Given that the HHI increases with disparities of market shares, it is not always the most useful measure of market concentration for the purpose of assessing coordination effects.
415 The market share of the largest four firms, or some other relatively small number, is an alternative measure used to determine the degree of concentration in a market. It is known as the CR4, or if the largest three firms in a market are used, the CR3. A post-merger CR4 threshold of concern of 75% was included in the 1999 and earlier iterations of the Merger Guidelines. Currently, a post-merger CR4 threshold of concern of 65% is included in merger guidelines published by the Canadian Competition Bureau,449 and a post-merger CR3 threshold of concern of 70% is included in the merger guidelines of the New Zealand Commerce Commission.450 Generally, the post-merger thresholds for concern, when using CRx measures, are also subject to the merged firm’s combined market share.
416 A limitation of the CR4 or CR3, however, or any other CRx threshold, is that it does not take account of the market shares of other participants in a market.
417 Given the respective limitations of both the HHI and CRx measures, the Tribunal considers it more useful to have regard to both measures to provide a more complete picture of market structure.
418 The average monthly market shares of participants in the home loans market between November 2021 and October 2022 were summarised in the ACCC’s Reasons for Determination at [6.45], from APRA monthly ADI statistics, in the following table:451
419 In the period between November 2021 and October 2022, the Major Banks made up approximately 74.5% of home loans nationally, and approximately [REDACTED]% of home loans in Queensland. Moreover, ANZ’s 13% market share (the smallest among the Major Banks) is marginally less than three times that of the 5th placed ADI, Macquarie, at 4.4%. It is more than four and half times that of the 6th placed ADI, Bendigo, at 2.8%.
420 Calculations of pre-merger and post-merger HHI for the home loans market were included in the following table in the ACCC’s Reasons for Determination at [6.47]:452
421 Both the pre-merger and the post-merger HHI are below the 2000 threshold and, in addition, the delta, at least for the home loans market, is less than the threshold of 100. Nevertheless, the figures are relatively high, and given the Major Banks’ 74.5% share of the home loans market would fall just shy of the CR4 threshold for concern specified in the 1999 iteration of the ACCC Merger Guidelines, the home loans market can readily be characterised as moderately concentrated.
422 Further, the Tribunal accepts that there was no evidence that Suncorp Bank had recently achieved a rapid increase in market share, driven innovation in the market, or tended to charge lower prices than other market participants.
423 The Tribunal accepts that Macquarie’s share of new loans is larger than its share of existing loans. It is necessary, however, to avoid conflating Macquarie’s net loan growth, which is skewed upwards by its relatively smaller and younger lending book, with its share of new loans. A more meaningful measure of Macquarie’s share of new loans is APRA data that shows the average value of home loans funded between November 2021 and October 2022. The APRA data suggests that Macquarie’s share of new loans is [REDACTED].453 Further, the data relied upon by ANZ, by reference to Dr Williams’ report,454 is taken from only one mortgage aggregator, Australian Financial Group. This data excludes loans that were initiated by other brokers and is likely to be skewed upwards for Macquarie because it relies almost exclusively (93%) on the broker channel to distribute its home loans, as demonstrated in the ACCC’s analysis of broker channel data, and included in the following table in its Reasons for Determination:455
424 By contrast, as the table records, [REDACTED].456
425 Where firms are more similar in terms of their size, products, costs, capacity and/or time horizons, they are more likely to be able to achieve a consensus on prices and/or outputs. If one or more firms are asymmetric, this will tend to result in divergent incentives. For example, a firm with lower costs and excess capacity may be more inclined to lower prices and expand their sales, undermining the coordinated outcome. Similarly, where a firm has a smaller market share than its rivals, it will tend to have more to gain, and less to lose from undermining coordination.
F.4.3.2. The applicants’ submissions
426 The applicants submit, by reference to Ms Starks’ report and the ACCC’s reasons at [6.215], that there is a lack of symmetry and alignment between the Major Banks.457 They submit their respective market shares vary significantly,458 the Major Banks are differentiated in other non-price attributes, such as turnaround times for loan approvals459 and there is a lack of symmetry in their funding base, products and geographic diversity.460
427 The applicants submit that the competitive advantages enjoyed by the Major Banks from greater scale, facilitates competition, not coordination. Moreover, they submit that those scale advantages do not otherwise diminish the lack of symmetry and alignment between the Major Banks.461
428 Next, they submit that a strategy to “price mid-market” in order to “deliver predictability of policy and process” does not amount to “accommodative pricing”. They submit a “mid-market price” is a competitive price and in any event, price is only one dimension of competition. They also submit that most of the symmetries and alignment referred to by the ACCC apply equally to smaller banks and, therefore, are not indicative of symmetry between the Major Banks.462
F.4.3.3. The ACCC’s submissions
429 The ACCC submits that although there are some asymmetries between the Major Banks, their ability to defray their costs over a much larger customer base and to raise capital more cheaply makes it difficult for other market participants to sustainably compete with them, particularly on price. It submits that the Major Banks benefit from [REDACTED].463
430 The ACCC submits that there is a level of symmetry and alignment between the Major Banks that is likely to make coordination between them feasible to initiate and sustain. It submits that they pursue similar business models targeted at supplying home loan propositions to all Australians with similar products,464 [REDACTED],465 similar credit ratings,466 similar lower capital requirements relative to their exposure than other competitors because of their AIRB approvals467 and each benefits from strong brand recognition.468
431 On balance, the Tribunal is satisfied that there remains considerable symmetries and alignment in the interests of the Major Banks.
432 The Tribunal accepts, however, that the following matters suggest that there are some material asymmetries between the Major Banks.
433 First, the market shares of the Major Banks vary significantly.
434 Both Mr Smith and Ms Starks included in their expert reports, calculations of market shares of the Major Banks in the national market for home loans from APRA monthly ADI data. Mr Smith calculated market shares based on average monthly shares between November 2021 and October 2022. The calculations revealed that ANZ has the smallest market share at 13%, approximately half of CBA’s share of 25.6%, significantly less than Westpac’s 21.4%, and less than NAB’s 14.5%.469 Ms Starks’ calculations of market shares of the Major Banks as at February 2023 were very similar. Ms Starks calculated (rounded to one decimal point) that ANZ had a market share of 13.2%, with CBA having 25.9%, Westpac having 21.4%, and NAB having 14.8%.470
435 Further, the market shares of the Major Banks have varied significantly since 2000. Comparisons of market shares that could be attributed to competition rather than acquisitions, in the period since 2000, are complicated by Westpac’s acquisition of St George Bank and CBA’s acquisition of BankWest. Relevantly, movements in the market shares of the Major Banks and other market participants in the period since 2000, are depicted in the following chart, which appears in the ACCC’s Reasons for Determination at [6.50]:471
436 The discontinuity after the dotted line reflects changes in the collection of data in mid-2019, being the inclusion of non-bank ADIs and changes in the definitions of various data items. After taking these discontinuities into account, it is apparent that CBA’s relative market share has increased since mid-2019, the market share of NAB has slightly decreased, and the market shares of both Westpac and ANZ have decreased more substantially than NAB’s market share.
437 The large increase in Westpac’s market share in 2010 is explained by its acquisition of St George Bank in 2008, and its move to being a single ADI. Similarly, the large increase in CBA’s market share in 2012 is attributable to the completion of its acquisition of BankWest in 2008.
438 It is apparent from the chart that (a) CBA’s market share relative to the other Major Banks, had declined in the period up to 2019 following its acquisition of BankWest, but has recently stabilised and increased somewhat, (b) NAB’s relative market share has decreased in the period since 2000 but has only decreased slightly since 2010 (having regard to the discontinuities in the data from mid-2019), and (c) the relative market shares of Westpac, since its acquisition of St George Bank, fell initially, was then stable for several years and more recently it has lost significant market share, and (d) the relative market share of ANZ had remained fairly static until it fell in 2020. Mr Campbell attributed this fall to problems that ANZ encountered with processing loan applications.472 Mr Campbell gave evidence that a [REDACTED].473
439 The movements in the market share of the Major Banks, in the period since the Westpac acquisition of St George and the CBA acquisition of BankWest, are depicted in the following table included in Ms Starks’ first report at [9.6]. The table depicts market shares in the home loans market for the period between 2012 and February 2023, based on APRA monthly ADI statistical data, including notional market shares for a combined ANZ/Suncorp Bank, and a combined Bendigo/Suncorp Bank.474 The table is subject to the same issues as the chart at [434] above, arising from the change in the denominator for market share calculations since mid-2019 in the APRA monthly ADI statistical data.
440 Second, no party suggested to the Tribunal that there had been any particular coordination on non-price aspects of competition. As Ms Starks concluded, there was no evidence of coordination on turnaround times,475 and there were significant differences in turnaround times from application to loan decision between the Major Banks.476
441 Third, there is at least some lack of symmetry among the Major Banks in their funding base, products and geographical diversity, some differences in the customer segments they focus on, and some variations in the size of their retail and business portfolios.
442 Notwithstanding these matters, overall, there remains a material symmetry and alignment in the interests of the Major Banks.
443 First, each of the Major Banks has a substantially greater market share than any other participant in the home loans market and has been in that position for many decades. The market shares of each of the two Major Banks with the lowest market share are still almost three times greater than Macquarie.
444 Second, the comparatively large market shares of the Major Banks gives them the benefit of significant scale, which is critical for funding the ever increasing fixed costs inherent in the supply of banking services. This includes, in particular, the investments in technology to drive efficiencies and to remain competitive both in terms of turnround times for home loan approvals, and to price competitively.
445 Third, each of the Major Banks has the same and higher credit ratings than other banks due, in part, to their D-SIB status, which gives it access to lower cost and a larger volume of wholesale funding than other banks. At the same time, by reason of their D-SIB status, each of the Major Banks is subject to the same prudential capital holding requirements, unlike other banks. Each of the Major Banks is also subject to the Major Bank Levy.
446 Fourth, each of the Major Banks monitor each other more closely than other banks, as illustrated in the following documents.
447 The board papers for the [REDACTED] board meeting on [REDACTED] June 2021 included a paper entitled [REDACTED].477
448 The pricing strategy of [REDACTED] was described in a [REDACTED] April 2022 memorandum to the board in the following terms:
[REDACTED], [REDACTED].478
449 An ANZ Exco update [REDACTED], that:
[REDACTED]. [REDACTED].479
450 Further, a paper reviewing ANZ’s variable home loan index rates in response to the change in the official cash rate at the RBA meeting in December 2022 [REDACTED].480
451 Fifth, each of the Major Banks has the benefit of strong brand recognition, an extensive branch footprint, is not as reliant on brokers as other banks, and offers a full range of banking products to all consumers.
452 Where rivals’ prices are clearly visible to each other, it will generally be easier to achieve a consensus on the terms of coordination, and to detect and punish cheating. In contrast, where prices are less observable, firms may be able to shave prices, and make substantial gains in market share before their cheating is discovered. Some sort of “focal point” such as a posted pricing or a “base rate” can assist in reaching a coordinated outcome or prices may also be adjusted in line with an input price index.
F.4.4.2. The applicants’ submissions
453 The applicants submit that there is no ability for the Major Banks to coordinate or punish deviation on price, given that the actual prices typically vary substantially from headline interest rates through discretionary discounts tailored to particular customers.481 They submit that the Major Banks only have a “delayed, imperfect, indirect and inferential”482 understanding of competitor pricing and the discount that might be offered to an individual customer by a competitor bank “can only be guessed at”.483
F.4.4.3. The ACCC’s submissions
454 The ACCC submits that headline interest rates for home loans are well-known because they are widely publicised.484
455 Next, the ACCC submits that the existence of discretionary discounts does not preclude coordinated conduct between the Major Banks. It submits that they have sufficient visibility as to the level of discretionary discounts to enable them to detect and respond to them because brokers, customers and potential customers provide feedback on competitors’ pricing.485
456 The widespread publication by the Major Banks of their standard variable rates for mortgages provides significant price transparency, and changes in their standard variable rates generally follow changes in the cash rate, which serves as a “focal point” for price adjustments.
457 The use of unpublished discretionary discounts, however, by the Major Banks in pricing home loans, necessarily detracts from price transparency. It can readily be accepted that the Major Banks monitor and take into account market intelligence that they might glean from brokers as to discretionary discounts offered by their rival Major Banks. Such market intelligence, however, falls well short of the immediacy and objectivity of published prices, such as the prices prominently posted on signboards at petrol stations, or online by airlines on their websites.
458 The Tribunal notes that there was some evidence that, at least, CBA may have moved away from offering unpublished discretionary discounts to consumers. In an ANZ board paper dated [REDACTED] December 2022 [REDACTED], it was stated that:
[REDACTED].486
459 More generally, prior to the more intense competition between the Major Banks during the COVID-19 pandemic, the ACCC included the following assessment of the prevalence of discretionary discounts, in its home loan price inquiry report in November 2020:
However, between 1 January 2019 and 31 December 2019, we observed two of the big four banks reducing or considering reducing their reliance on discretionary discounting to provide more price transparency to borrowers. A number of lenders, including non-big-four banks and non-ADI lenders, also do not offer, or are less reliant on, discretionary discounts.
The move towards greater price transparency has occurred without direct regulatory intervention. Rather, these developments have been primarily motivated by demand-side and other competitive pressures, with these lenders seeking to:
• simplify discounts to benefit the borrower experience
• increase their new lending volume
• adopt pricing structures similar to those that are offered by other lenders487
460 Ultimately, the impact of the increasing use of brokers on price transparency is likely to be more acute for borrowers than the Major Banks. A broker is able to provide a borrower with actual figures for comparative discretionary discounts offered by the Major Banks in circumstances where there is likely to be an alignment of interest between the broker and the borrower, not least because of the regulatory obligation on brokers to act in the best interests of their clients. Such an alignment of interests is less likely in a negotiation between a broker and a Major Bank where it could be expected that the broker would be pressing for a larger discretionary discount and would be providing information on other discretionary discounts in a manner that would tend to overstate rather than understate the availability and magnitude of those comparative discretionary discounts.
F.4.5. Consumer choice frictions
461 Where customers are sticky and less likely to switch suppliers, the incentive to undercut a coordinated outcome will be reduced. Customer stickiness may be a consequence of high search and/or switching costs, or result from brand loyalty or consumer behavioural bias and/or risk aversion. In contrast, active customer switching can encourage competition between suppliers, as they are more likely to gain from lowering prices.
F.4.5.2. The applicants’ submissions
462 The applicants submit that consumer choice frictions are not substantial and have reduced over time for the following reasons.
463 First, the use of brokers has led to a reduction in search and switching costs and has facilitated price competition. The applicants submit that brokers are an important driver of competition in the home loans market because they facilitate price transparency for consumers and identify opportunities for repricing and refinancing. They submit, by reference to Mr Campbell’s estimates, that brokers account for a large and increasing share of customer acquisition, particularly for many smaller banks whereby brokers account for at least half, with estimates of up to 70% of the total market, including [REDACTED]% for ANZ, [REDACTED]% for Suncorp Bank and an estimated 90-95% for Macquarie and [REDACTED] respectively.488 Further, they submit that the imposition on brokers of an obligation to act in customers’ best interests, combined with brokers’ incentive to encourage switching because of their remuneration arrangements, has further increased the likelihood of customer switching.489
464 Second, regulatory reform has made switching easier. The Consumer Data Right (CDR) as a component of the Open Banking reforms enables a consumer to share securely their financial data, including transaction history and account information, with accredited third parties. Hence, a consumer applying for a mortgage can give their consent to a bank accessing their banking details which streamlines the loan application and approval process.490
465 Third, banks have encouraged switching through streamlined switching processes (including ANZ’s “simpler switch”), cashback offers and introductory rate pricing.491 They submit that consumer choice frictions are not substantial is evidenced by increased refinancing and repricing492 and the fact that a significant proportion of customers have their home loan with a lender that is not their MFI.493 For example, in the three years to December 2022, refinancing increased by an estimated [REDACTED]% to ANZ and [REDACTED]% away from ANZ and repricing requests and price reductions [REDACTED] over equivalent periods from 2019 to 2023. Further, static market shares obscure the rate of churn from refinancing, repricing and repayment. For example, ANZ must replace [REDACTED]% of its loan book annually to maintain market share and if it does not, it will lose market share.494
F.4.5.3. The ACCC’s submissions
466 The ACCC submits that despite the increasing use of brokers, actual switching rates remain low. It submits that a large proportion of customers of Major Banks consider them to be their MFI495 and do not consider switching their home loan from that institution.496 In this regard, it cites evidence that 66% of customers have their home loan with their MFI.497 It submits that this reluctance is reinforced by the scale, longevity and perceived safety and stability of the Major Banks compared with smaller providers.498
467 Further, the ACCC submits that the data relied upon by the applicants, summarised at [465] above, is misguided. For example, ANZ figures are based on external refinancing as a proportion of the total value of new loan commitments and external refinancing activity, rather than as a proportion of total credit outstanding. The attrition rate of [REDACTED]% for ANZ’s loan book also relevantly includes attrition resulting from property sales and principal repayment and is, therefore, not an accurate measure of switching behaviour.
468 The Tribunal accepts that, as a general proposition, brokers have reduced search and switching costs and facilitated price transparency for consumers.499 Whilst the ability for price transparency to facilitate price competition is limited by banks’ ability to provide unpublished discretionary discounts, the Tribunal accepts that brokers can provide greater price transparency for individual consumers, and are a significant distribution channel for banks, particularly second-tier banks and other ADIs.
469 The Tribunal, however, shares the concerns expressed by the ACCC about the data relied upon by the applicants to demonstrate the prevalence of switching in the home loans market. The significance of external refinancing to the competitive dynamics in the market is better understood as a proportion of total credit outstanding, not as a proportion of new loan commitments. Further, the time series data on the absolute level of refinancing is potentially misleading, in that it focuses on the dollar value of refinancing. Given the recent significant increases in house prices, this also has the tendency to overstate increases in switching behaviour. A more reliable guide would be annual levels of home loan external refinancing as a proportion of total home loans outstanding, as presented in figure 11 of the ACCC’s Reasons for Determination over a 15-year period.500
470 Moreover, recent increases in switching are likely to be attributable, in large part, to the initial attraction of fixed rate home loans for relatively long periods during the sustained period of low interest rates, and then the decision by most borrowers to revert to variable home loans when their fixed rate terms expired, during a period in which interest rates were significantly and quickly increasing.
471 The Tribunal notes that the CDR was intended to facilitate consumers’ ability to search for and switch to alternative lenders’ offerings. The parties, however, did not take the Tribunal to any evidence that it had a material impact on switching behaviour.
472 Further, although Mr Campbell gave evidence that research by Roy Morgan revealed that the percentage of borrowers who were prepared to obtain a home loan from a lender other than their MFI increased from 27% in 2017 to 34% in 2021,501 the position as at 2021 remained that 66% of borrowers were not prepared to look beyond their MFI. As at August 2022, the Major Banks were the MFI for 73.7% of borrowers.502 Given that the dominant position of the Major Banks as the MFI of borrowers, the impact of brokers on switching behaviour remains relatively less significant than submitted by the applicants.
473 A survey conducted by the Australian Mortgage Council that was published in December 2022 revealed that the main reasons home loan borrowers who took out a loan in the last 5 years from one of the Major Banks, due to a recommendation from a broker (27%), was decreasing, and because of an existing banking [REDACTED], was increasing.
474 At the same time, the survey revealed that home loan borrowers who took out a loan in the last 5 years from a regional bank due to a recommendation from a broker [REDACTED], was increasing and an existing banking relationship [REDACTED], was decreasing.503
475 Finally, in assessing the likelihood of switching, it is necessary to recognise that while the “receiving lender” has incentives to make refinancing easier, the “losing lender” does not have such incentives, and the cost of discharging loans is an onerous process that deters facilitating refinancing.
F.4.6. Barriers to entry and expansion
476 Where entry or expansion is a significant threat going forward, it is likely to undermine coordination. Successful coordination will raise margins above competitive levels and attract the entry or expansion of new rivals, lowering prices and margins, and shifting market shares. Expectation of significant entry or expansion is likely to reduce the future payoffs from coordination and encourage more competitive behaviour to retain market share.
477 Where new entrants, however, are likely to remain on the fringes of the market, and not have the capacity to significantly expand, this will be insufficient to undermine coordinated outcomes. Entrants may also confine themselves to particular customer groups, and not extend their competitive influence across the broader market.
F.4.6.2. The applicants’ submissions
478 The applicants submit that banks can and have entered and expanded successfully in the home loans market. They submit that barriers to entry are low and barriers to both entry and expansion are declining because regulatory requirements for entry are not insurmountable,504 branch presence is no longer necessary to compete505 and new entrants are not burdened by legacy information technology systems and can take advantage of more efficient and cost-effective open API and cloud-based technology.506
479 In this regard, the applicants point to what they characterise as Macquarie’s meaningful and recent rapid growth, whereby it has increased its market share by 3.22 percentage points since 2017 (a 179.84% increase), and submit that it is growing increasingly quickly (five times the aggregate growth of all ADIs, and 18% of total market growth), all without a branch network.507 They submit that to refer to Macquarie’s growth as occurring over a 10-year period is a mischaracterisation of its success, particularly because it switched from a white label origination model to focus on its current direct model at the end of 2018.508
480 They also submit that numerous non-bank lenders that have entered the market are competing effectively509 and the Major Banks have invested significantly in developing new information technology systems to responding to innovation by new entrants and existing market participants, such as fully automated application and assessment procedures.510
481 The applicants submit that the internal documents of the Major Banks demonstrate [REDACTED].511
F.4.6.3. The ACCC’s submissions
482 The ACCC submits that the advantages of scale in banking conferred on the Major Banks, in particular, lower costs of capital, lower prudential capital requirements and more favourable credit ratings, make it difficult for new entrants to compete in a manner that meaningfully constrains the Major Banks.512
483 The ACCC also submits that, contrary to the applicants’ submissions, Macquarie’s experience highlights that barriers to entry and expansion are high. It submits that Macquarie was only able to achieve a “mid-single digits” market share after 10 years of significant capital investment and only by targeting a specific low risk category of customers and not seeking to compete “across the board”.513
484 Ms Starks concluded that while barriers to entry in the home loan market were low, barriers to expansion were high, and although they had fallen in recent years, they were likely to remain higher for banks other than the Major Banks and Macquarie.514
485 The Tribunal shares Ms Starks’ view that barriers to expansion in the home loans market are likely to remain high notwithstanding the reduced importance of branch networks, recent entry into the market, and the success of Macquarie in increasing its market share.
486 The Tribunal accepts that the reductions in the market shares of the Major Banks over the past 10 years, the increase in the market share of Macquarie, and [REDACTED],515 suggest that barriers to entry and expansion have reduced.
487 Over the past 10 years, there has been a significant fall in the aggregate market share of the Major Banks in the home loans market. By reference to Ms Starks’ market share table reproduced at [439] above, the Major Banks had a market share of 85.26% in 2012 but by 2022 (using the December 2022 figures), their market share had fallen to 75.20%. After taking into account the change in the denominator in the market share table in the period from 2019, this represents an approximately 7 percentage point decline in the market share of the Major Banks, principally in the market shares of ANZ, NAB and Westpac. The Tribunal views that loss of market share as materially reducing, but certainly not eliminating, the prospect of coordinated conduct by the Major Banks in the home loans market. The ACCC’s submission that this reduction is relatively insignificant because it only amounts, on average, to a reduction of 1% a year (presumably intended to be a reference to one percentage point), spread across three banks, is not persuasive. It injects an unduly static approach to the analysis to focus on average annual reductions, rather than the extent of a reduction over a broader time horizon.
488 The significance of the aggregate fall in market share of the Major Banks is reflected in internal documents of the Major Banks. Those documents demonstrate [REDACTED].516 [REDACTED].517
489 Further, in a paper prepared in August 2019 for a NAB board strategy day (NAB 2019 Board Strategy paper),518 the market summary included the following assessment of the general banking competitive environment:
490 The NAB 2019 Board Strategy paper also included a slide entitled “DISRUPTION IS HERE. NAB’S COMPETITOR SET IS EXPANING, PUTTING PRESSURE ON MARKET SHARE AND MARGINS”. The competitor set was then divided into “TRADITIONAL BANKS” and “DISRUPTORS/PLATFORMS”. The Traditional Banks included Bendigo, Macquarie and Suncorp Bank and the Disruptors/Platforms included various entities under the headings, “Neobanks”, “Non-Bank Lenders”, “Non-Bank Fintechs” and “Big Tech”.519
491 More directly, for present purposes, the NAB 2019 Board Strategy paper recorded that the Major Banks had lost market share in the home loans market in the period from May 2017 to May 2019. The loss of market share, based on “APRA Banking Statistics, May 2019” was calculated to be, in aggregate, a loss of 1.5 percentage points. The largest contributors to the increase in the market share of “Others” was stated to be “Macquarie, HSBC & ING” (one percentage point in aggregate). It was also stated that further market share had been lost to non-bank lenders that were not regulated by APRA, such as Athena. Finally, it stated that “Disruptors are also entering the market with improved application processes and aggressive pricing”.520
492 [REDACTED]. [REDACTED]:
• [REDACTED].521
493 Notwithstanding the reductions in aggregate market share, it remains the case that the cost of implementing and updating information technology systems that are sufficiently robust and reliable to service large numbers of customers, remains a substantial hurdle and a significant barrier to expansion. The impact of this hurdle is further compounded for regional banks due to their smaller scale and higher costs per unit.522 This smaller scale means technology investments necessary to enhance the regional banks’ competitive positions have lower returns as costs are spread across a smaller customer base.523
494 Notably, the investment in technology by the Major Banks and Macquarie dwarfs the regional banks. In FY23, $924 million, or 46%, of CBA’s $1,998 million investment spend was incurred in relation to productivity and growth,524 largely directed to modernisation and digitisation of processes and systems,525 and $630 million was spent on risk and compliance.526 Similarly, of Westpac’s $1,922 million investment spend in FY23, $1,194 million was incurred on risk and regulatory costs, and $728 million on growth and productivity.527 Between 2020 and 2022, Macquarie spent between $400 to $500 million per annum on technology,528 and this spend is forecast to increase to $[REDACTED] by FY26.529 In comparison, (a) Bendigo incurred $202.7 million in total annual investment spend in FY23, with $112.7 million directed towards “Foundational Technology” and $61.4 million spent on risk and compliance,530 (b) Suncorp Bank incurred $[REDACTED] in total,531 and (c) Bank of Queensland’s annual investment spend totalled $326 million, however, this was inclusive of $82 million incurred due to the ME Bank integration.532
495 Having sufficient scale is critical, as Mr Elliott explained in his examination before the ACCC:
You know, we have to invest literally billions of dollars to stay ahead in competition. Whether it’s cyber security, whether it’s fraud and scam management, whether it’s digitisation of services and allowing people to do more and more and more things on their mobile phone. And so your ability to defray that cost over a bigger customer base gives you an advantage.533
496 Again, even after making due allowance for the likely inherent advocacy in his answer, the following evidence given by Mr Elliott in his ACCC examination about why he had assessed that SGL had decided to sell Suncorp Bank, highlights the importance of scale in acquiring the necessary technology to compete effectively:
[REDACTED] We are half the size of CBA. We are not much less, you know, it’s almost the same for Westpac. So we are also in that same struggle and the challenge of being able to invest At the scale required. We are not talking small amounts of money here. I mean, we have already spent over a billion dollars building the ANZ Plus platform, you know? When I started as CFO 10 years ago, ANZ would spend, like the other banks, $800,000, $900,000, $1 billion a year on new stuff. Yeah? Some of it's regulatory, a lot of it was customer benefit. Today, we are spending over $2 billion a year, yeah? And our revenue is lower, yeah? So it’s really that ability to invest at scale. So if we don’t if we are not successful here, we will battle on. But it will actually mean we will be less competitive, I think, because inherently, having a smaller base to defray that cost and get those benefits means it will take longer and it will be more difficult for us.534
497 The significance of scale and technology was also emphasised in an annual strategy review presentation to the Macquarie boards (Macquarie Group Limited and Macquarie Bank Limited) in November 2022. It was noted that rising interest rates and tough wholesale funding conditions had created a very challenging environment for “sub-scale players such as fintechs and small banks”, and the key strategic priorities included a statement that:
Ongoing investment in technology and data is required to deliver a differentiated customer experience, meet compliance requirements and increase efficiency. 535
498 The presentation also included an overview of the competitive landscape in which it was observed that the Major Banks have “invested heavily in digital transformation and delivering digital experiences”, competition has intensified due to a great focus on creating digital home loan products, the collapse of Xinja and Volt Bank highlight the difficulty for neobanks to build a scaleable business model, and, more generally, fintechs, neobanks and non-bank lenders “struggle to compete in the current funding environment”. 536
499 As explained at [113]-[115] and [119]-[121] above, the benefit of scale for the cost of funds, in particular, for wholesale funds, makes it more difficult for new entrants and smaller banks to compete with the Major Banks.
500 Macquarie’s remarkable growth over the past few years in its share of the home loans market does not establish that barriers to expansion in the market are low. The increases are from a very low base, and its market share is still only one third of the lowest market share of the Major Banks. Professor King attributes Macquarie’s success to two principal factors: good processes and a low cost of funds.537 The former is a reflection of a substantial investment in technology, for example, in 2023 alone it invested a total of $595 million.538 It has a credit rating only one notch below the Major Banks from S&P Global and Fitch, only two notches below the Major Banks from Moody’s,539 and it has IRB accreditation. Moreover, Macquarie built up its deposit base, being the cheapest source of funds, before expanding its home loan offering, and it is also in a largely unique position given its very significant financial resources and brand awareness in Australia. Importantly, Macquarie has pursued a strategy of targeting what might colloquially be described as the “low hanging fruit”. Macquarie has principally targeted its offering at borrowers with a loan to value ratio of less than 80% to PAYG taxpayers and relies heavily on brokers to originate loans, thus dispensing with the need to establish and maintain a branch or direct sales presence. It has not sought to compete more broadly across all sectors of the home loans market.540
F.4.7. Recent increased price competition
501 It was common ground before the Tribunal that there had recently been evidence of stronger competition in the home loans market.
502 The ACCC accepted in its Reasons for Determination that there was evidence of relatively strong price competition across 2021 and 2022 for new home loans, including widespread use of promotional and cashback offers to attract customers.541
503 The evidence considered by the ACCC in its Reasons for Determination included RBA data that indicated that the lending spread of the Major Banks narrowed across 2021 and 2022 but had remained approximately the same or had increased from at least 2016 to 2020.542
504 There was no agreement, however, as to whether this stronger competition was likely to continue, or as to the factors giving rise to the recent increase in competition in the home loans market.
F.4.7.2. The applicants’ submissions
505 The applicants contend that the recent price competition between the Major Banks, Macquarie and other smaller banks demonstrates that there is effective and presently vigorous competition in the home loans market. They contend that even if competition were to reduce in intensity, that would not suggest that present competition is an exception to any alleged past coordination. The long-term trend since at least 2018 is of increasing and strong competition driven by structural and behavioural changes. The applicants advance the following submissions in support of those two contentions.
506 First, the ACCC and Ms Starks wrongly suggest that competition in the market is unlikely to endure.543 The applicants submit that suggestion depends implicitly on the idea of a “return” to past coordination (or, as put by Ms Starks, a “propensity towards coordination”),544 which overstates the extent to which competition is attributable to transient macroeconomic factors. The applicants submit that it also gives insufficient weight to longer term trends of increasing competition, reflected in declining market shares and profitability measures including ROE and NIM.545
507 Second, the process of competition may ebb and flow, but the long term trend is one of increasing and strong competition since at least 2018.546 The applicants submit that structural and behavioural changes in the market have increased competitive intensity, including the growth of Macquarie (along with other banks), the entry and expansion of non-bank lenders and fintechs, the increased activity of mortgage brokers, and consumers increasingly seeking refinancing and repricing for lower interest rates.547 They submit that these longer term trends are reflected in data showing that Westpac, NAB and ANZ have lost market share over time (particularly to Macquarie)548 and that their ROE and NIM have progressively declined since 2000.
508 Third, the ACCC wrongly discounted the data that indicated that ROE and NIM for the Major Banks have progressively declined since 2000. They submit that the ACCC was mistaken in concluding that reductions in bank NIM could be explained by increases in equity ratios in response to tighter APRA capital standards and reductions in the cash rate target. They submit that the ACCC’s error was due to it misinterpreting RBA analysis, which instead demonstrated that NIM should increase, not decrease, with higher capital requirements.549 They submit that NIM has decreased despite higher capital requirements, is consistent with the proposition that competition has prevented ANZ and other larger banks from being able to maintain, let alone increase, prices to maintain ROE or NIM.550
509 The applicants also submit that the international comparisons of bank ROE and NIM relied on by the ACCC provide no useful insight into competition in the home loans market, including because of the absence of evidence of competitive conditions in those international markets.551 They submit that the ACCC likewise ignores or marginalises evidence from the internal documents of competitors which specifically attribute reductions in NIM to competitive intensity.552
510 Fourth, recent fluctuations in total mortgage lending by the Major Banks and other lenders indicate that, far from the Major Banks accommodating each other to protect the status quo, they are competing vigorously to win business. They submit that APRA data for the period June to September 2023 shows that recent reductions in cashback offers by some banks is not demonstrative of a lack of competition or evidence that competition is short lived.553
511 The APRA data to the end of September 2023 indicates that, in the period June to September 2023, the total mortgage lending increased by approximately $18.175 billion. Mr Campbell gave evidence that the total mortgage lending of CBA did not increase in line with the total market. If the status quo had been maintained CBA’s total home lending would have increased by approximately $4.6 billion. In fact, its total lending reduced by $4.102 billion. In contrast, the total mortgage lending of Westpac increased by $6.495 billion; that of NAB increased by $2.267 billion and that of ANZ by $5.433 billion. The total mortgage lending of Macquarie increased by $4.186 billion. The total mortgage lending of the regional banks (Bendigo, Bank of Queensland and Suncorp Bank) reduced minimally by $477 million, and the total mortgage lending of other institutions increased by $4.295 billion.554
512 The Tribunal was taken to a number of internal documents of banks by the applicants, which cite competitive pressure as a factor driving decreases in NIM. [REDACTED],555 [REDACTED].556 [REDACTED].557 NAB’s FY20-FY24 Strategic and Financial Plan noted NIM “is expected to decrease 5bps in FY20 due to continued competitive margin pressures in Housing Lending and the lower interest rate environment”.558
513 The applicants submit that this recent experience in the market confirms that all banks must remain competitive on dimensions of price, policy and process or they will lose market share.559
F.4.7.3. Bendigo’s submissions
514 Bendigo submits that recent evidence of increased competition in the home loans market between the Major Banks should be given no or little weight in the competition analysis, for the following reasons.
515 First, the evidence of recent competition in the market appears to be short term only. It submits, by reference to the opinion of Ms Starks, that the recent increase in pricing competition may be driven by a sudden increase in refinancing demand in response to the rapid increase in interest rates, after a long period in which rates were low. It submits that short-term suspension of some aspects of coordination is supported by reports in May 2023 that CBA, Westpac and NAB have announced that they are ending their cashback offers for home loans.560
516 Second, both Ms Starks and Professor King have given evidence that the home loans market is structurally conducive to coordinated conduct, irrespective of whether that conduct is apparent.561 It submits that the effects on competition of the proposed acquisition must, therefore, be understood in this context.562
517 Third, the evidence tends to show that to the extent that the Major Banks were at some point pricing more competitively, it is not sustainable for smaller banks to compete. Bendigo submits, by reference to Ms Baker’s transcript of examination, [REDACTED].563
F.4.7.4. The ACCC’s submissions
518 The ACCC submits that in assessing the significance of recent price competition in the home loans market, it is necessary to have regard to the long history of muted competition and the most unusual macroeconomic context in which recent competition has taken place. It submits that given the unusually large number of fixed-rate home loans written during the height of the COVID-19 pandemic that are now expiring and requiring refinancing, it cannot be presumed that the recent levels of competition will endure.564
519 The ACCC refers to statements by representatives of the Major Banks that recent competitive conditions are atypical, or may change and steps have already been taken to reduce the volume of capital allocated to home loans, cash backs and other incentives.565
520 The ACCC submits that the declining ROE and NIM achieved by the Major Banks since about 2000 cannot be attributed to increased competition. It submits that the pronounced fall in ROE in 2016 is largely attributable to the need for the Major Banks to raise additional equity to meet tighter prudential standards imposed by the RBA.566 It submits that it is unsurprising that NIM has reduced in the period between at least 2010 and 2022, given the cash rate has fallen over that time, because as was observed in an RBA discussion paper:567
Low rates can still affect profitability even when spreads remain constant. Bank’s net interest margins (NIMs) - the difference between their interest income and interest expenses (as a share of assets) will fall with interest rates if spreads remain constant. This occurs because some of banks’ assets are funded by equity. So as the returns on these assets fall, so does the return on equity (all else equal).
521 The ACCC submits that the preferred metric to assess long-term trends in the market for home loans is the lending spread, which shows whether, in fact, banks are choosing to reduce their own margins to win or retain customers.568 It points to an analysis undertaken by the RBA that shows lending spreads for the Major Banks remained approximately the same in the period from 2010 to 2020 and only reduced in 2021 and 2022.
522 Ms Starks observes that when “firms find themselves facing higher than average market demand, they will have an increased short-term incentive to deviate from coordination to capture a large share of that transient demand”.569
523 The Tribunal accepts that the recent volatility in interest rates is likely to have significantly increased demand for the refinancing of home loans and created an attractive environment to increase market share and thus reduced the prospect of coordination. The fiscal response to the COVID-19 pandemic created a large supply of relatively cheap money and led to a significant increase in low interest fixed term loans. This was then followed by a rapid tightening of the cash rate, and a mortgage “cliff”, as borrowers fixed term loans expired.
524 The Tribunal does not accept, however, that the recent increase in competition in the home loans market can be attributed solely to recent external economic challenges, and fluctuations in interest rates. The Tribunal is, therefore, not of the view that the market is necessarily likely to return to conditions more conducive to coordination once these matters have subsided. The Tribunal accepts, however, that other considerations less conducive to coordination, such as (a) the increasing use of brokers to source home loans, who are also subject to regulatory requirements to put consumers’ interests first, (b) the growth in the market share of Macquarie, (c) the increasing use of technology to both reduce banks’ turnaround times for applications, and allow consumers to compare offerings from different banks more readily, together with (d) the increasing use of electronic settlements through PEXA (an online digital settlement platform), are likely to remain and become even more significant in the foreseeable future.
525 The Tribunal also accepts that there has been a long-term trend of declining ROE and NIM, since 2000, for the Major Banks.
526 Mr Elliott gave evidence that Australian bank average ROE above 10-year bond rates has declined by 42% from 14.6% in 2007 to 8.4% in 2021,570ANZ’s cash ROE has declined from 19.6% in 2007 to 14% in 2015, to 11.9% in 2017, and to 10.4% in 2022, only marginally above ANZ’s cost of capital at 9.75%, as at September 2022.571 ANZ’s ROE on back book home loans [REDACTED]. ANZ’s ROE on front book home loans [REDACTED].572 Further, SGL observes that regional banks, such as Suncorp Bank, have generated lower returns than the major banks since the mid-2000s.573
527 Mr Elliott also gave evidence that ANZ’s NIM for its Australia division, including Australian Retail and Commercial Business, has declined from 2.75% in 2016 to 2.25% for retail and 2.10% for commercial in 2022.574 (575)
528 The long-term trend of declining ROE can be seen in the following RBA chart:576
529 The Major Banks average NIM has also declined from 3.3% in 1999 to below 2.0% in 2022, 577 as illustrated in the following RBA chart:578
530 These long-term declines in ROE and NIM represent a significant shift in returns and margins but the extent to which these shifts are attributable to exceptional macro-economic factors and regulatory reform, rather than competition, is difficult to discern. The two considerations, however, are not mutually exclusive. It is much easier to coordinate a pass through to borrowers of an increase in the cash rate, than to coordinate a pass through of costs associated with regulatory reform, such as changes to prudential standards, or compliance costs. Mr Elliott gave evidence that ANZ has not been able to offset increases in regulatory costs associated with increased capital and prudential requirements, the CDR and the Major Bank Levy, with increased revenue.579 An ANZ strategy paper, prepared for a meeting in May 2023, attributed approximately [REDACTED]% of the [REDACTED] in home loan ROE to regulatory interventions and approximately [REDACTED]% to increased competition.580 The Tribunal accepts, on balance, that a not insignificant quantity of the shifts in returns and margins is due to increased competition but the dominance of the Major Banks has not been materially diminished.
531 The profitability of the Major Banks remains amongst the highest in the world for major banks, as demonstrated in the following graph prepared by the ACCC from data provided by ANZ pursuant to a request for information dated 21 June 2023:581
532 The following assessment of the current conditions in the home loans market in the SGL February 2022 Analysis, as at early 2022, points to the dominant market position of the Major Banks:
The banking sector in Australia in 2016 was broadly similar to today, with major banks aggressively taking market share from regionals and challengers – over 80% of gross loans were provided by the major banks in 2016. The only sign of disruption of market scale has been Macquarie in mortgages, but despite a level of “noise” this has been at the margins with its share increasing from 1.8% in 2016 to 3.3% by the end of 2021.582
533 Macquarie has since increased its market share of the home loans market to over 5% but the observation that the only sign of disruption was from Macquarie and, even then, only at the margins, is telling.
534 The Major Banks have also recently released their FY23 results. Each of the Major Banks reported overall increases in ROE since FY22. The increases reported were 54 basis points for ANZ,583 120 basis points for NAB,584 130 basis points for CBA,585 and 199 basis points for Westpac.586
535 At the same time, however, each of the Major Banks and Macquarie reported declines in their NIM on home loans.
536 NAB reported a 7 basis point decline in its NIM between FY22 and FY23, which it attributed primarily to home lending competition, and its key considerations identified for its NIM for the first half of FY24 included:
• Home lending (HL) competition headwinds expected to continue.587
537 In a similar vein, CBA reported in its FY23 presentation that “Continued competitive pressures” for “domestic home loans” had seen a six basis point reduction in group margin between the first and second halves of FY23.588
538 Similarly, Westpac reported a reduction of 10 basis points in its NIM on home loans between 2022 and 2023,589 ANZ reported a 5 basis point decline,590 and Macquarie reported “[i]mproved average margins from the rising interest rate environments” but they were “offset by changes in portfolio mix and ongoing lending competition”.591
539 Ms Starks states that the conduciveness of the home loans market to coordinated conduct arises from price transparency, the concentration in the market power of the Major Banks, ease of communication in pricing changes, multi-market contact, a frequency of interaction between competitors, entry and expansion barriers, consumer choice frictions, and a general lack of innovation.592
540 The Tribunal is satisfied that the home loans market is conducive to coordination but that the conditions for coordination have recently reduced and are likely to continue to reduce in the foreseeable future.
541 The following characteristics of the home loans market suggest that the market is conducive to coordination.
542 First, the degree of concentration in the home loans market and the significant market shares of the Major Banks, makes the market particularly vulnerable to coordination. The Major Banks have significantly more in common with each other than with other banks, as discussed at [442]-[451] above.
543 Second, the transparency of the RBA cash rate, and the publication by the Major Banks of their standard variable rates for mortgages, also make the home loans market vulnerable to coordination.
544 Third, both the Productivity Commission in its 2018 report,593 and the ACCC in its 2018 Residential Mortgage Price Inquiry report,594 identified historical pricing conduct by the Major Banks in the home loans market, that they concluded was consistent with coordinated conduct.
545 Fourth, the recent increase in price competition in the market is attributable, at least in part, to the exceptional volatility in interest rates, giving rise to increased market demand that could be expected to increase the short-term incentives to deviate from coordination, to capture a larger market share.
546 The degree to which the home loans market is subject to coordination, however, is tempered by the following characteristics of the market.
547 First, there is a material asymmetry in the market shares of the Major Banks and their respective market shares have not remained constant. CBA has increased its market share in the period since 2000, whilst the other Major Banks have lost market share. The market shares of Westpac have decreased more substantially than the market shares of NAB and ANZ, during the period since 2000.
548 Second, Macquarie has recently acquired a significant market share, albeit from a low base, and together with smaller players such as HSBC and AMP Bank, has reduced the market shares of the Major Banks. The collective market share of the Major Banks has declined by some 7 percentage points over the last decade, principally lost to Macquarie.
549 Third, as a general proposition, the increasing use of brokers in the home loans market has reduced consumer choice frictions and facilitated greater switching. Automation of settlement processes through the use of PEXA has also reduced settlement frictions.
550 Fourth, the ability of the Major Banks to detect deviations from any coordinated conduct remains constrained by the prevalence of unpublished discretionary discounts on headline interest rates. Any information provided by brokers to a Major Bank as to the prevalence and quantum of discretionary discounts offered by other Major Banks must, necessarily be incomplete, involve some delay, and be subject to some scepticism.
551 Fifth, barriers to entry to the home loans market have reduced because (a) the importance of a branch presence has declined because the majority of home loans are now sourced through the broker channel, and (b) digital and technology platforms have become key sources of differentiation. While scale remains important to be able to invest in technology, new entrants are not encumbered with legacy systems, and can invest in new platforms at lower cost.595 Nevertheless, the cost of establishing the necessary technology systems to be able to offer a competitive product in the home loans market, represents a significant sunk cost, and, thus, a material barrier to both entry and expansion, irrespective of whether a firm must replace a legacy system.
F.5. Would the Proposed Acquisition increase the likelihood or sustainability of coordination in the home loans market?
552 The relevant competition inquiry for the Tribunal to undertake, in the home loans market, is to determine whether the Proposed Acquisition would not be likely to have the effect of substantially lessening competition in the home loans market, having regard to an increase in the likelihood of coordination between the Major Banks.
553 In order to undertake that inquiry, it is first necessary to identify any change in competitive constraints, which might affect the likelihood of coordination between the Major Banks in the home loans market, on the assumption that the Proposed Acquisition proceeds. It is then necessary to compare that position with competitive constraints that would likely be imposed on ANZ in each of the No Sale counterfactual and the Bendigo Merger counterfactual, that might affect the likelihood of coordination.
F.5.2. Likelihood of increase in coordination if the Proposed Acquisition proceeds
F.5.2.1. The applicants’ submissions
554 The applicants submit that any contention that an acquisition by the fourth largest bank of the ninth largest bank would lead to a structural change that would increase the propensity to coordination is “highly speculative and without foundation”, 596 for the following reasons.
555 First, the Proposed Acquisition would result in a de minimis increase in ANZ’s market share in the order of only 2.3 percentage points and thus would not materially increase the symmetry in market shares between ANZ and the largest bank, CBA (25.6%).597 The applicants submit that even with the Proposed Acquisition, ANZ would remain the smallest of the Major Banks, in overall banking system assets and marginally overtake NAB (14.5%) to be third in home loans (15.3%) behind CBA (25.6%) and Westpac (21.4%).598
556 Second, asymmetries in other important competitive attributes such as turnaround times and credit policy would continue to undermine the stability of coordination. The applicants submit that the Proposed Acquisition would also not materially alter ANZ’s funding base or focus, nor would it be likely to increase materially the degree of symmetry in cost structures so as to increase ANZ’s incentives to coordinate.599
557 Third, the Proposed Acquisition is not likely to change ANZ’s incentives to compete as it has every incentive to retain the Suncorp Bank customers that would otherwise be targeted by ANZ’s competitors.600 The applicants submit that ANZ’s incentives to maximise the prospect of retaining and growing Suncorp Bank customers means it is likely to offer those customers equal or better systems, features, pricing or other benefits.601
558 Fourth, the removal of Suncorp Bank would not have any material impact on competition in the home loans market because it is not a vigorous or effective competitor. The applicants submit that Suncorp Bank’s market share is “small, relatively stagnant and [REDACTED].602 They also submit that Suncorp Bank lags behind other banks in non-price competition, including because it has [REDACTED] average turnaround time, is poorly ranked in online and mobile banking, and [REDACTED] in the technology required to substantially improve its performance.603
559 Fifth, the Proposed Acquisition would not reduce the constraint exerted by other smaller banks, particularly Macquarie. The applicants submit that the ACCC’s contention that Macquarie would not undermine coordination but an enlarged Bendigo would, is untenable given Macquarie’s recent pricing and growth.604
F.5.2.2. The ACCC’s submissions
560 The ACCC advances the following submissions on the impact of the Proposed Acquisition on the likelihood of coordination in the home loans market.
561 First, the Proposed Acquisition will increase concentration in the home loans market and reduce the level of asymmetry among the Major Banks.605
562 Second, the increase in ANZ’s share of the home loans market will mean that it has “more to lose and less to gain” from any deviation from coordinated behaviour by the Major Banks.606 The ACCC submits that a key reason ANZ is prepared to pay $4.9 billion in the Proposed Acquisition is because it is “the equivalent of many years of organic system growth”, as previously stated by Mr Elliott.607
563 Third, the Proposed Acquisition will also increase symmetry between ANZ and the other Major Banks in other respects. The ACCC submits that it will result in a “material” 22% increase in ANZ’s retail deposits, making its retail deposits percentage closer to the other Major Banks in circumstances where Mr Elliott has given evidence that ANZ’s funding base is currently “very different”.608 It submits that the Proposed Acquisition will also result in an approximate [REDACTED]% increase in ANZ’s Australian retail and commercial earnings, which is another way ANZ will become more like the other Major Banks.609
564 The Tribunal does not consider that the Proposed Acquisition would materially increase the prospect of coordination between the Major Banks in the home loans market.
565 First, the Proposed Acquisition would not lead to any substantive change in the structure of the market. ANZ’s market share would only rise from 13.16% to 15.56%, using the most recent market shares figures in Ms Starks’ table reproduced at [439] above.610 The increase would see ANZ move from fourth to third in market share, just above NAB with a 14.76% share. The substantial asymmetry in the market shares of the Major Banks would remain, given the much larger market shares of CBA (25.92%) and Westpac (21.35%).611
566 Second, the increase in retail deposits held by ANZ will not materially move its funding base closer to the other Major Banks. The ACCC’s reference to a 22% increase in retail deposits is apt to mislead. The source relied on by the ACCC for that proposition makes clear that it is a 22% increase in total deposits from $174 billion to $179 billion compared with a 17% increase in mortgages.612 Retail deposits as a percentage of mortgages would only increase from 53% to 55%.613 The Tribunal is not persuaded that this movement would materially increase the likelihood of coordination in the home loans market. In its Reasons for Determination, the ACCC was only able to point to the evidence of Mr Elliott that the diversity in ANZ’s funding base influenced how it approaches its strategy for attracting deposits and funding home loans, and then speculate that:
the ACCC considers the Proposed Acquisition may have an impact on the way that ANZ competes to retain and build market share in the home loans market. As it stands, ANZ's net position from its commercial and institutional banks (deposits minus lending) currently provides a source of funding for home loans that may not be available to other banks to the same extent as ANZ. By comparison, the Proposed Acquisition means ANZ’s funding base for home loans will look more comparable to the other major banks, as ANZ’s alternate source of funding is likely to be reduced and ANZ will move to having a higher proportion of funding from retail deposits compared to its previous position. While it is uncertain precisely what impact this would have on the strategies that ANZ engages in to compete for deposits and home loans, it may operate to further align ANZ’s incentives with the other major banks, and increase ANZ’s incentives to coordinate in home loans.614
(Footnotes omitted.)
567 The concerns expressed by the ACCC do not rise above conjecture.
568 Third, the Proposed Acquisition would not otherwise have any material impact on any of the other conditions for coordination in the home loans market. It would not alter the conditions for price transparency, or consumer choice frictions.
569 Fourth, the removal of Suncorp Bank is unlikely to have any material impact on the likelihood of coordination as the Tribunal is satisfied, for the reasons advanced below in addressing the No Sale counterfactual, that Suncorp Bank is not a vigorous or effective competitor in the home loans market.
F.5.3. Likelihood of comparative decrease in coordination in the No Sale counterfactual
F.5.3.1. The applicants’ submissions
570 The applicants submit that there would be no relevant difference in Suncorp Bank’s position in the No Sale counterfactual. They submit that Suncorp Bank’s position will continue to be modest, commensurate with its small and relatively static market share, and it will pose no disruptive constraint to ANZ or the other Major Banks, for two principal reasons.615
571 First, SGL’s organic strategy for Suncorp Bank is [REDACTED] market share. The applicants submit that in recent history, Suncorp Bank has achieved only modest growth and it revised its FY23 growth target [REDACTED].616 In this regard, they submit that Suncorp Bank’s current organic strategy, set out in the Suncorp Bank Business Plan FY23-FY25 (Suncorp Bank FY23-25 Plan) and as will be set out in its organic strategy for FY24-26, forecasts a [REDACTED].617 They submit, relevantly, that the forecasts do not contemplate Suncorp Bank [REDACTED], having achieved only 0.12% growth in home loans market share under the FY21-23 business plan.618
572 Second, SGL’s organic strategy for Suncorp Bank does not cater for [REDACTED]. The applicants submit that the Suncorp Bank FY23-25 Plan exposes the need for transformation but [REDACTED].619 They submit that SGL’s organic plan for Suncorp Bank [REDACTED].620
F.5.3.2. The ACCC’s submissions
573 The ACCC submits that [REDACTED], the Suncorp Bank FY23-25 Plan forecasts a future of improving competitiveness [REDACTED] to Suncorp Bank’s technology.621 It submits that the evidence establishes that Suncorp Bank sees itself as capable of growing its business and increasing its profitability with only “modest progress” toward improving its current technology.622
574 The ACCC submits that the Tribunal should consider whether SGL’s [REDACTED] reflects, in part, its disposition to sell Suncorp Bank since at least 2018.623 It submits that the Tribunal may infer that SGL’s organic plans for Suncorp Bank would likely change in the No Sale counterfactual, particularly because the evidence suggests that the decision [REDACTED] has been a choice rather than an imperative.624
F.5.3.3. The applicants’ submissions in reply
575 The applicants submit that the ACCC is wrong to contend that Suncorp Bank would materially increase its competitiveness in the No Sale counterfactual, for the following reasons.
576 First, there is no “disconnect” between SGL’s submissions and its internal business records. The applicants submit that the Suncorp Bank FY23-25 Plan describes a bank that successfully arrested a precipitous decline and aspires to target [REDACTED] of previously lost market share in home loans. They submit that the targeted home loan growth rates were “aspirational goals rather than easy goals”, which have not been met thus far.625 They also submit that Suncorp Bank actively moderated its home loan growth in FY23, and the draft Suncorp Bank FY24-FY26 business plan [REDACTED].626
577 Second, the ACCC essentially ignores Dr van Horen’s evidence, which explained the challenges confronted by a small bank competing against larger banks in an increasingly capital-intensive industry. The applicants submit that this evidence must be accepted in circumstances where the ACCC examined the witness, embraced his evidence, and then took it further.627
578 Third, Suncorp Bank’s [REDACTED] level investment [REDACTED] is not a mere preference or consequence of an [REDACTED] borne of a “predispose[ition] to selling Suncorp Bank”.628 The applicants submit that Dr van Horen made clear in his s 155 examination that SGL [REDACTED] because it has committed to investors that it will not compromise ROE, CI ratios and other key metrics.629
579 Ms Starks concluded that it would be unlikely that the removal of Suncorp Bank as an independent competitor would reduce competitive constraints on the Major Banks to such an extent that it would make coordination more likely, effective or stable, because Suncorp Bank is not a particularly effective competitor in the home loans market.630
580 The Tribunal is satisfied that in the future with the Proposed Acquisition, when assessed against the future without the Proposed Acquisition, in the No Sale counterfactual, that the Proposed Acquisition would not be likely to have the effect of, substantially lessening competition in the home loans market.
581 As submitted by the applicants, Suncorp Bank has recently only achieved modest growth in its small and relatively static market share. SGL’s current organic strategy for Suncorp Bank does not envisage [REDACTED], and as submitted by the applicants, it appears to [REDACTED] to defend Suncorp Bank’s present market share. There was no evidence before the Tribunal from which it could be inferred that Suncorp Bank would likely impose any material constraint on the Major Banks in the future in the No Sale counterfactual.
582 The Tribunal accepts that given SGL had been looking at a possible divestiture of Suncorp Bank since at least 2016, its incentives to commit to additional technology expenditure that might be considered necessary for it to defend and increase market share and profitability, might be significantly reduced. Certainly, once the Proposed Acquisition was announced it would be surprising if strategy papers recommending or reviewing the need for significant additional technology expenditure were prepared and presented to the SGL board.
583 What is striking, however, is that in the potential divestment strategy papers presented to the SGL board, prior to the announcement of the Proposed Acquisition, an organic strategy for Suncorp Bank was invariably presented as an alternative to any divestment option.631 The analysis of alternative options in the strategy papers made clear that any divestiture of Suncorp Bank had to be value accretive to SGL and its shareholders, compared with its retention.632 In that context, there was little incentive for SGL not to factor into its valuation of the organic option for Suncorp Bank, likely achievable future technology expenditure and the likely benefits that could be obtained from that expenditure in terms of market share and profitability.
584 The extent of SGL’s [REDACTED], in order to enable it to compete effectively in the home loans market, was highlighted by Dr van Horen in his s 155 examination. Dr van Horen readily accepted a proposition put to him by the ACCC in his s 155 examination that the technology expenditure challenges facing Suncorp Bank had been [REDACTED] in the Suncorp Bank FY23-25 Plan.633 Moreover, Dr van Horen also readily accepted propositions put to him by the ACCC, to the effect that Suncorp Bank [REDACTED].634
F.5.4. Likelihood of comparative decrease in coordination in the Bendigo Merger counterfactual
F.5.4.1. The applicants’ submissions
585 The applicants submit that a merged Bendigo/Suncorp Bank is no more likely to increase ANZ’s incentives to compete or to constrain any coordination in the home loans market by the Major Banks, for the following principal reasons.
586 First, a merged Bendigo/Suncorp Bank will not be a more effective competitor, let alone strong enough to result in a material loss of market share sufficient to disrupt any coordination, that would not already be disrupted by competition from Macquarie and other banks. The applicants submit that it is wholly speculative that a merged Bendigo/Suncorp Bank would win materially more market share compared to the factual, by offering a “different business model” or targeting “different niches in the competitive fringe”.635
587 The applicants further submit that the increased scale that Bendigo would obtain from acquiring Suncorp Bank would not materially increase its effectiveness as a competitor.636 They submit that although increased scale would give the merged entity a larger base over which to spread its fixed costs and enable greater investment in technology, any synergies from the merger would be delayed for at least four to six years.637 In addition, they submit that any merged entity would have to address various information technology legacy issues prior to receiving the benefits of any synergies.638 Nor, they submit, given the relatively small increase in scale compared with the size of the Major Banks could it simply be inferred that the increase in scale would enable it to fund transformative technology that would enable it to provide a greater constraint on the Major Banks.639
588 The applicants also submit that slightly greater scale alone is not sufficient to create a vigorous and effective competitor.640 They submit that in order to compete effectively, other banks, such as Macquarie in home loans, Rabobank in agribusiness banking and Judo Bank in SME banking have offered compelling and disruptive strategies.641 They submit that there is no evidence that a merged Bendigo/Suncorp Bank would implement a changed, compelling strategy642 and any contention that a merged Bendigo/Suncorp Bank would win materially more market share compared with the factual, by offering a “different business model” or by targeting “different niches in the competitive fringe”, is wholly speculative.643
589 Second, the Proposed Acquisition does not have the effect of preventing any pro-competitive impact from Bendigo obtaining greater scale.644 They submit that the potential for Bendigo to merge with another bank, such as Bank of Queensland or ING in a future with the Proposed Acquisition must also be considered in this context.645 They submit that given their similar comparative market shares, a merger between Bendigo and either Bank of Queensland or ING would exert a similar degree of competitive pressure on ANZ in a future with the Proposed Acquisition as a merged Bendigo/Suncorp Bank in a world without the Proposed Acquisition would. Thus, they submit that any competitive constraint offered by a merged Bendigo/Suncorp Bank in the Bendigo Merger counterfactual could be expected to be replicated by a merged Bendigo/Bank of Queensland or a merged Bendigo/ING in a future with the Proposed Acquisition.646
590 Relatedly, the applicants submit that if the Tribunal were to find it plausible that Bendigo would materially increase its competitiveness by acquiring Suncorp Bank in the counterfactual, it should also find it plausible that Bendigo would achieve a similar result through a different acquisition in the factual. If so, they submit, the Bendigo Merger counterfactual would add nothing to the competition analysis.647
591 Third, a merged Bendigo/Suncorp Bank would incur a significant cost dis-synergy from the change in Suncorp Bank’s credit rating and having to pay the Major Bank Levy. It submits that this would be in the order of $[REDACTED] million to $[REDACTED] million per year in additional funding costs and at least $[REDACTED] million per year for the Major Bank Levy. They submit that this would reduce (rather than improve) Bendigo’s ability to offer competitive interest rates on loans.648
592 Fourth, achieving IRB status is presently speculative for Bendigo, and its prospects would not be improved by acquiring Suncorp Bank.649 In this regard, the applicants submit that any additional retail deposits that might be attracted because of a perception that the larger entity would be a safer investment, would have to be sourced at additional cost as it is not plausible that the merged entity would receive a credit rating uplift.650
593 The applicants further submit that it is uncertain and speculative whether Bendigo’s competitiveness would improve even if it achieved IRB accreditation. They submit that the merged Bendigo/Suncorp Bank would incur a substantial day one capital impost if it were to obtain IRB accreditation, that would take nine years to recoup. Hence, they submit that it is not credible to suggest that Bendigo’s position would be different if it were to obtain IRB accreditation or even if accreditation enabled it to offer lower rates on low-risk loans, Macquarie is already a vigorous and effective competitor for such loans.651
F.5.4.2. Bendigo’s submissions
594 Bendigo submits that a merged Bendigo/Suncorp Bank would operate as a substantially enhanced competitive constraint on the Major Banks and decrease future coordinated conduct by them. It submits that in contrast to Macquarie’s focus [REDACTED]. 652
595 Bendigo submits that if it were able to achieve half the growth achieved by Macquarie in the last decade, which has increased its market share by a factor of four in that time, a merged Bendigo/Suncorp Bank would achieve a market share of 10.4%. Bendigo observes that Professor King expects that some and potentially most of that increase in market share would come from the Major Banks and that any such steady decrease in the market share of the Major Banks would undermine their ability to engage in coordinated conduct.653
596 Bendigo also observes that both Professor King and Ms Starks concluded that the Proposed Acquisition will substantially lessen competition in the home loans market in the future compared with the Bendigo Merger counterfactual.654
597 Bendigo submits that in a future with the Bendigo Merger counterfactual, Bendigo’s market share in the home loans market would almost double from 2.8% to 5.2%.655 It submits that the merged entity would become the fifth largest competitor in that market, and the number of institutions with a greater than 5% market share would expand from five to six.
598 Bendigo submits that the merged Bendigo/Suncorp Bank would be in a position to invest, and spread its costs over a larger customer base, enabling it to offer more competitive pricing. It submits that the Bendigo Merger counterfactual would represent a substantial change in market dynamics and substantially increase the ability of the merged entity to exert competitive constraint on the Major Banks. It submits that it would enable the merged entity to provide a stronger regional bank alternative to Australian consumers and would likely result in substantially greater competition on price, service and innovation when compared with the Proposed Acquisition.656
599 Bendigo submits that the scale of a merged Bendigo/Suncorp Bank would enable Bendigo to increase investment in innovation and technology, and accelerate the delivery of Bendigo’s digital capabilities by spreading its fixed costs across a larger balance sheet.657 [REDACTED].658
600 Bendigo submits that the increased scale provided by the Bendigo Merger counterfactual would see an immediate increase in the merged entity’s national share of deposits from 2.5% to 4.3%.659 It submits that this would enable Bendigo to [REDACTED] and the merged entity would become a more attractive alternative bank for deposits.660 It submits that a merger with Suncorp Bank, would enable the merged entity to “leverage its enhanced scale to increase its access to lower cost funds such as deposit funding”.661
601 The Tribunal is also satisfied that the Proposed Acquisition, when assessed against the Bendigo Merger counterfactual, would not be likely to have the effect of substantially lessening competition in the home loans market.
602 In the future without the Proposed Acquisition, in the Bendigo Merger counterfactual, the Tribunal is not persuaded that a merged Bendigo/Suncorp Bank would reduce the likelihood of coordinated conduct in the market more than Bendigo and other non-Major Banks in the future with the Proposed Acquisition, for the following reasons.
603 First, the Tribunal accepts that Bendigo currently provides a competitive offering in the home loans market with strong deposit backing providing a cheap source of funds, a substantial branch network, and a unique community bank model with an emphasis on customer service levels and good customer ratings.662 It has been pursuing IRB accreditation for [REDACTED], and if it is achieved, this will reduce their marginal cost of funds and permit them to price their home loans more sharply.
604 Second, while the Bendigo/Suncorp Bank merger is a realistic commercial possibility, it is far from certain, and there are significant execution challenges and risks. This includes, in particular, likely delays in the timing of the realisation of anticipated synergies, and the impact of those delays on whether the merger would be value accretive to both Bendigo and SGL.
605 It is not commercially realistic to expect that the postulated synergies for the merged entity could be realised within the time periods contemplated in the [REDACTED]. The need to complete the integration of Bendigo’s existing core digital platforms and the additional complications arising from the need to negotiate amendments with the Queensland government would likely require considerable time and resources to be allocated before any substantive cost benefits could be realised by the merged entity.663 Further, the merger could well be characterised as a “merger of equals” and, therefore, more likely than an acquisition of a small bank by a much larger bank to have substantial integration costs, and take a longer time to realise the synergies.
606 These delays in realising the postulated synergies would make it more difficult for the merged entity to invest, in the short to medium term, in the necessary technology to enable it to compete more effectively with the Major Banks and Macquarie, in the home loans market.
607 Third, Bendigo’s small scale has limited its ability to disrupt any coordinated conduct by the Major Banks. Rather, Bendigo has maintained a reasonably steady market share by offering a differentiated product under the pricing umbrella of the Major Banks. Bendigo’s relatively small existing scale has placed them at a disadvantage with respect to the marginal cost of funds from wholesale financial markets. Lack of scale also affects its ability and incentive to invest in tech systems and digital platforms.
608 A doubling of scale might allow a merged Bendigo/Suncorp Bank to have a greater ability and incentive to disrupt coordination by the Major Banks. In particular, a merger with Suncorp Bank could (a) reduce Bendigo’s present marginal cost of funds, allowing the merged entity to sharpen its pricing, and (b) allow the merged entity to spread fixed costs over a larger number of consumers, increasing the incentive to invest in improved digital services, and processing of loans.
609 The evidence before the Tribunal, however, does not support a likely material reduction in the marginal cost of funds for Bendigo in the Bendigo Merger counterfactual. Rather, the following evidence pointed to no likely material improvement in marginal cost, at least in the short term, from a merged Bendigo/Suncorp Bank, (a) uncontradicted expert evidence from Dr Howell that there was little prospect of an increase in credit ratings for the merged entity, other than a single uptick in the Moodys’ rating and a substantial downgrade for Suncorp Bank,664 and (b) the evidence of Mr Ali and Dr van Horen that the prospect of Bendigo achieving IRB accreditation would not be assisted, and if anything, would be made more difficult if Bendigo were to merge with Suncorp Bank.665
610 Fourth, the market share of a merged Bendigo/Suncorp Bank would still be substantially below that of each of the Major Banks. Bendigo’s market share would increase from 2.75% to 5.14% following the merger compared with market shares of between 13.16% and 25.92% for the Major Banks.666 While this represents an almost doubling of market share, the Tribunal does not accept that the Bendigo Merger counterfactual would represent a substantial change in market dynamics, and nor does it accept that the increase in market share would substantially increase the ability of the merged entity to exert competitive constraint on the Major Banks.
611 Fifth, and relatedly, the increase in scale and investment envelope achieved by the merged entity would be unlikely to be sufficient to enable it to fund the necessary technology to enable it to match the technology now deployed and in the process of being developed by the Major Banks and Macquarie. Unlike Macquarie, a merged Bendigo/Suncorp Bank would not have access to equivalent parental financial support and would have to confront and overcome significant technology legacy issues and integration challenges. When combined, the annual investment spends of Suncorp Bank and Bendigo Bank in FY23, outlined at [494], would amount to approximately just less than a tenth of CBA and Westpac’s annual investment spend, and approximately half of Macquarie’s retail investment spend.
612 Even a bank with the scale of ANZ, as explained by Mr Elliott in his s 155 examination, is pursuing the Proposed Acquisition in order for it to be able to obtain sufficient scale to fund its digital transformation program, and address cyber security, fraud and scam management in order to enable it to “stay ahead in competition”, and compete effectively with the larger Major Banks, (CBA and Westpac).667
613 Sixth, the merged entity lacks a sufficiently differentiated product, cost base or supply chain to otherwise enable it to act as a disruptor or “maverick” and, thereby, impose a constraint greater than its market share might otherwise have indicated. Disruption in the home loans market has principally come from Macquarie. Unlike Bendigo and Suncorp Bank, Macquarie has captured a substantially increased market share in the past 10 years. Macquarie’s market share increased from 0.49% in 2012 to 5.01% by February 2023, a net increase of 4.52%.668 In the same period, Bendigo’s market share increased by only 0.5% and Suncorp Bank’s market share fell by 0.39%.669
614 Although the market share of the merged entity would be equivalent to Macquarie’s 5.01% market share as at February 2023,670 Macquarie has a very different model targeting loans with relatively low loan to value ratios and PAYG taxpayers. Unlike Bendigo and Suncorp Bank, Macquarie does not have an extensive branch network, it is heavily dependent on brokers, employs a highly automated model and has significant financial support from a parent company.
615 Eighth, any postulated greater constraint that might be imposed by a merged Bendigo/Suncorp Bank, on the ability of the Major Banks to engage in coordinated conduct in the home loans market, could potentially also be expected to be achieved, at least in part, by a merger between Bendigo and [REDACTED]. The Tribunal accepts, however, that a merger with [REDACTED] would likely be a less attractive proposition for Bendigo given Bendigo’s strong preference, to date, for a merger with Suncorp Bank. Nevertheless, [REDACTED].
616 For all these reasons, the Tribunal is satisfied that a merged Bendigo/Suncorp Bank would not be likely to impose a materially greater constraint on the Major Banks than the two banks operating independently, and by Macquarie and other regional and independent banks. The Tribunal has also concluded, that a merged Bendigo/Suncorp Bank is unlikely to have any meaningful impact on the prospects for successful coordination by the Major Banks because of the uncertainties as to the execution of the merger, the likely significant delays in the realisation of synergies, and the absence of any compelling evidence that the merged entity would benefit from a lower cost of funds. In addition, the Tribunal is also satisfied other recent significant changes to the home loans market around brokers, technical developments and consumer behaviour have reduced the risk of coordination.
G. COMPETITIVE EFFECTS OF THE PROPOSED ACQUISITION IN AGRIBUSINESS BANKING MARKETS
617 The only competitive concerns raised in the agribusiness banking markets in Queensland were in relation to unilateral effects, namely the effect of removing competition between ANZ and Suncorp Bank, being the parties to the Proposed Acquisition, taking account of the likely response by other existing and potential market participants. No coordinated effects concerns were advanced to the Tribunal.
618 In order to assess the competitive effects of the Proposed Acquisition on agribusiness banking markets in Queensland, it is convenient to address first, the likely competitive effects in a future with the Proposed Acquisition, and then to compare those effects with the likely extent of competition in the future without the Proposed Acquisition, having particular regard to the No Sale counterfactual and the Bendigo Merger counterfactual.
619 The first question largely turns on the likely impact on competition of the removal of Suncorp Bank as an independent competitor in agribusiness markets in Queensland. That issue requires a consideration of the extent to which (a) there is competitive overlap between ANZ and Suncorp Bank, (b) Suncorp Bank’s offering is differentiated and competitive, (c) ANZ is likely to maintain Suncorp Bank’s relationship managed model, (d) there is a meaningful constraint imposed by the threat of entry or expansion by other banks, and (e) the broker channel is effectively utilised to source agribusiness customers.
G.2. The principal contentions of the parties
G.2.1. The applicants’ principal contentions
620 The applicants submit that the removal of Suncorp Bank as a competitor in the supply of products and services to agribusiness customers would not have a meaningful competitive effect. They submit that irrespective of whether there is a discrete product market for the supply of banking products and services to agribusiness customers, and regardless of the geographic dimension of the market, Suncorp Bank’s offering to those customers is not materially differentiated.671
G.2.2. Bendigo’s principal contentions
621 Bendigo submits that the Proposed Acquisition will likely lead to a substantial lessening of competition in the Queensland agribusiness banking markets by removing Suncorp Bank as a vigorous and effective competitor for agribusiness customers, when compared with both the No Sale counterfactual and the Bendigo Merger counterfactual.672 It submits that the following propositions relevant to competitive effects in the supply of banking products and services to agribusiness customers in Queensland emerge from the evidence before the Tribunal, (a) agribusiness markets are concentrated, (b) Suncorp Bank offers a differentiated and competitive agribusiness product, (c) ANZ and Suncorp Bank compete meaningfully for agribusiness customers, (d) the constraint imposed by potential entry or expansion is limited, and (e) brokers are not widely used in agribusiness banking and do not impose a meaningful constraint.673
G.2.3. The ACCC’s principal contentions
622 The ACCC submits that after the Tribunal has evaluated the evidence as a whole, including the expert evidence of Professor King and Ms Starks, it could properly conclude that it is not satisfied that the Proposed Acquisition would not, and would not be likely to, substantially lessen competition in the market for agribusiness banking in Queensland.674
623 The ACCC points, in particular, to evidence that, it submits, establishes that supply in the agribusiness banking market in Queensland is concentrated, Suncorp Bank’s offering is differentiated and competitive, there is competitive overlap between ANZ and Suncorp Bank, the constraint imposed by the threat of entry or expansion is limited and brokers are not widely used in agribusiness banking.675
G.3.1. The applicants’ submissions
624 The applicants submit that both their analysis and the ACCC’s analysis suggests that the agribusiness market is not concentrated nationally. They submit that the HHI does not exceed the ACCC’s threshold following the Proposed Acquisition, on their analysis, or only slightly exceeds it, on the ACCC’s analysis, and has a delta less than 100. They submit that in Queensland, in a future with the Proposed Acquisition, it is likely to result in a moderate increase in concentration in Queensland, but the geographic overlap area will remain relatively unconcentrated.676
625 The ACCC submits that there is no definitive source of market share data for agribusiness banking markets in Queensland, but, on any view, ANZ and Suncorp Bank are two significant players in a concentrated market.677 It submits that quantitative share data kept by APRA shows that Suncorp Bank and ANZ are the [REDACTED] and [REDACTED] largest agribusiness lenders in Queensland, and following the Proposed Acquisition, ANZ will become the [REDACTED] largest agribusiness lender, and the [REDACTED] largest Queensland agribusiness lenders will account for almost [REDACTED]% of agribusiness lending.678 It submits that there is, therefore, force in a submission provided by Brennan Mayne Agribusiness (BMAgBiz), a specialised agribusiness consulting firm based in Central Queensland, to the ACCC (BMAgBiz submission), that there “are already only a small number of options when considering moving banks”, and the Proposed Acquisition “would result in a significant reduction in the already limited options”.679
626 The Tribunal considers that (a) irrespective of whether there are national or local/regional markets in Queensland for the supply of agribusiness banking products, the markets are concentrated, and (b) the Proposed Acquisition will significantly increase concentration in those markets, particularly in Queensland.
627 There is no data in evidence from which market shares could reliably be calculated for local and regional agribusiness banking markets in Queensland. The Tribunal, however, was provided with APRA data of agribusiness lending both nationally, and in Queensland. In its Reasons for Determination, the ACCC included the following table based on that data:680
628 Nationally, ANZ and Suncorp Bank have the [REDACTED] and [REDACTED] largest shares of agribusiness lending, respectively, of the top 10 agribusiness lenders that report their agribusiness loans to APRA. In Queensland, the position is [REDACTED], Suncorp Bank has the [REDACTED] largest share [REDACTED] , and ANZ has the [REDACTED]largest share [REDACTED]. A combined ANZ/Suncorp Bank would have the [REDACTED] largest share [REDACTED] of agribusiness lending, nationally, and the [REDACTED]largest share [REDACTED] in Queensland.681
629 In its Reasons for Determination, the ACCC also included the following table containing its calculations from the APRA data of pre-acquisition and post-acquisition HHIs nationally, and in Queensland for agribusiness lending:682
630 It is readily apparent from the HHI calculations that any national or local/regional Queensland markets for agribusiness loans is concentrated. The post-acquisition HHIs, both nationally and in Queensland, exceed the ACCC concentration threshold of 2000. The delta nationally, is only one point below the threshold of 100, and for Queensland, well in excess of 100.
631 In the absence of any statistical data for market shares in the postulated local/regional agribusiness banking markets in Queensland, Ms Starks identified the following 12 towns in Queensland, in which ANZ and Suncorp Bank currently overlap in agribusiness lending: Ayr, Bundaberg, Cairns, Townsville, Mackay, Rockhampton, Emerald, Roma, Dalby, Toowoomba, Goondiwindi and Chinchilla/Miles.683
632 Ms Starks noted that if the Proposed Acquisition proceeds, in Ayr and Chinchilla/Miles, there would only be one competitor bank present, Rabobank in Ayr, and Westpac in Chinchilla/Miles. In Cairns and Chinchilla/Miles, there would be no regional bank, although in Cairns there would be NAB, Westpac, and CBA. In each of the other remaining towns identified by Ms Starks, there would remain at least four banks, including the merged ANZ/Suncorp Bank, if the Proposed Acquisition proceeds.684 Of these towns, Bank of Queensland is present in Toowoomba and Rockhampton.685 Bendigo’s agribusiness offerings are delivered through its dedicated brand, Rural Bank. In response to a request for information from the ACCC, Bendigo stated that “there is no area in Queensland that [Bendigo’s] relationship managers are unable to service”.686
633 The Tribunal notes that except for the towns of Bundaberg and Chinchilla/Miles, Professor King identified direct overlap between ANZ and Suncorp Bank’s agribusiness locations in the same towns as those identified by Ms Starks.687 Professor King did, however, acknowledge that there “may be further overlap” in other agribusiness locations in Queensland.688
634 Further, and relatedly, Mark Bennett has given evidence that, in his view, Suncorp Bank and ANZ are the third and fourth largest agribusiness banks, respectively, in Queensland.689 Mark Bennett states that, in his view, NAB and Rabobank are the first and second largest agribusiness banks in Queensland.690 Mark Bennett also gave evidence that the only regional banks, which have a material presence in Queensland, are Suncorp Bank and Bank of Queensland.691
G.4. Competitive overlap between ANZ and Suncorp Bank
G.4.1. The applicants’ submissions
635 The applicants submit that Suncorp Bank is one of several rivals in agribusiness in Queensland and its removal as a competitor to ANZ will lead neither to a significant increase in concentration in Queensland agribusiness (locally or statewide), nor to the elimination of a particularly vigorous competitor.692
636 The applicants submit that ANZ and Suncorp Bank are not each other’s closest competitors and that their respective agribusiness portfolios are largely complementary by geography and industry, and by customer size.693 The applicants submit that ANZ seeks diversity to manage climate, currency and export risks, which means that its national agribusiness portfolio is not overly concentrated in any location or subsector, and only [REDACTED]% of its portfolio is located in Queensland.694 By way of contrast, they submit that Suncorp Bank is more active in Queensland where it services customers in [REDACTED]. Further, they submit that the evidence does not show that ANZ and Suncorp Bank are or will be particularly close competitors in Queensland in beef, which comprises over half of agribusiness lending in Queensland, or in dairy, which comprises less than 1% of agribusiness in Queensland, and is also declining.695
637 The applicants also submit that ANZ focuses on [REDACTED], whereas the majority [REDACTED] of Suncorp Bank’s current agribusiness portfolio comprises small customers under [REDACTED], and almost [REDACTED]% of its customers are under $[REDACTED] million.696 The applicants submit that Suncorp Bank does not have the same capacity as ANZ to service customers over $[REDACTED] million, because it does not have the capability or risk appetite to offer, at scale, transactional banking, overdraft, equipment financing and foreign exchange products and services to these customers.697 They also submit that Suncorp Bank is often not agribusiness customers’ MFI and it typically provides lending as a secondary bank with other banks providing transactional and other services.
638 Further, the applicants submit that if the Proposed Acquisition proceeds, ANZ would face effective competition in agribusiness lending nationally and in Queensland, particularly from both NAB and Rabobank, which would [REDACTED]and impose a greater competitive constraint than Suncorp Bank. They note that NAB [REDACTED] its agribusiness market share from [REDACTED]to [REDACTED]and to [REDACTED].698 They also note that [REDACTED], for which it is likely to compete closely with ANZ.699
639 Bendigo submits that ANZ and Suncorp Bank are competitors across a broad range of agribusiness customers and are particularly close competitors for small to medium agribusiness customers in certain areas of regional Queensland. It submits that there is evidence of ANZ and Suncorp Bank winning agribusiness customers from each other,700 and third parties have observed that ANZ and Suncorp Bank are strong competitors.701
640 Bendigo accepts that ANZ has a focus on larger agribusiness customers, and Suncorp Bank’s traditional customer base tends towards small to medium agribusinesses but submits that there remains significant overlap and competitive tension.702
641 Bendigo also submits that there is significant geographic overlap in ANZ and Suncorp Bank’s agribusiness banking operations in Queensland, and that they appear to overlap in the following 12 towns identified by Ms Starks: Ayr, Bundaberg, Cairns, Townsville, Mackay, Rockhampton, Emerald, Roma, Dalby, Toowoomba, Goondiwindi and Chinchilla/Miles.703
642 The ACCC submits that ANZ and Suncorp Bank presently compete for at least some of the same customer segments of the agribusiness banking market in Queensland.704 It relies on the evidence of Mark Bennett to the effect that Suncorp Bank and ANZ are effective in winning business from each other in Queensland and he generally considers that ANZ and Suncorp Bank “seek similar customers” in customer segments with total business limits in the range from $[REDACTED] to $[REDACTED] million.705 It also refers to Ms Starks’ evidence about the towns in which ANZ and Suncorp Bank’s agribusiness bankers’ operations overlap in Queensland.706
643 The Tribunal considers that there is some competitive overlap between ANZ and Suncorp Bank’s agribusiness offerings in Queensland, not least because of the 12 Queensland towns identified by Ms Starks, in which ANZ and Suncorp Bank overlap.
644 As Bendigo submits, the BMAgBiz submission observed, from a third party perspective, that ANZ and Suncorp Bank were “strong competitors”.707 The Tribunal notes that the BMAgBiz submission was advanced in the context of the applicants’ authorisation application before the ACCC. Nevertheless, BMAgBiz, which has been servicing rural clients in Queensland, Western Australia, New South Wales and the Northern Territory, for 30 years, submitted that Suncorp Bank was “particularly strong in agribusiness lending in Queensland”, and “[i]t is highly competitive with small to mid-sized borrowers i.e. less than $50M”.708 It also submitted that the 1.4% increase in ANZ’s share of the national lending market, to 15.4%, is not indicative of the impact that the Proposed Acquisition would have on the agribusiness “lending market in which Suncorp is a major player (particularly in Queensland) & in which ANZ & Suncorp are direct competitors”.709
645 Mark Bennett gave evidence that ANZ and Suncorp Bank are “winning agribusiness customers from each other”.710 Despite being qualitative and somewhat equivocal, both Bendigo and the ACCC rely on Mark Bennett’s evidence that:711
Within Queensland, my perception is that Suncorp Bank does a reasonably good job of servicing its clients, and has loyalty from them, but that ANZ is able to win business from Suncorp Bank customers.
646 Mark Bennett also gave evidence that “Suncorp Bank can also be effective in winning business from ANZ”,712 and:
Generally, I consider that ANZ and Suncorp Bank seek similar customers in the BB and the lower end of the SD segments, and service those customers with a similar business model.713
647 Whilst the Tribunal accepts that there is some competitive overlap between ANZ and Suncorp Bank, it considers that ANZ competes more closely with NAB, Rabobank and Westpac. ANZ focuses on customers with complex needs, high turnover and debt ranging between $[REDACTED] million and $[REDACTED] million.714 By contrast, Suncorp Bank generally focuses on less complex customer segments, including family intergenerational farmers, with most agribusiness customers having banked with Suncorp Bank for “at least for 10 years”,715 with borrowing needs ranging between $[REDACTED] million and $[REDACTED] million.716
648 Further, Rabobank is present in all Queensland towns in which ANZ is present, and of the 12 towns identified by Ms Starks, where ANZ and Suncorp Bank overlap, CBA and Rabobank are present in 10, Westpac is present in eight, and NAB is present in seven of those towns. In Ayr, only one competitor, Rabobank, would remain, but it is relevantly within reasonable driving distance to Townsville, where NAB, Rabobank, Westpac and CBA are present.717
649 The Tribunal notes that in Cairns and Chinchilla/Miles, there would be no sizeable regional bank present. Both Bendigo and Bank of Queensland are also present in Cairns, but their shares of agribusiness lending in Queensland are currently small.718 Bendigo’s share of agribusiness lending in Queensland in FY22 was only [REDACTED]%, and Bank of Queensland’s share for the same period was [REDACTED]%.719
650 Rabobank is, however, located in Atherton, a reasonable drive from Cairns, and is also located in multiple towns within reasonable driving distance from Chinchilla/Miles.720 Further, there is evidence from Bendigo, in response to an ACCC request for information that there is “no area in Queensland that [Bendigo’s] relationship managers are unable to service” their agribusiness customers.721
G.5. Is the Suncorp Bank offering differentiated and competitive?
G.5.1. The applicants’ submissions
651 The applicants submit that Suncorp Bank’s offering to agribusiness customers is neither materially differentiated from other bank’s offerings nor particularly competitive, for the following reasons.
652 First, Suncorp Bank does not compete materially differently to other banks in the market (including ANZ, NAB and Rabobank) on price or non-price factors and there is no evidence it is more flexible in service or pricing than ANZ. The applicants submit that the evidence does not support the contention that Suncorp Bank is materially more competitive in servicing “non-standard” agribusiness banking needs. They submit that Suncorp Bank tends to have a higher volume of business in Queensland because of its historical footprint rather than any material difference in the quality or price of its offering to customers.722
653 Second, Suncorp Bank’s relationship-managed model is not unique, or more personalised and attentive than other banks. The applicants submit that it can readily be replicated and is relevantly, [REDACTED] than ANZ’s relationship model.723 They submit that the sentiment of “care” is commonplace among agribusiness executives, and it cannot necessarily be inferred that a bank’s customer service is materially better based only on the number, or ratio, of relationship managers to customers.724
654 Third, in any event, the relationship model that ANZ uses for medium-sized agribusiness customers is comparable with Suncorp Bank’s offering.725
655 Fourth, while the provision of relationship bankers may be appreciated by agribusiness customers, their effectiveness has more to do with the individual banker than their employer bank. The applicants submit that competitiveness in this respect is a function of the personal reputation of the banker, which in part depends on word-of-mouth recommendations.726
656 Further, the applicants submit, in response to the ACCC’s submissions, that to merely identify some points of differentiation in Suncorp Bank’s current offering, is nothing more than a static analysis. They submit that to show that competition would be materially reduced, either some feature in the acquired entity that cannot be replicated, or some other reason why competition would be suppressed, must be identified, for example, that the remaining entities are likely to be less competitive on pricing or service levels for an identified reason.727
657 Bendigo submits that Suncorp Bank is a strong competitor in agribusiness markets in Queensland. It submits that (a) Suncorp Bank has a differentiated offering to ANZ, including on price and non-price factors such as customer service and its ability and willingness to serve non-standard agribusiness needs,728 (b) the larger banks such as ANZ, are more focused on high-volume traditional agribusiness lending,729 and (c) in contrast, Suncorp Bank may be willing to understand complex or unique client situations and fund those clients through a more flexible and dynamic approach to lending.730
658 The ACCC submits that consistently with its strong share in the Queensland agribusiness banking market, Suncorp Bank is presently an effective competitor with a differentiated offering that is attractive to many agribusiness customers.731
659 The ACCC submits that Suncorp Bank differentiates itself from the Major Banks on the basis of its customer service and care. The ACCC submits that the Major Banks have a lower percentage of agribusiness customers that are relationship-managed compared with Suncorp Bank.732 It submits that about [REDACTED]% of Suncorp Bank’s agribusiness customers are relationship-managed, and that its organic plan involves maintaining this “relationship management approach”.733 In contrast, it submits that the Major Banks utilise relationship management for the following percentages of agribusiness customers: ANZ [REDACTED],734 CBA [REDACTED],735 Westpac [REDACTED],736 and NAB [REDACTED].737
660 The Tribunal is satisfied that Suncorp Bank is a relatively strong competitor in agribusiness markets in Queensland. Relevantly, however, the Tribunal is not satisfied that Suncorp Bank’s agribusiness offering is materially differentiated from Rabobank, and regional banks such as Bank of Queensland, and Bendigo, or that it has an agribusiness offering that is unique, and unable to be replicated, for the following reasons.
661 First, it is apparent that Suncorp Bank has a relatively stronger presence in agribusiness markets in Queensland, holding the [REDACTED]largest share [REDACTED] in agribusiness lending in Queensland, whereas it holds the [REDACTED] largest share [REDACTED] nationally.738
662 Second, a strategy update in a memorandum prepared for the [REDACTED], highlights that Suncorp Bank benefits from a stronger presence in Northern Queensland.739 The [REDACTED] memorandum is particularly instructive not least because it was a contemporaneous business record, instead of a submission made to the ACCC, or evidence adduced from an interested party in an ACCC examination.740 It is important to acknowledge, however, that the [REDACTED] memorandum provided a competitive review of the “Australian rural lending market”, and was, therefore, not limited to competitive conditions in Queensland local/regional markets for agribusiness lending.
663 The following conclusions and findings in the memorandum are of particular significance:
(a) [REDACTED]. [REDACTED]. [REDACTED].741
(b) [REDACTED]:742
• [REDACTED]
• [REDACTED].
• [REDACTED].
(c) [REDACTED],743 [REDACTED]744 and [REDACTED].745 [REDACTED]:746
[REDACTED].
664 The references to [REDACTED].
665 Third, the Tribunal accepts that there are some material differences between Suncorp Bank’s relationship model and the model employed by ANZ.
666 ANZ uses three relationship models for its agribusiness customers,747 by classifying its agribusiness customers as Small Business Banking (SBB), Business Banking (BB), or Specialist Distribution (SD) customers.
667 ANZ uses a locally managed model for agribusiness customers with banking needs that are classified as complex (approximately [REDACTED]% by number and [REDACTED]% of funds under management). These customers are managed by a relationship manager (for SD and BB customers), and a business banking manager (for SBB customers) that generally travel to the customer.748
668 ANZ uses a remote managed model for customers with complex/intermediate banking needs (approximately [REDACTED]% by number and [REDACTED]% of funds under management). These customers are managed by small business managers in the specialised agribusiness team at the ANZ National Business Centre in Melbourne.749
669 ANZ uses a portfolio management model for customers with simple banking needs (approximately [REDACTED]% by number and [REDACTED]% by funds under management). These customers are managed through self-service options, and a team of SBB specialists by phone, email, in-person meetings at branches, or online.750
670 Suncorp Bank operates a “relationship-focused” model for customers who have borrowings from Suncorp Bank. Approximately [REDACTED]% of its agribusiness customers in Queensland (who have borrowings with Suncorp Bank) are relationship managed.751
671 Suncorp Bank’s relationship managers usually travel to the premises of a customer, or a third-party location near the customer’s premises, such as a local coffee shop. Branches are used where it is a customer’s preference to meet at a branch.752 The distance relationship managers are required to travel depends on the region. Relationship managers based in Toowoomba may have customers based locally, and therefore are only required to travel for two hours to meet customers. In contrast, relationship managers based in [REDACTED] may regularly be required to travel [REDACTED] hours, and in some cases relationship managers could be required to travel [REDACTED] hours to visit a customer.753
672 The Tribunal accepts that ANZ’s relationship model for agribusiness customers with simple banking needs is inferior to Suncorp Bank’s model. Although, Suncorp Bank recently attempted to [REDACTED] their relationship-focused customer servicing model for agribusiness customers to [REDACTED] customers that provided more than $[REDACTED] in business. Suncorp Bank, however, [REDACTED].754
673 ANZ’s relationship model for agribusiness customers, classified as having complex or complex/intermediate banking needs ([REDACTED]% by number, and [REDACTED]% by funds under management), however, is not materially different from that offered by Suncorp Bank to its agribusiness customers.
674 Fourth, the ratio of bankers to customers in Suncorp Bank’s relationship model is inferior to that used by [REDACTED], and similar to that employed by [REDACTED].
675 The relationship model used by Suncorp Bank for agribusiness customers provided for a banker-customer ratio of [REDACTED], and for “higher tiered portfolios”, and an additional assistant relationship manager to support the relationship manager at a [REDACTED] ratio.755 This ratio is higher than that utilised by [REDACTED](nationally) and [REDACTED](in Queensland),756 and [REDACTED] (nationally) and [REDACTED](in Queensland),757 and similar to that employed by [REDACTED](nationally) and [REDACTED](in Queensland).758
676 The importance of relationship management for agribusiness clients was highlighted in an August 2020 memorandum to [REDACTED].759 [REDACTED].760
677 Fifth, responses provided by NAB,761 [REDACTED],762 and [REDACTED],763 to requests for information from the ACCC, highlight that irrespective of any differences in relationship models, competitor banks generally strive to develop models that focus on the “customer experience”, and the provision of high quality relationship management services.
678 For example, in response to a s 155 notice issued by the ACCC, NAB outlined the roles and responsibilities of its agribusiness managers, and noted that they are “expected to consider customer needs, both in the immediate future and with the objective of building long term customer relationships”.764 NAB also described a typical day for an agribusiness manager, which again, highlights the importance of customer engagement for agribusiness customers:765
Assessing suitability, risk and economic value of customer proposals including customer pricing decisions or recommendations;
Sharing insights with customers about business trends including contributing to customers’ business strategies;
Proactively leveraging industry and commercial contacts to build industry networks and create new business opportunities;
Preparing and assessing customer proposals including working with internal stakeholders to assess and fulfil customer requests; and
Coaching and supporting junior team members.
679 Sixth, the Tribunal does not accept that Suncorp Bank’s stated focus on personal relationships and customer care provides a meaningful distinction to, at least Rabobank’s, personalised offering to agribusiness customers.
680 Suncorp Bank’s “care value proposition” was explained in speaking notes prepared for a business banking board strategy session in April 2022 in the following terms:766
Servicing customer’s locally
As a proof point of our Care value proposition, a core tenet of the strategy is to service customers locally. While this does not mean a full decentralisation of the Business Banking model, but rather a regional business model that is structurally aligned to Agribusiness. This will enable us to better understand our customers’ needs, attract new and different talent, and better support local communities through employment and banking local businesses.
681 Mr Cleland provided the following explanation in his s 155 examination of the “care” component of Suncorp Bank’s strategy for agricultural banking customers:767
MR CLELAND: Empathy. Listening. Understanding. Being there in the moments that matter. So in agri this is particularly important. So when there’s floods, when there’s droughts we don't walk away. We sit around the kitchen table and help the customer navigate through that. Really important. Um competitors, are there in the good times. It gets tough, walk away.
MS CARISTO: Which competitor - did you have any particular competitors - - -
MR CLELAND: I think the [REDACTED] - - -
MS CARISTO: Okay.
MR CLELAND: - - - are known for being more commercial and less caring. I would say we’re more caring.
682 The Tribunal does not accept that this sentiment of “care”, as described in internal SGL documents, and by Mr Cleland, is unique to Suncorp Bank.
G.6. Maintaining Suncorp Bank’s relationship managed model
G.6.1. The applicants’ submissions
683 The applicants submit that there is no reason to expect that ANZ would not maintain Suncorp Bank’s relationship model and focus on service quality for agribusiness customers, if the Proposed Acquisition proceeds, for the following reasons.
684 First, ANZ would have little incentive not to maintain Suncorp Bank’s specific relationship management model and its focus on service quality given there is a material degree of churn in the market that means banks need to continue to attract customers to maintain market share. By way of example, they note that ANZ has about [REDACTED]% attrition each year due to paydown, amortisation or refinancing for agribusiness customers away from ANZ.768
685 Second, it is unlikely that the particular service offering presently provided by Suncorp Bank’s agribusiness bankers would disappear under ANZ because ANZ presently offers a responsive relationship-based model to customers with lending of $[REDACTED] or more, or complex needs.769
686 Third, and relatedly, ANZ views the Proposed Acquisition as an opportunity to acquire Suncorp Bank’s relationship bankers and it has every incentive to work hard to retain them post-acquisition. The applicants submit that if Suncorp Bank’s relationship bankers are a competitive strength, ANZ would have good reason to keep them functioning in the same way. If they are not, then they do not matter to the competition analysis.770
687 Bendigo submits that the removal of Suncorp Bank is likely to result in a substantial lessening of competition given the specific and diverse needs of agribusiness customers. It submits that Suncorp Bank has a specific relationship model and specific focus on service quality, which ANZ has little incentive to maintain. Rather, Bendigo submits, ANZ has strong incentives to align Suncorp Bank’s relationship model for agribusiness customers with its own.771 It submits that ANZ has strong incentives to do so, including because it considers that customers “prefer conducting their banking through digital means” and that technological innovations can “reduce (and often eliminate) the need to meet a relationship manager face-to-face to obtain a product.”.772
688 The Tribunal is not persuaded that it is likely that ANZ would cease to offer a relationship model comparable to Suncorp Bank, in a future with the Proposed Acquisition. In a future with the Proposed Acquisition, ANZ will likely have every incentive to retain Suncorp Bank’s relationship bankers, if they are viewed as a competitive strength.
689 In her supplementary report, Ms Starks accepted that there was relatively little direct evidence of what ANZ would do with Suncorp Bank in the medium term. She accepted that ANZ’s statement that “Following the Proposed Acquisition, Suncorp Bank customers will continue to have access to a relationship-led service model, where applicable” was plausible given they “currently use relationship managers in a very similar way in the SME market”. Ms Starks then noted, however, by reference to evidence given by Mr Elliott, that ANZ plans to “automate processes and offer digital service propositions”, which she stated suggests “a move away from personalised banking for customers who do not qualify for relationship-managed banking”.773
690 Mr Elliott’s evidence was as follows:
The technology and platform we have built, and will continue to improve upon, will enhance ANZ’s ability to provide high quality, compelling experiences to our customers at pace. This includes making it easy and convenient to open accounts, reducing the time to get a home or business loan, providing customers with data-driven insights to improve their financial wellbeing and empowering customers to self-serve digitally at a time that is convenient for them. In addition, our new technology makes it easier (and lower risk) to launch and modify our products and services and helps reduce the risk of remediation and ensure compliance with our obligations. With its already significant investment, and progress, in digital transformation, ANZ is well placed to deliver better outcomes for Suncorp Bank's customers over time compared to outcomes for those customers if Suncorp Bank remains owned by the Suncorp Group. As I noted above at paragraph 63(b), our ability to do this will be enhanced due to the scale that acquiring Suncorp Bank brings.774
691 Mr Elliott’s evidence suggests that ANZ’s plans are, at least in part, directed at introducing efficiencies in back office functions rather than introducing new “digital services”.775 As submitted by the applicants, any investment in digital automation of processes should reduce administrative burdens, and lead to a greater maximisation of the time, quality and number of relationships with customers.776At the same time, however, the reference to “empowering customers to self-serve digitally at a time that is convenient for them” does suggest, as Ms Starks observed, a move away from personalised banking for those customers who do not qualify for relationship banking.777
692 Perhaps the most useful insight into ANZ’s plans for Suncorp Bank’s agribusiness model post-acquisition emerges from the “Project Parker: Business Bank Target Operating Model Playbook” as at 26 August 2022 developed by ANZ, in collaboration with Deloitte and Suncorp Bank (ANZ Playbook). It is a document of some 400 pages and articulates the conceptual design and implementation roadmap for ANZ’s initial [REDACTED] month interim operating model for the Suncorp Bank’s business bank, including agribusiness banking services.778
693 The ANZ Playbook included explanations of service models and coverage ratios to be used, including for a portfolio described as “Agribusiness & Regional”. The service model for that portfolio was for a dedicated regional manager/assistant regional manager to service a portfolio of agribusiness/regional customers for loans in excess of $[REDACTED] million at an approximate banker-to-customer ratio of [REDACTED]. For customers with lending values less than $[REDACTED] million, regional managers were to be permitted to provide relationship management services to them if the regional managers believed that there was a future growth opportunity for these customers. It was also proposed that the [REDACTED] ratio would reduce to [REDACTED] for higher tiered customers and regional managers servicing these customers would be assisted by a second assistant regional manager.779
694 The ANZ Playbook also included a chart forecasting an average four year cumulative annual growth rate of [REDACTED]% for Suncorp Bank’s business bank post-acquisition and annual growth rates for the Agribusiness & Regional portfolio that would give rise to a [REDACTED] year cumulative annual growth rate of [REDACTED]%.780 These estimates evidence [REDACTED].781
695 Further, the Tribunal is satisfied that Rabobank and regional competitors, which target similar customer segments to Suncorp Bank, would also have the incentive and capability to replicate Suncorp Bank’s model. In this regard, even if ANZ aligned Suncorp Bank’s relationship model more closely with its own, the Tribunal considers that Rabobank and regional competitors, such as Bendigo (through Rural Bank) or Bank of Queensland, are either offering a similar level of service and/or could expand their existing offer in competition with the merged firm.
G.7. Barriers to entry or expansion
G.7.1. The applicants’ submissions
696 The applicants submit that the principal barrier to entry and expansion for existing banks is obtaining specialised agribusiness bankers.782 They submit that this not a sufficient barrier to limit the threat of expansion or entry. They point to Rabobank’s recent increase in its market share for agribusiness banking from [REDACTED], in the [REDACTED] years leading up to 2022 to become the [REDACTED] largest supplier in Queensland, and Judo Bank’s recent entry and success in attracting agribusiness bankers from ANZ in Queensland. They submit that barriers to expansion are low for existing banks such as Bank of Queensland, Judo Bank and Bendigo, not least given [REDACTED].783
697 The ACCC submits, by reference to Professor King’s evidence, that although barriers to expansion in agribusiness are moderate, barriers to entry are high, and the constraint imposed by new entrants is limited. It submits that Professor King’s evidence is consistent with submissions by Bendigo and BMAgBiz,784 which highlighted that barriers to becoming a full-service agribusiness operator were high. It submits that this contention is also supported by Mark Bennett’s evidence which emphasised that (a) the significant investment in human resources required by a bank, and the often “multi-year effort” to establish relevant relationships, to enter and compete in a regional agribusiness banking market successfully, and (b) the considerable competition for agribusiness bankers.785
698 Moreover, the ACCC submits that the success of Rabobank and Judo Bank do not support the conclusion that barriers to entry and expansion are low.786 It submits that, in the case of Rabobank, it expanded gradually, over a number of decades.787 It submits that to the extent there is private survey firm data suggesting that Rabobank has expanded quickly in the last three years, that data should be treated with caution, not least because it estimates a merged ANZ/Suncorp Bank would be the largest agribusiness lender in Queensland, with a [REDACTED]% market share, with its next largest competitor being [REDACTED] and [REDACTED]. It submits that if such data was correct, the three largest banks following the Proposed Acquisition would hold a combined share of agribusiness lending in Queensland of [REDACTED]%, which would suggest that the adverse effects on competition would be very significant.788 It further submits that Judo Bank’s recent entry does not show that barriers to entry and expansion are low because it remains a small competitor and has also acknowledged it faced significant set-up and operational costs in establishing a new branch network.789
699 The Tribunal accepts that sourcing specialist agribusiness bankers is a relatively low barrier to entry or expansion, and that some banks, including ANZ790 and Judo Bank,791 provide banking services to agribusiness customers through generalist bankers who were trained and supported by agribusiness bankers, or through specialist agribusiness bankers.
700 Specialist agribusiness bankers are not a “scarce resource” in the same sense as gold or iron ore deposits. It is a resource that can be obtained in the labour market and more agribusiness bankers can be trained and acquire the relevant experience over time. Further, to the extent that ANZ reduced the level of service to agribusiness customers in the wake of the Proposed Acquisition, the pool of available specialised agribusiness bankers would be increased.
701 There is also evidence from banks, including NAB, that their agribusiness bankers are, in any case, not required to have any “specialist knowledge or expertise”. NAB stated in response to an ACCC information request that it did not require its agribusiness managers to have any specialist knowledge or expertise but did recommend that they have tertiary qualifications (such as a degree in Business, Commerce or Agribusiness), commercial banking experience, and a desire to develop a deep understanding of the agribusiness industry and geographic specific knowledge to complement core technical commercial banking knowledge and experience.792
G.8.1. The applicants’ submissions
702 The applicants submit that brokers have begun to supplant the role of relationship managers in understanding agribusiness borrowers’ bespoke needs and sourcing appropriate products to meet those needs.793
703 The applicants submit that brokers, including specialised agribusiness brokers, create competitive tension and drive material amounts of lending (including refinancing) in agribusiness. They submit that brokers are likely to make customer relationships less significant as a point of differentiation because customers have a direct relationship with their broker, particularly as experienced agribusiness bankers switch from banking to brokering.794 In this regard, they submit that Suncorp Bank, in particular, has recently lost a number of relationship managers to become brokers, taking customers with them.795
704 The ACCC submits that brokers are not widely used in agribusiness and are, therefore, likely to have a limited effect on competition.796 They submit that the applicants’ summary of the proportion of broker originated agribusiness loans in Queensland is incomplete. The ACCC notes that in 2022 [REDACTED]797 and Bendigo798 [REDACTED], and that [REDACTED]799 and [REDACTED]800 had less than [REDACTED]%. The ACCC submits that there is therefore reason to question the effectiveness of brokers in ensuring or enhancing competition in agribusiness banking.
705 The extent to which banks currently use brokers for originating agribusiness loans in Queensland, varies significantly.
706 In 2022, Bendigo had no broker originated agribusiness lending in Queensland, although it noted that it had only recently launched its broker model. Bendigo otherwise claimed that Queensland was one of the lowest states for broker penetration, with less than 10% of new agribusiness loans originated through third parties.801
707 [REDACTED].802 [REDACTED],803 [REDACTED].804 ANZ’s estimate of broker originated agribusiness loans was slightly higher, reporting that in 2022, approximately [REDACTED]% (by value) of agribusiness loans were originated through brokers, with a higher proportion for SBB and BB customers, and a lower proportion for SD customers.805 The percentage of broker originated loans for ANZ can be further broken down as follows, [REDACTED]% for loans for less than $[REDACTED] million, [REDACTED]% for loans for between $[REDACTED] million and $[REDACTED] million, and [REDACTED]% for loans in excess of $[REDACTED] million.806
708 In contrast, approximately [REDACTED]% of Suncorp Bank’s new agribusiness loans in Queensland (by value) are broker originated.807 The equivalent percentages for [REDACTED] are [REDACTED]%,808 and for [REDACTED], [REDACTED]%.809
709 The Mortgage and Finance Association of Australia observed in its submission to the ACCC that there is insufficient data on brokers in agribusiness but in answer to the following question from the ACCC:
whether brokers generally compare loan offers from lenders on their panels only or give preference to those lenders’ offers (rather than seeking other options)810
710 The Mortgage and Finance Association of Australia relevantly responded:811
In terms of agri loans specifically, we understand that often this is very individualised and the types of business loans sourced for agri customers could range from crop funding, equipment finance, farm machinery etc. As an extension to this comment, agri brokers are highly specialised and the way in which agri-brokers work on behalf of their customers is generally through working closely with a business/agri manager within a lender. We are not aware of the existence of panels with respect to agri-lending.
711 The Tribunal accepts that the use of brokers may well reduce the significance of the customer relationship model as a point of distinction between banks engaged in agribusiness lending. The Tribunal, however, is not persuaded that the current relatively varied and limited utilisation of the broker channel in agribusiness lending in Queensland is sufficient to have any material impact on the significance of the relationship model.
G.9. Constraint imposed by Suncorp Bank in the No Sale counterfactual
G.9.1 The applicants’ submissions
712 The applicants submit that the application of the correct forward-looking approach demonstrates that Suncorp Bank would be unlikely to impose any material constraint on other market participants if the Proposed Acquisition did not proceed.812
713 The applicants submit that Suncorp Bank and ANZ are not, and would not be, particularly close competitors in the No Sale counterfactual.813
714 Moreover, even if the Tribunal concluded that Suncorp Bank’s relationship model was unique, the applicants submit that the viability of the relationship managed model faces the following limitations: (a) consolidation and increasing financial complexity limit the viability of relationship managed models without investment in digital platforms to generate efficiencies, (b) competition from brokers is replacing any constraint that Suncorp Bank’s relationship model has or would impose, in the No Sale counterfactual,814 and (c) there is no probative evidence that Suncorp Bank’s quality and service in the No Sale Counterfactual would not remain under ANZ ownership, in a future with the Proposed Acquisition.815
715 Bendigo submits that there is some evidence that in the future without the Proposed Acquisition, Suncorp Bank would focus on growing its agribusiness portfolio and competing more aggressively. It submits, by contrast, the Proposed Acquisition will reduce the number of competing firms in every local market where ANZ and Suncorp both operate, and bring together two effective competitors so their direct competition ceases. It submits that this will lead to a substantial reduction in consumer choice and competition in non-price aspects, particularly for agribusiness customers with bespoke needs. Bendigo submits that both Professor King and Ms Starks agree that on this basis the Proposed Acquisition is likely to lead to a substantial lessening of competition in the agribusiness markets compared with the No Sale Counterfactual.816
716 The ACCC submits that there are various possibilities as to what would happen to the agribusiness portfolio of Suncorp Bank in the No Sale counterfactual. The ACCC submits that two of those possibilities are for Suncorp Bank to continue to compete in agribusiness lending as it currently does, or a divestment of its agribusiness lending to a second-tier bank, as suggested by Judo Bank in its submission to the ACCC, in order to focus on loans to other segments. The ACCC accepts that the implications for competition would necessarily depend on the identity of the purchaser but submits that for present purposes, it cannot be assumed that competition would necessarily be lessened by a decision by Suncorp Bank to exit the market for agribusiness lending.817
717 The Tribunal is not satisfied, for the following reasons, that there is any persuasive evidence from which it could conclude that Suncorp Bank’s present offering to agribusiness customers in Queensland is likely to be more competitive in the No Sale counterfactual.
718 First, as discussed at [672] above, the evidence suggests that in the No Sale counterfactual, Suncorp Bank would either continue to provide the same level of service to its agribusiness customers, or would provide a reduced level of service, as it has already unsuccessfully sought to provide by reducing the number of customers serviced by relationship managers.
719 Second, there is very limited evidence before the Tribunal that Suncorp Bank intends to grow its agribusiness portfolio and compete more aggressively in local/regional markets in Queensland. Bendigo advances the submission that Suncorp Bank intends to compete more aggressively by reference to Dr van Horen’s first witness statement,818 in which he gave evidence that, consistently with the Suncorp Bank FY23-25 Plan, SGL intends to grow Suncorp Bank’s agribusiness portfolio by building on its position in beef in Queensland to win customers in New South Wales, Victoria and South Australia in cotton, broadacre and dairy agribusinesses.819 Dr van Horen also gave evidence that Suncorp Bank is focussed on growing its business in family-owned enterprises.820 Dr van Horen’s evidence does not extend, however, to any discussion of any plans by SGL to grow Suncorp Bank’s agribusiness portfolio in local/regional markets in Queensland.
720 Further, the Suncorp Bank FY23-25 Plan states under the heading “Growth trajectory” that:
As we begin to roll out our refreshed business banking strategy, we expect the portfolio to [REDACTED]. Growth will initially be driven through our commercial and agribusiness portfolios, and [complement] stronger SME growth in outer years.821
721 Relevantly, the Suncorp Bank FY23-25 Plan does not discuss growth of Suncorp Bank’s agribusiness portfolio in Queensland. There is, therefore, insufficient evidence before the Tribunal to conclude that Suncorp Bank’s agribusiness portfolio in local/regional markets in Queensland is likely to become more competitive in the No Sale counterfactual.
722 Third, the Tribunal accepts that Suncorp Bank’s offering to agribusiness customers is likely to be less competitive if the trend towards consolidation continues and, therefore, the needs of agribusiness customers increase in complexity.822 As the applicants submit, in the absence of investment in digital capabilities to reduce the need for manual work and intervention, an investment that SGL cannot appear to make at acceptable ROE, Suncorp Bank’s service offering to agribusiness customers with increasingly complex needs, will be less effective.823
723 Further, and relatedly, to the extent that Suncorp Bank’s offering for small agribusiness customers was differentiated from its competitors, in a future without the Proposed Acquisition in the No Sale counterfactual, Suncorp Bank has decided to [REDACTED].824 This decision was made after Suncorp Bank attempted to cease offering relationship managers to agribusiness customers with less than $1 million in borrowings but then reversed this decision following resistance from relevant agribusiness customers.825
724 Fourth, consistently with the Tribunal’s conclusion at [149] to [159], in respect of the alternative portfolio divestment strategies postulated by the ACCC, the Tribunal is not persuaded that there is a sufficient basis to conclude that SGL would improve its competitive position in agribusiness banking markets in Queensland by divesting Suncorp Bank’s agribusiness lending to a second-tier bank. There is little evidentiary support for this submission.
G.10 Constraint imposed by merged Bendigo/Suncorp Bank in the Bendigo Merger counterfactual
G.10.1. The applicants’ submissions
725 The applicants submit that a merged Bendigo/Suncorp Bank is not likely to be a more vigorous or effective competitor than Bendigo or Suncorp Bank alone. They submit that Bendigo is not a significant competitor in the locations in which ANZ and Suncorp Bank overlap and, therefore, in those locations, an acquisition by Bendigo would simply mean that Suncorp Bank has a different owner.826 Moreover, they submit that if the Proposed Acquisition proceeds, there is a possibility of Bendigo merging with Bank of Queensland, and any competitive constraint posed by a merged Bendigo/Suncorp Bank could be replicated.827
726 Bendigo submits that in a future without the Proposed Acquisition, a merged Bendigo/Suncorp Bank would be likely to impose a substantially enhanced competitive constraint in agribusiness banking markets in Queensland compared with the Proposed Acquisition.828
727 Bendigo submits that the activities of Suncorp Bank and Bendigo in both the Queensland agribusiness banking markets and nationally, are complementary rather than competitive. It submits that in Queensland, Bendigo’s share of farm lending is just [REDACTED]%, whereas Professor King estimates that Suncorp Bank’s share sits at approximately [REDACTED]%.829 It submits that the Bendigo Merger counterfactual would, therefore, only lead to [REDACTED] in the concentration of supply of agribusiness banking services in Queensland. It submits that by contrast, Bendigo has a [REDACTED] market share in farm lending in South Australia ([REDACTED]%), Western Australia ([REDACTED]%), Victoria ([REDACTED]%), and Tasmania ([REDACTED]%), while Suncorp Bank’s shares in these states are trivial.830 It further submits that Bendigo and Suncorp Bank have complementary business models, each having a focus on relationships and bespoke bank lending.831
G.10.3. The ACCC’s submissions
728 The ACCC submits, by reference to Ms Starks, that a merged Bendigo/Suncorp Bank would be at least as strong a constraint on ANZ as Suncorp Bank is currently in the agribusiness market in the No Sale Counterfactual.832 It submits, by reference to Professor King’s evidence, that competition would also potentially increase in a future without the Proposed Acquisition, in the Bendigo Merger counterfactual, because of the complementarity of Bendigo and Suncorp Bank’s agribusiness banking portfolios.833 It submits, therefore, that the applicants’ submission that a merged Bendigo/Suncorp Bank would not impose a greater constraint than Bendigo or Suncorp Bank alone, is contrary to the weight of the expert evidence.834
729 The Tribunal is not satisfied, for the following reasons, that there is any persuasive evidence, from which it could conclude that Suncorp Bank’s present offering to agribusiness customers in Queensland is likely to be materially more competitive in the Bendigo Merger counterfactual.
730 First, Ms Starks concluded that a merged Bendigo/Suncorp Bank would be at least as strong a constraint on ANZ as Suncorp Bank is in the No Sale counterfactual:
[REDACTED]. As such, the [Bendigo]/Suncorp entity will be at least as strong a constraint on ANZ as Suncorp is in the status quo counterfactual. Therefore, relative to this counterfactual, I consider there is a real chance that the acquisition will give rise to an SLC.835
731 The Tribunal accepts that a merged Bendigo/Suncorp Bank would impose as strong a constraint on ANZ as Suncorp Bank in the No Sale counterfactual. As explained at [717] to [724] above, however, the Tribunal has concluded that there is not any persuasive evidence from which it could conclude that Suncorp Bank’s offering to agribusiness customers in Queensland is likely to be more competitive in the No Sale counterfactual. The evidence suggests that in the No Sale counterfactual, Suncorp Bank would either continue to provide the same level of service to its agribusiness customers or would provide a reduced level of service, as it has already unsuccessfully sought to provide by reducing the number of customers serviced by relationship managers.
732 Second, Professor King’s opinion went further than Ms Starks. He expressed the view that competition is potentially increased:
In my opinion, as a matter of economics, compared to the alternative buyer counterfactual, the acquisition of Suncorp Bank by ANZ is likely to substantially lessen competition in at least some local/regional agribusiness banking markets in Queensland. In particular, I note that while the acquisition of Suncorp Bank by Bendigo and Adelaide Bank would lead to a modest increase in market share for the merged entity in the supply of agribusiness banking services in Queensland, the complementary nature of the two banks in the provision of agribusiness banking services across Australia, and the strength of the Rural Bank brand, mean that competition is potentially increased by the acquisition of Suncorp Bank by Bendigo and Adelaide Bank compared to the status quo.836
733 Given Bendigo’s larger market shares in other states, its ownership of Rural Bank and its specialist agribusiness knowledge, and the existing resources available to it on a national level, a merged Bendigo/Suncorp Bank could produce a somewhat stronger competitor than Suncorp Bank alone. Nevertheless, the Tribunal is satisfied that any increase in strength would not be at a sufficient level to suggest that the Proposed Acquisition would be likely to have the effect of substantially lessening competition in Queensland agribusiness markets, compared with the Bendigo Merger counterfactual.
734 Third, for present purposes, the critical issue is competition in agribusiness banking markets in Queensland. As Professor King acknowledged, the addition of Bendigo’s market share of agribusiness banking in Queensland would only lead to a modest increase in the merged entity’s market share.837 Moreover, the analysis of competitor presence in local/regional agribusiness markets centred on Queensland towns, undertaken by Ms Starks, did not record Bendigo (or Rural Bank) as the supplier of banking services in any of the towns from which ANZ and Suncorp Bank provided agribusiness banking services.838 Bendigo did, however, claim that its bankers could cover the whole of Queensland, irrespective of where they were physically located.839
735 Mark Bennett gave evidence that Bendigo is not a significant competitor in regional Queensland agribusiness markets and, therefore, in many locations, an acquisition by Bendigo of another regional bank, such as Suncorp Bank, would make little difference.840 Mr Lane, the ANZ Queensland State Manager of Business Banking, also gave evidence to this effect and states that a merged Bendigo/Suncorp Bank is unlikely to achieve sufficient scale to enable it to compete more strongly.841
736 That Bendigo is not currently a significant competitor in local/regional agribusiness markets in Queensland is supported by its [REDACTED] market share in agribusiness lending in Queensland, being ranked [REDACTED], with a [REDACTED]% market share.842
737 The Tribunal is satisfied that the Proposed Acquisition, when assessed against either the No Sale counterfactual or the Bendigo Merger counterfactual, would not be likely to have the effect of substantially lessening competition in agribusiness banking markets in Queensland. In undertaking the necessary single evaluative judgment, the Tribunal gave particular attention to the following matters.
738 First, as explained above at [717]-[724], the Tribunal is satisfied that Suncorp Bank, which is a relatively strong competitor in Queensland, is not likely to impose any greater competitive constraint in the No Sale counterfactual, than it does at present. The Tribunal is also satisfied that a merged Bendigo/Suncorp Bank is not likely to impose any materially greater constraint on Queensland markets beyond that currently imposed by Suncorp Bank in the factual.
739 Second, as noted above in relation to the home loans market, while a merger between Bendigo and Suncorp Bank is a realistic commercial possibility, it is far from certain and there are significant execution challenges and risks, in particular, likely delays in the realisation of anticipated synergies and the impact of those delays on whether the merger would be ultimately value accretive to both Bendigo and SGL.
740 Third, many of the competitive concerns identified by Ms Starks in her first report fell away when she was provided with evidence of the distances that ANZ’s agribusiness relationship managers travelled in Queensland. In a response to an ACCC request for information, ANZ reported that there were no local areas in Queensland where its relationship managers were unable to visit and relationship managers servicing agribusiness customers were typically assigned a vehicle, which required to drive more than 18,000 km each year.843 Examples of areas visited included relationship managers in Townsville driving as far as Mt Isa, and other locations across North-West Queensland, and Cairns based relationship managers driving to Weipa.844
741 This evidence caused Ms Starks to opine that the geographic dimensions of the local/regional agribusiness markets in Queensland may well be much larger than a single major town.845 This, in turn, led Ms Starks to conclude that it was less likely that there were “one or more local markets where there will be a real chance of an SLC because the merger causes there to be only three or fewer competitors left”.846 Moreover, it caused Ms Starks to change her assessment of whether there was a real chance that the Proposed Acquisition would substantially lessen competition in the agribusiness markets from “Real chance of SLC” to “Cannot rule out SLC”.847
742 Fourth, contrary to the submissions of the applicants, the Tribunal considers that the supply of loans to agribusiness customers in the national market and in local/regional Queensland markets are concentrated, and that the Proposed Acquisition will lead to a significant increase in the concentration of those markets, particularly in Queensland.
743 The Tribunal is satisfied, however, that ANZ will remain constrained by other competitors such as NAB and Rabobank, and that Suncorp Bank’s agribusiness offering is not particularly unique, or unable to be replicated by other competitors. Critically, the Tribunal is satisfied that barriers to expansion in agribusiness markets in Queensland are relatively low, and, therefore, following the removal of Suncorp Bank as an independent competitor, there will be other competitors, capable of exerting a material constraint on the merged ANZ/Suncorp Bank.
744 There are other banks presently offering banking services to agribusiness customers in Queensland, which have the ability and incentive to expand their agribusiness banking services, if ANZ were to increase prices or reduce the levels of service, if the Proposed Acquisition proceeds. These include Bank of Queensland that has a specific Queensland focus, and Bendigo that has a significant agribusiness presence in other states and the ability, through Rural Bank, to expand its agribusiness presence in Queensland. In addition, Rabobank has an extensive presence in Queensland and is a specialist agribusiness bank, and CBA has [REDACTED] its share of agribusiness lending in Queensland in the [REDACTED] years to [REDACTED] from [REDACTED]% to [REDACTED]%.848
745 In his first witness statement, Mark Bennett gave evidence in support of this view, stating, “[a]gribusiness is currently an attractive sector for banks, in that it contains many strong potential customers, who typically seek fully secured transactions, which generate reasonable returns for banks”.849 He also stated that there is “a very limited history of losses in modern agribusiness banking”, citing foreclosure statistics published by the Australian Bureau of Agricultural and Resource Economics and Sciences for the period since 2016.850
746 For the foregoing reasons, the Tribunal is satisfied that the Proposed Acquisition would not be likely to have the effect of substantially lessening competition in any local/regional Queensland markets for agribusiness banking products.
H. COMPETITIVE EFFECTS OF THE PROPOSED ACQUISITION IN SME BANKING MARKETS
747 As was the case for agribusiness banking markets, the only competitive concerns raised in the local/regional SME banking markets in Queensland, were in relation to unilateral effects. No coordinated effects concerns were advanced to the Tribunal.
748 In order to assess the competitive effects of the Proposed Acquisition on SME banking markets in Queensland, it is again convenient to address first, the likely competitive effects in a future with the Proposed Acquisition, and then to compare those effects with the likely extent of competition in the future without the Proposed Acquisition, having particular regard to the No Sale counterfactual and the Bendigo Merger counterfactual.
749 The first question largely turns on the likely impact on competition of the removal of Suncorp Bank as an independent competitor in SME banking markets in Queensland. That issue requires a consideration of the extent to which (a) there is competitive overlap between ANZ and Suncorp Bank, (b) Suncorp Bank’s offering is differentiated and competitive, (c) there is a meaningful constraint imposed by the threat of entry or expansion by other banks, and (d) the broker channel is effectively utilised to source SME customers.
H.2. The principal contentions of the parties
H.2.1. The applicants’ principal contentions
750 The applicants submit that the Proposed Acquisition would not be likely to have the effect of substantially lessening competition in the supply of banking products to SME customers, regardless of whether there exists a discrete market for the supply of those products and services or the geographic dimension of the market.851
751 The applicants note that (a) no expert concluded that there was a real chance that the Proposed Acquisition would substantially lessen competition in the supply of banking products and services to SME customers, (b) the ACCC conceded that its conclusion to the contrary was “finely balanced”, and (c) Bendigo did not advance any contentions in respect of SME banking customers.852
H.2.2. The ACCC’s principal contentions
752 The ACCC submits that there is a sufficient basis in the evidence upon which the Tribunal could properly conclude that it is not satisfied that the Proposed Acquisition would not be likely to have the effect of substantially lessening competition in the market or markets for SME banking in Queensland.853
753 The ACCC points, in particular, to Ms Starks’ opinion, that she could not “be sure that there is no local market where the merger causes there to be only three or fewer competitors left”, and she could not “rule out that under ANZ’s leadership Suncorp will change its business model to be more like ANZ, reducing the number of non-major banks that operate based on a more personalised and flexible approach by one”.854
754 In addition, the ACCC submits that the evidence establishes that SME banking markets in Queensland are concentrated, Suncorp Bank is an effective competitor, not all banks rely heavily on brokers to acquire SME customers, and barriers to entry remain.855
H.3.1. The applicants’ submissions
755 The applicants submit that even if the supply of SME banking products and services is considered separately from the broader business banking market, the segment is not concentrated, and the Proposed Acquisition is not materially likely to increase concentration. They submit that although it is impossible to calculate concentration with any reasonable degree of precision, given there is no defined cohort of SME customers, the available data indicates supply is not concentrated nationally or in Queensland.856
756 The ACCC submits that it would be open for the Tribunal to conclude that the Queensland SME banking market or markets are concentrated.857 It submits that this conclusion is supported by the data it has obtained and analysed for 2022, which demonstrates that (a) the combined shares of the Major Banks for both SME lending and SME deposits in Queensland exceeds [REDACTED]%, (b) the share of both SME lending and SME deposits for Suncorp Bank in Queensland is approximately [REDACTED]% in each case, (c) Suncorp Bank’s share of SME lending in Queensland exceeded ANZ’s share of [REDACTED]%, (d) Suncorp Bank and ANZ were the [REDACTED] and [REDACTED] largest SME lenders in Queensland, and the [REDACTED] and [REDACTED] largest banks for SME deposits in Queensland, respectively, and (e) the Proposed Acquisition will result in ANZ’s market share in SME lending in Queensland increasing, by almost double, to [REDACTED]% and in SME deposits, its share would increase to [REDACTED]%.858
757 The ACCC submits that this data suggests that the Proposed Acquisition would materially increase concentration in what are already highly concentrated SME banking markets (or market) in Queensland.859
758 The Tribunal considers that SME banking markets in Queensland can fairly be characterised as concentrated.
759 In its Reasons for Determination, the ACCC included the following table, which summarised its calculations of the respective average market share of firms engaging in SME lending and SME deposits, from quarterly data provided by the top 11 banks in the period January 2022 to December 2022 (including an aggregate figure for a combined ANZ/Suncorp Bank) (SME market share table):860
760 The ACCC acknowledged that the calculations would likely overstate the market shares because data from other small banks and non-bank lenders, such as vehicle finance providers, have not been included. The ACCC nevertheless concluded in its Reasons for Determination, that the data provided a reasonable indication of the 11 leading banks’ market shares nationally, and in Queensland.861 The Tribunal shares that view.
761 The SME market share table, subject to the qualification above, shows the Major Banks in Queensland have a [REDACTED]% share of SME lending and an [REDACTED]% share of SME deposits. In addition, the table reveals that in Queensland, only Suncorp Bank, with the [REDACTED] largest market share ([REDACTED]%) in SME lending and the [REDACTED] largest share in SME deposits ([REDACTED]%), and [REDACTED], with the [REDACTED] largest market share ([REDACTED]%) in SME deposits and sixth largest share in SME deposits ([REDACTED]%), came close to ANZ, which is the Major Bank with [REDACTED] share of SME lending and deposits.862
762 It is readily apparent from the SME market share table, that if the Proposed Acquisition proceeds, ANZ’s market share of SME lending, nationally, would remain below that of the other Major Banks. In Queensland, however, ANZ’s market share in SME lending [REDACTED] and move it from [REDACTED] to [REDACTED] place in market share of SME lending. Similar, but not as significant, increases in market share would also be gained for SME deposits, both nationally and in Queensland.
763 Notwithstanding the Tribunal’s conclusion that SME banking markets in Queensland are concentrated, the evidence shows that regional and second-tier banks compete strongly for SME customers with smaller transaction sizes. A [REDACTED] board presentation in April 2021 reported that second-tier lenders [REDACTED].863 A SGL board submission in May 2021 concluded that the Major Banks had a strong foothold in the $1 million to $10 million turnover segment for SME customers but the “smaller players” had more success in the less than $1 million turnover segment. It reported that the non-Major Banks nationally had a market share of approximately [REDACTED]% of the $50,000 to $1 million segment but only [REDACTED]% of the $1 million to $10 million segment.864
764 Internal bank documents also suggest that Judo Bank has recently emerged as a “challenger” bank in the supply of banking products to SME customers. Judo Bank is a dedicated supplier of banking products to SME customers, and, in particular, [REDACTED].865
765 Mr Lane gave the following evidence:
In Queensland, the competitors that I see most often in the BB segment are (in order) NAB, CBA, Westpac, and, more recently, Judo Bank. I also see Suncorp Bank and BOQ from time to time, but they are not as strong. I rarely see Bendigo and Adelaide Bank/Rural Bank. 866
H.4. Competitive overlap between ANZ and Suncorp Bank
H.4.1. The applicants’ submissions
766 The applicants submit that ANZ and Suncorp Bank are not particularly close competitors in business banking. They submit that ANZ does not regard Suncorp Bank as a particularly strong competitor in business banking and does not set prices or make product changes in response to Suncorp Bank. They submit that ANZ and Suncorp Bank have different geographic presence and capabilities to serve medium and larger business customers, and this is reflected in very limited refinancing between ANZ and Suncorp Bank.867
767 The ACCC submits that Suncorp Bank exerts a “measure of constraint” on its competitors including ANZ in the supply of banking services to SME customers in Queensland.868 It seeks to rely on evidence from Mr Rankin, the ANZ Managing Director of Commercial and Private Banking, that Suncorp Bank (together with Bendigo and Bank of Queensland) exert “a degree of competitive pressure on ANZ Commercial”, although less than the Major Banks, and each was able to win more business in their “home” regions which reflected “the value of their brand and greater presence in those regions”.869
768 The Tribunal does not consider that ANZ and Suncorp Bank are particularly close competitors in the supply of banking products to SME customers in Queensland.
769 Mr Rankin’s evidence, read as a whole, makes clear that the principal constraints on ANZ, in the supply of business banking services relevant to SME banking customers, are the Major Banks, together with strong competition in specific product areas from (a) Judo Bank, Macquarie and non-bank providers in a range of specialist areas, and (b) “Regional Banks”, being Bendigo, Suncorp Bank and Bank of Queensland. As the ACCC noted, however, Mr Rankin expressly states the constraint imposed by the “Regional Banks” is less than that imposed by the Major Banks and is qualified by the phrase “a degree of”.870 Moreover, the ACCC’s reliance on the extracts from Mr Rankin’s evidence referred to above need to be considered in the context of the full text from which they were extracted:
While these banks tend to win a higher volume of business in their ‘home’ regions due to their historical footprint, my assessment is that this largely reflects the value of their brand and greater physical presence in those regions rather than any material geographic differences in any of the drivers of competition I set out above (service, credit policy, product features, price or origination options).871
770 In 2022, ANZ acquired, by way of refinancing, [REDACTED] SME business loan facilities, in an aggregate amount of $[REDACTED] and lost, by way of refinancing, [REDACTED] SME business loan facilities, in an aggregate amount of $[REDACTED]. Only [REDACTED]% of the aggregate amount of business loan facilities acquired by ANZ in 2022 were acquired from Suncorp Bank. Only [REDACTED]% of SME business loan facilities lost by ANZ in 2022 were lost to Suncorp Bank.872 [REDACTED] and [REDACTED] account for the largest amounts of SME loans refinanced from Suncorp Bank, although in recent months, [REDACTED] has captured the [REDACTED] share of SME loans refinanced from Suncorp Bank.873
771 Suncorp Bank is limited in its capacity to effectively compete for larger SME customers, because it is unable to offer more complex end-to-end banking services and more complex internet banking services,874 and its competitive strength in Queensland is, to a significant extent, reflective of its brand recognition and greater physical presence.875 [REDACTED].876 ANZ views [REDACTED].877
H.5. Is the Suncorp Bank offering differentiated and competitive?
H.5.1. The applicants’ submissions
772 The applicants submit that Suncorp Bank’s offering to SME customers is not materially differentiated from other competitors’ offerings, for the following reasons.
773 First, the applicants submit that Suncorp Bank is not a more vigorous or effective competitor than any other competitor. They submit that to the contrary, Suncorp Bank’s market share of SME customers is relatively small [REDACTED], being the MFI for only [REDACTED]% of its business customers, compared with a market average of 78%. They submit that it focuses on limited customer segments, it has been unable to [REDACTED], it does not have an end-to-end digital process (many of its processes remain manual), its turnaround times are [REDACTED], and it does not have the capability or risk appetite to offer, at scale, the transactional banking products necessary to meet larger business customers’ requirements.878
774 Second, the applicants submit that Suncorp Bank’s customer service offering is not unique or sufficient to impose a material competitive constraint. They submit that many other banks use a relationship management model for some or all of their SME customers (typically those with higher value loans or more complex banking needs, with most customers portfolio managed by generalist bankers using phone and digital channels).879
775 The applicants acknowledge that comparing the ratio of relationship managers to customers is an imperfect proxy to assess quality of an offering, but note that, in any event, ANZ has a [REDACTED] ratio of relationship managers to customers for most small to medium businesses than Suncorp Bank.880
776 The applicants also submit that (a) Suncorp Bank’s ability to compete in business banking is impeded by [REDACTED] infrastructure because, unlike ANZ, it has not made significant investments in automation and digitisation to improve speed and quality of service for SME customers and many of its SME processes are manual, (b) Suncorp Bank’s brand recognition and branch network is not unique, and its higher volume of business in Queensland reflects its historical footprint not material differences in drivers of competition, and (c) Suncorp Bank’s business banking net promoter score does not indicate material differentiation in its customer service offering.881
777 The applicants submit that ANZ would remain constrained by stronger competitors in the supply of SME banking products to customers if the Proposed Acquisition proceeds.882
778 The applicants submit that ANZ faces effective competition nationally and in Queensland from larger banks, particularly NAB, CBA, Westpac and Rabobank, and smaller banks including Macquarie, Judo Bank and Bank of Queensland. They submit that [REDACTED]. [REDACTED]. [REDACTED].883
779 The applicants also submit that even if competition occurs on a local/regional basis, in the future with the Proposed Acquisition, there will be no town in Queensland which would have fewer than five alternative bank branches (noting that having a bank branch is not necessary to compete for or supply banking products and services to SME customers). Competition between these banks occurs on the basis of price, quality of service, product features, and credit policy and is dynamic as non-banks lenders and fintechs force banks to respond to changes in customer expectations by investing in and providing new technology to consumers.884
780 As noted at [754] above, the ACCC submits that Suncorp Bank is presently an effective competitor in markets for the supply of banking products to SME customers in Queensland and exerts a measure of constraint on its competitors (including ANZ).885
781 The ACCC submits that (a) there is evidence that Suncorp Bank is competitive and responsive (albeit not market-leading) on price and regarded as competitive on non-price aspects, (b) Suncorp Bank has a strong customer service and care offering, (c) Suncorp Bank’s established network, brand recognition and customer relationships in Queensland enable it to exert strong competitive constraints in that State, (d) Suncorp Bank has a higher ratio of relationship managers to customers for most small and medium-sized businesses, and (e) Suncorp Bank has started taking steps to address the technology issues impacting its competitiveness in SME banking, and further investments may be made if the Proposed Acquisition does not proceed.886
782 Notwithstanding the limitations to the data in the SME market share table, it shows that Suncorp Bank has a relatively small national market share but is a stronger competitor in Queensland. Suncorp Bank is generally also included in other banks’ SME price monitoring and benchmarking assessments, although it is often listed alongside a range of other competitors, including other second-tier and regional banks such as Bank of Queensland, Macquarie and Bendigo.887
783 In an internal board submission dated April 2022 (April 2022 board submission), SGL noted that Suncorp Bank faced “[REDACTED]” including “[REDACTED]”, which may have contributed to its declining market shares in SME.888 The April 2022 board submission goes on to note that SGL had taken a number of steps to address deficiencies in Suncorp Bank’s SME offering, including [REDACTED].889
784 In the period between 2019 and 2022, Suncorp Bank’s share of SME lending declined from [REDACTED]% to [REDACTED]% in Queensland.890 By December 2022, as set out in the SME market share table, Suncorp Bank’s market share in SME lending in Queensland had further declined to [REDACTED]%. Despite this trend of loss of market share, particularly in SME lending in Queensland, the Tribunal accepts that Suncorp Bank’s offering in local and regional SME banking markets in Queensland presently exerts some competitive pressure on ANZ, particularly for customers with turnovers less than $1 million.
785 The Tribunal also accepts that Suncorp Bank’s offering is differentiated from ANZ and the other Major Banks but considers, for the following reasons, that Suncorp Bank’s offering is not materially differentiated from other regional and second-tier banks, in particular, Bank of Queensland.
786 First, as noted above at [782], the April 2022 board submission expressly acknowledges that in SME markets, Suncorp Bank had “[REDACTED]”.891 The April 2022 board submission goes on to state that:892
Business Banking is experiencing above average growth driven by a post COVID economic rebound, optimistic business sentiment and favourable trading conditions. [REDACTED].
787 Second, to the extent that Suncorp Bank’s customer service and “care” offering is competitive, the Tribunal does not accept that it imposes a materially stronger constraint, or is unique, or not easily replicable by other competitors. The ACCC relies on Suncorp Bank’s high business banking net promoter score (NPS) to suggest that Suncorp Bank’s SME offering is materially differentiated from the Major Banks.893 Suncorp Bank’s NPS is higher than the Major Banks’ average NPS. Relevantly, however, as at January 2023, Suncorp Bank’s NPS of [REDACTED] was significantly lower than Bendigo’s NPS of [REDACTED], and similar to the NPS of Bank of Queensland and NPS of Bankwest, which had increased their scores to [REDACTED] and [REDACTED], respectively.894 The similar NPS of Suncorp Bank and NPS of Bank of Queensland, in particular, suggests that whilst Suncorp Bank’s customer service offering is materially differentiated from the Major Banks, its offering is much less differentiated, when compared with other second-tier and regional banks.
788 Further, whilst Suncorp Bank provides a [REDACTED] of relationship managers to customers than ANZ, for businesses with lending between $[REDACTED] million and $[REDACTED] million,895 the ACCC relevantly stated in its Reasons for Determination that relationship management services were more relevant to medium rather than small businesses.896 Hence, as the applicants submit, comparing the ratio of relationship managers to customers is an imperfect proxy to assess the effectiveness of customer service offerings.897 Further, and relatedly, Suncorp Bank’s lack of investment in developing digital SME processes means that many of its SME processes are still manual and, therefore, less efficient than other competitors such as ANZ, rather than being reflective of a better “care” or customer service offering.
789 Third, Suncorp Bank’s offering is similar to Bank of Queensland’s SME offering in Queensland. In its Reasons for Determination, the ACCC states in respect of the removal of Suncorp Bank as an independent competitor in SME markets:898
[O]ther competitors appear to impose a competitive constraint. Bank of Queensland will impose competitive constraints most like Suncorp Bank, given the similarities in their scale, physical presence and product range.
H.6. Barriers to entry and expansion
H.6.1. The applicants’ submissions
790 The applicants submit that ANZ will remain constrained by the threat of expansion and new entry, which is an existing feature of the market. They submit that potential barriers to expansion identified by the ACCC, such as acquiring personnel and branch presence, and barriers to entry including regulatory barriers, are not significant. They further submit that Judo Bank has successfully attracted specialist SME bankers from ANZ in Queensland, branch presence is not necessary in order to compete successfully as demonstrated by the experience of Macquarie and Judo Bank and regulatory barriers have not prevented the establishment of over 100 new ADIs nationally.899
791 The ACCC submits that barriers to entry in SME banking remain high given the regulatory capital requirements, operational costs, and funding disadvantages that new entrants face, relative to incumbents.900
792 The ACCC accepts that barriers to entry are higher for SME deposits compared with SME lending because of the difference in regulatory requirements but submits an effective competitor in SME banking is likely required to offer both deposit products and loan products. This is due to many SME customers also wanting transactional banking products and preferring to hold multiple business banking products together with the same institution.901
793 The ACCC also submits that whilst Judo Bank is an example of recent market entry in SME banking, [REDACTED], and additionally, has been impeded in its ability to establish a branch network due to significant set-up and ongoing operational costs.902
794 It is necessary to distinguish between barriers to entry and barriers to expansion. The Tribunal accepts that barriers to expansion may be comparatively limited. While barriers to entry at the industry level remain significant, all the main players in the banking sector already participate in SME banking in Queensland, so the critical issue is barriers to expansion not barriers to entry.
795 The Tribunal accepts that attracting skilled SME banking personnel is not a significant barrier to expansion in SME banking markets for banks with an existing banking presence and operations in Queensland and nor is an extensive branch presence essential in order to be able to compete for SME banking customers.
796 Further, in its submission to the ACCC, ANZ had pointed to new entry into banking markets by Judo Bank, Prospa, Liberty Financial Group, Tyro Payments and Zeller.903 With the exception of Judo Bank, these were entries into banking markets in general, not SME banking markets. As Judo Bank submitted to the ACCC, it has been the only successful recent entrant into the business banking market.904 Judo Bank pointed to what it described as “very real and continued barriers to entry for new entrants”, that included regulatory capital settings, operational costs, funding disadvantages, and risk weighting settings.905
797 The Tribunal accepts that barriers to entry are lower for SME lenders that do not offer SME deposit products as they are not required to acquire an ADI licence and, therefore, are not subject to higher standards of prudential regulation. As the ACCC submitted, however, an effective competitor in SME banking is likely required to offer both deposit products and loan products,906 and this proposition is supported by market research conducted by Cameron Research, which demonstrates that SME customers require a cluster of banking services and prefer to acquire those services from a single bank:907 see also [363] above.
H.7.1. The applicants’ submissions
798 The applicants submit that brokers play an increasing role in driving competition for SME customers, and are critical for new entry and expansion, particularly for new, online and non-bank lenders, and originate a significant proportion of new SME loans. They submit that banks are making significant investments in developing their broker relationships as they recognise that SME customers are increasingly turning to commercial brokers for their business lending needs and ANZ has made significant investments to be a bank of choice for brokers.908
799 The ACCC accepts that brokers have played an increasing role in driving competition in SME lending in recent years but submits there is evidence to suggest that brokers’ ability to drive competition between banks in SME lending may be more limited. It submits that not all banks rely heavily on brokers to acquire customers, and brokers play a larger role in originating loans with non-bank and online lenders than they do with the established banks.909
800 The ACCC also submits that there is also evidence that (a) customer switching rates are low and customer “stickiness” is high, and (b) banks with a retail customer base also have a competitive advantage in gaining and retaining SME customers.910
801 The Tribunal accepts that brokers are performing an increasingly important role in the supply of banking products to SME customers.
802 Brokers currently originate approximately 40% of SME business for banks overall, including [REDACTED]% for ANZ and [REDACTED]% for Suncorp Bank.911
803 The significance of brokers was highlighted in a memorandum [REDACTED]:912
[REDACTED], [REDACTED]. [REDACTED].
804 Mr Rankin gave evidence that ANZ is currently engaging in work to manage its relationship with brokers, including by facilitating training and professional development courses for brokers, launching a rapid refinance process for business lending up to $1 million and developing a [REDACTED].913 The Mortgage and Finance Association of Australia (MFAA) in their response to an ACCC request for information noted that according to the Productivity Commission Research Paper into small business access to finance, some lenders had reported about 25% to 30% of new business originated from commercial brokers. The MFAA noted, however, that from their observations, the share of loans originated from commercial brokers varied “significantly across different lenders based on their business models”. The MFAA stated that it was very likely that non-bank and online lenders source a much larger proportion of their customers through brokers than larger banks with a physical presence.914
H.8. Constraint imposed by Suncorp Bank in the No Sale counterfactual
805 The parties generally relied on their submissions on the current competitive constraint imposed by Suncorp Bank in SME banking markets as providing the best guide to the constraint that Suncorp Bank would impose in the No Sale counterfactual.
806 For the reasons advanced above at [768]-[771], the Tribunal considers that an independent Suncorp Bank in the No Sale counterfactual would be likely to impose a relatively limited constraint in markets for SME banking in Queensland. Further, the Tribunal considers that the presently foreshadowed increased expenditure on technology systems would assist in maintaining rather than expanding market share.
H.9. Constraint imposed by merged Bendigo/Suncorp Bank in the Bendigo Merger counterfactual
H.9.1. The applicants’ submissions
807 The applicants submit that a merged Bendigo/Suncorp Bank is unlikely to impose any materially greater competitive constraint than Bendigo or Suncorp Bank separately, if the Proposed Acquisition proceeds. They submit that there would be no increase in competitive constraint in the Bendigo Merger counterfactual, given Bendigo is currently not present, or has only limited presence, in many of the locations in which ANZ and Suncorp Bank supply banking products and services to SME customers. In contrast, they submit that competitive constraints imposed by Bendigo may well be greater if the Proposed Acquisition proceeds than in the Bendigo Merger counterfactual because Bendigo could merge with Bank of Queensland and expand its presence in the locations in which ANZ and Suncorp Bank supply banking products and services to SME customers.915
808 The applicants submit that to the extent that relationship management has a viable commercial future, whether or not Suncorp Bank remains under independent ownership will make no difference to its availability. They submit that relationship management is not unique to Suncorp Bank, but even if it were, there is no reason why, if that was attractive to customers, another entity would not provide the same service.916
809 They submit that the Suncorp Bank’s relationship managers will either be providing the same relationship management at ANZ in competition with other participants in the market, or if that does not occur for some reason, they will almost certainly be providing those same services for another competitor.917
810 Neither Bendigo, nor the ACCC, advanced specific submissions addressing the constraint that would otherwise be imposed by a merged Bendigo/Suncorp Bank in SME banking markets in Queensland, in the Bendigo Merger counterfactual.
811 The SME market share table suggests that Bendigo presently has a limited and small market presence in Queensland, [REDACTED]. [REDACTED].918
812 A merged Bendigo/Suncorp Bank is, therefore, unlikely to impose any significantly greater constraint than Suncorp Bank would provide in the No Sale counterfactual on other competitors in SME banking markets in Queensland.
813 On balance, the Tribunal is satisfied that the Proposed Acquisition would not be likely to have the effect of substantially lessening competition in SME banking markets in Queensland.
814 The Tribunal accepts that SME banking markets in Queensland are relatively concentrated, and there are relatively high barriers to entry. Barriers to entry are relatively high because of the regulatory barriers to becoming an ADI. While a new entrant may offer SME loans without becoming an ADI, it would not be able to offer SME or other deposit products and, therefore, would be at a significant competitive disadvantage because it would have to rely on typically more expensive alternative sources of funds.
815 The Tribunal accepts that there will be some loss of competition for SME banking customers from the removal of Suncorp Bank as an independent competitor, if the Proposed Acquisition proceeds.
816 The Tribunal notes that the applicants, Bendigo, and the ACCC did not specifically address competition in the future without the Proposed Acquisition in the No Sale counterfactual for SME banking markets in Queensland. The Tribunal has concluded that in the No Sale counterfactual, Suncorp Bank is likely to (a) remain an effective competitor in Queensland SME banking markets by reason of its substantial market share, with an established branch network, brand recognition and customer relationships in Queensland, and (b) continue to take steps to seek to address its technological issues, and inability to differentiate in the market, in order to seek to retain and expand its market share.
817 In the Bendigo Merger counterfactual, Suncorp Bank would be removed as an independent competitor, but a merged Bendigo/Suncorp Bank would raise fewer competition concerns because of, (a) the more limited geographic overlap between Bendigo and Suncorp Bank, particularly in comparison to ANZ and Suncorp Bank, and (b) the lower levels of market concentration for a merged Bendigo/Suncorp Bank.
818 These considerations, however, are outweighed by the following matters.
819 First, as noted above in relation to the home loans market, while a merger between Bendigo and Suncorp Bank is a realistic commercial possibility, it is far from certain and there are significant execution challenges and risks, in particular, likely delays in the realisation of anticipated synergies and the impact of those delays on whether the merger would be ultimately value accretive to both Bendigo and SGL.
820 Second, there will still be a large number of competitors remaining in the SME banking markets in Queensland, including each of the Major Banks, as well as Rabobank and a range of regional banks. If the Proposed Acquisition proceeds, ANZ would have strong incentives to compete against the Major Banks, the regional banks and Judo Bank to maintain its increased customer cohort, and the other three Major Banks would continue to exert a significant competitive constraint on ANZ.
821 Third, barriers to expansion in SME banking markets in Queensland are relatively low. There is no evidence that existing banks offering SME banking products have encountered any difficulties in attracting SME bankers. Nor is an extensive branch presence necessary to compete, as demonstrated by Macquarie and Judo Bank. Banks that lack a local presence may potentially offer cash management services for SME customers anywhere there is a post office and other services such as account opening and loans could be effectively offered online or potentially in a branch in a regional centre.
822 Fourth, ANZ and Suncorp Bank are not particularly close competitors in the supply of banking products to SME customers in Queensland. As noted at [765] above, this is reflected in the very limited refinancing between ANZ and Suncorp Bank. Further, Suncorp Bank focuses on servicing the lower end of the SME banking market and several other competitors would remain in that space.
823 Fifth, any loss of competition and consumer welfare because ANZ determined it was not profit maximising to continue to offer the differentiated service offering provided by Suncorp Bank to its SME customers could be expected to be addressed by banks with similar product offerings. Banks with similar product offers would be more likely to compete for these customers as they would be less likely to cannibalise their existing customer base. Bank of Queensland, and to a lesser extent, Bendigo and Judo Bank, given their much smaller existing market share, would appear to be the most likely banks that would fill any vacated product space.
824 Sixth, the increasing use of brokers in SME lending in Queensland is likely to assist SME customers to switch, and to drive competition between existing banks.
825 Further, the Tribunal notes that the ACCC referred to Ms Starks’ opinion, that she could not “be sure that there is no local market where the merger causes there to be only three or few competitors left”, and she could not “rule out that under ANZ’s leadership Suncorp will change its business model to be more like ANZ, reducing the number of non-major banks that operate based on a more personalised and flexible approach by one”.919
826 Tests of could not “be sure” and could not “rule out”, however, are not the relevant test under s 90(7)(a) of the CCA. They cannot be used as substitutes or proxies for the statutory language of “would not have the effect, or would not be likely to have the effect, of substantially lessening competition”. Ms Starks’ opinion that she could not “rule out” a substantial lessening of competition in SME banking markets in Queensland if the Proposed Acquisition proceeds does not address the relevant question.920
I. PUBLIC BENEFITS AND DETRIMENTS OF THE PROPOSED ACQUISITION
827 Given the conclusion of the Tribunal that it is satisfied in all the circumstances that the Proposed Acquisition would not be likely to have the effect of substantially lessening competition, it is strictly not necessary to address the net public benefits test in s 90(7)(b) of the CCA.
828 Nevertheless, in light of the extensive submissions made by the parties directed at whether the Proposed Acquisition would result, or be likely to result, in a net benefit to the public, the Tribunal has determined that it is appropriate to address those submissions.
829 The Tribunal also notes that pursuant to Direction 1 of the directions made on 15 December 2023, it has addressed the written submissions of the State of Queensland dated 6 October 2023, for the purpose of clarifying information provided to the ACCC in connection with the making of its Reasons for Determination. It was, therefore, not necessary for the Tribunal to have regard to statements of the Queensland Treasurer and Minister for Trade and Investment made in the Queensland Legislative Assembly on 16 June 2023, or to consider the operation of s 8 of the Parliament of Queensland Act 2001 (Qld), in respect of those statements.
I.1.1. The principal contentions of the parties
I.1.1.1. The applicants’ principal contentions
830 The applicants submit that the Proposed Acquisition will result in significant and measurable public benefits. They submit these benefits are “merger specific”921 for the Proposed Acquisition and include (a) costs-savings driven by productive efficiencies from integration synergies,922 (b) increases in prudential safety of Suncorp Bank,923 (c) reduced cost of funding for Suncorp Bank,924 (d) an increase in the Major Bank Levy,925 (e) SGL will become a “pureplay” insurer,926 and (f) substantial benefits for the Queensland economy.927 They submit that the substantial benefits for the Queensland economy are to be delivered pursuant to agreements that are “causally connected” to the Proposed Acquisition, and not “merely coincident” with the Proposed Acquisition.928
831 The applicants submit that the Proposed Acquisition will not give rise to any meaningful competitive detriments.929
I.1.1.2. Bendigo’s principal contentions
832 Bendigo submits that the public benefits claimed by the applicants, to the extent that the Tribunal might accept that they should be taken into account pursuant to s 90(7)(b), would or would also likely arise in the Bendigo Merger counterfactual.930 More specifically, it submits that any benefits from SGL’s divestment of Suncorp Bank would arise, equally, in the Bendigo Merger counterfactual, and it is also likely that equivalent productive efficiencies from integration synergies, and an increase in the Major Bank Levy would also be achieved.931
833 Bendigo submits any public benefits that might arise from the Proposed Acquisition would be outweighed by (a) competitive detriments in the home loans market and the agribusiness banking markets, and (b) harm to the Australian retail banking industry by entrenching the oligopoly market structure of the Major Banks, and removing the only real chance for a second-tier bank such as Bendigo to challenge effectively the Major Banks, by a step up in scale.932
I.1.1.3. The ACCC’s principal contentions
834 The ACCC advances the following principal contentions in response to the benefits of the Proposed Acquisition, claimed by the applicants.
835 First, any improvements to SGL’s insurance business would be small and could equally be expected to arise in the Bendigo Merger counterfactual.933
836 Second, there is likely to be a public benefit realised from integration synergies, but in a lower amount than claimed by the applicants,934 and “some measure” of integration synergies are also likely to be realised in the Bendigo Merger counterfactual.935 In addition, it submits that the extent of the synergies remains uncertain, the synergies may be further delayed and the extent of any pass through of benefits to consumers will depend on the degree to which any savings relate to fixed or variable costs savings, and the intensity of competition following the Proposed Acquisition.936
837 Third, any benefits from improving the prudential safety of Suncorp Bank are unlikely to be material because there is no meaningful risk of Suncorp Bank failing. Further, any benefits would be offset by requirements to tie up additional capital, and the Proposed Acquisition would lead to a larger ANZ that would thereby increase systemic risk as the consequences of failure would be more widespread and significant.937
838 Fourth, any increase in the Major Bank Levy would not be a meaningful public benefit because any increase to government revenue is offset against the risk of harm that larger banks pose.938 Further, any perceived benefit could be expected to arise also in the Bendigo Merger counterfactual as a merged Bendigo/Suncorp Bank would equally be likely to be subject to the Major Bank Levy, and the quantum of the increase would be larger because both Bendigo and Suncorp Bank would be captured.939
839 Fifth, there is potentially some benefit in the form of lower funding costs, but any benefits are likely to be small, and at least partially, offset by higher contributions for the Major Bank Levy and higher capital requirements. Further, Ms Starks’ view is that these savings are unlikely to constitute efficiency benefits and the extent of any pass through to consumers is uncertain.940
840 Sixth, the commitments made by the applicants to the State of Queensland cannot be taken into account as public benefits under s 90(7)(b) because they are only coincident with the Proposed Acquisition, and do not result from the conduct for which authorisation is sought. The ACCC further submits that any benefits to Queensland may well give rise to correlative detriments in other States, and to the extent that any benefits may be profitable to other banks, including Bendigo, those banks could be expected to pursue them if the Proposed Acquisition does not proceed.941
841 The ACCC submits that the most significant detriments of the Proposed Acquisition are its likely impacts on competition in the markets for home loans, retail deposits, agribusiness banking and SME banking.942
842 The ACCC submits that the Proposed Acquisition involves a substantial public detriment, being the loss of an attractive acquisition target for an existing smaller bank and, in turn, the loss of the most meaningful opportunity for a second-tier bank to better compete through an increase in scale.943 The ACCC contends that this detriment extends beyond market specific detriments, and should be brought to account as public detriments under s 90(7)(b) of the CCA.944
843 The ACCC submits that given the substantial public detriments that are likely to result from the Proposed Acquisition, and the limited public benefits that might be considered likely to be realised, the Tribunal might properly conclude that it is not satisfied that the public benefits from the Proposed Acquisition would outweigh the resulting public detriments. 945
I.2.1. SGL becoming a “pureplay” insurer
844 The Tribunal accepts that the SGL conglomerate structure is not providing its insurance business, and more importantly, its consumers, with any advantages. Rather, it would appear that there are only, or, at least, largely diseconomies of scope reflected in the conglomerate discount [REDACTED].946
845 The Tribunal also accepts that (a) SGL may well be better placed to confront the challenges posed by increasing natural disasters, attributable to climate change, without the distraction of operating a second-tier bank and (b) SGL’s divestiture of Suncorp Bank may result in increased capitalisation from the removal of the conglomerate discount. Any increase in market capitalisation might well only reflect market confidence in future earnings, which may or may not increase value to consumers.
846 Further, the Tribunal accepts that to the extent that there is a real resource saving from increased efficiency, or SGL being able to compete better in insurance markets for the benefit of consumers, this would constitute a public benefit.
847 The Tribunal, however, is of the view that each of these benefits could accrue independently of the Proposed Acquisition by SGL divesting Suncorp Bank by a spin off or a sale to a bank, other than a Major Bank.
I.2.2. Productive efficiencies from integration synergies
848 The applicants claim that the annual cost synergies from productive efficiencies that could be realised from the Proposed Acquisition would be approximately $260 million, pre-tax, and would commence to flow some four to six years after the Proposed Acquisition.947 This figure represents approximately [REDACTED]% of the cost base of Suncorp Bank.948
849 Both the quantum of those synergies, and the time by which they could be achieved, have been subject to extensive due diligence by ANZ. Ms Higgins gave evidence of the extent of the due diligence undertaken by ANZ with the assistance of external and subject matter experts, since December 2021, to determine the likely value of the integration synergies that could be realised from the Proposed Acquisition. Ms Higgins provided a comprehensive explanation of the steps taken to estimate Suncorp Bank’s cost base, identify and estimate cost synergies and one off integration costs, and finalise other due diligence estimates.949
850 Further, the ANZ estimate of the percentage of the synergies claimed to the cost base of Suncorp Bank is consistent with Barrenjoey’s calculation in a November 2021 presentation to the SGL board that the average synergy realisation from mergers and acquisitions of large Australian banks in the period 1995 to 2020, as a percentage of the target’s cost base, was 34%.950 Further, Ms Starks referred, in her first report, to an analysis that [REDACTED].951
851 Equally, Mr Smith’s estimate of $[REDACTED] million in integration costs, which is close to [REDACTED] the estimated run rate synergies, appears to be reasonable when compared with the [REDACTED] average of integration costs to run rate synergies incurred in mergers and acquisitions of large Australian banks in the period between 1995 and 2020.952
852 The Tribunal accepts that real resource cost savings from spreading fixed costs, labour cost reductions and branch rationalisations represent gains in productivity efficiency and, therefore, constitute public benefits, to the extent that they are merger specific and more likely to be realised in the future with, rather than without, the Proposed Acquisition.
853 The Tribunal is satisfied that the cost savings identified by the applicants are more likely in the future with the Proposed Acquisition. Although [REDACTED], for the reasons provided at [292]-[322] above, while it might be a realistic commercial possibility, there remains considerable uncertainty as to whether the merger would occur in that counterfactual. Further, and more relevantly, the time period in which those synergies could be realised is uncertain because Bendigo was not able to undertake any due diligence953 of the quantum and timing of the likely costs savings.
854 The Tribunal accepts, however, that costs savings from branch closures that have already occurred, estimated by Mr Smith to be in vicinity of $[REDACTED] million,954 need to be excluded (representing only some [REDACTED]% of the claimed run rate synergies). The Tribunal also accepts that it would be necessary to net off integration costs. Mr Smith estimates that the net present value of total costs savings expected from the Proposed Acquisition would be between $[REDACTED] and $[REDACTED] (after excluding pecuniary savings and netting off integration costs and dis-synergies).955
I.2.3. Prudential safety benefits
855 Following the Proposed Acquisition, Suncorp Bank, would become part of ANZ and would thus become a D-SIB and be subject to higher prudential standards and regulatory oversight by APRA. At the same time, it would have reduced costs of funds, and access to a lower cost of capital than it currently enjoys.
856 In evaluating the extent of any prudential benefits that might result from the Proposed Acquisition, it is important not to double count the benefit of reductions in risk from increased prudential requirements, as explained by Dr Carmichael,956 and the benefit of reductions in the cost of funds, by reference to Mr Smith’s analysis of costs savings, to the extent that this reflects the same underlying reduction in risk.
857 The Tribunal accepts that there would be some prudential benefits from the Proposed Acquisition, but, ultimately, is not persuaded that they would be likely to be significant.
858 First, while it follows that an increase in prudential requirements and becoming a part of a larger and more diversified bank would make it less likely for Suncorp Bank to fail, it is doubtful that any benefit to Suncorp Bank would be substantial. Given Suncorp Bank’s existing support from SGL and financial position, any current risk of failure is negligible. The difference may be no more, as suggested by Dr Carmichael, than a difference between Suncorp Bank being a bank that was an “unquestionably strong bank”, and Suncorp Bank becoming a bank that was “an unquestionably stronger bank”.957 It follows that any reduction in systemic risk to the Australian financial system would similarly not be significant.
859 Second, the greater capital requirements and level of prudential scrutiny associated with becoming a D-SIB bank, that would lead to more capital being tied up, together with increased regulatory costs that would be incurred by both APRA and ANZ, would reduce the value of any prudential benefits that might otherwise have arisen from the Proposed Acquisition.
860 Third, Suncorp Bank’s access to a higher credit rating,958 and a lower cost of capital as a result of the Proposed Acquisition would either (a) reflect a real resource cost saving associated with a reduction in underlying risk that would have been already taken into account, if there was a substantive reduction in risk, or (b) would not otherwise be a public benefit to the extent it constituted a transfer of risk. The transfer of risk is transferring risk to the government and taxpayers because Suncorp Bank becomes part of a bank that is considered “too big to fail”.
861 Fourth, it is not apparent, contrary to Mr Smith’s evidence,959 how a lower cost of funds would represent a productive efficiency gain attributable to a reduction in risk associated with an expectation of support from ANZ. No reason was advanced as to why ANZ could be expected to provide greater support to Suncorp Bank than SGL. To the extent that any increase in Suncorp Bank’s credit rating following the Proposed Acquisition reflected a real reduction in risk, it would seem more likely that this was due to increased prudential requirements associated with being a D-SIB bank, and becoming part of ANZ’s more diverse portfolio.
I.2.4. Increase in Major Bank Levy
862 Following its acquisition by ANZ, Suncorp Bank would become subject to the Major Bank Levy. The benefit alleged by the applicants is that this would result in an increase of taxation revenue of approximately $24 million per annum, and public benefits from the use to which that revenue could be put.960
863 This alleged benefit is different from the net benefit found by the Tribunal in Application for Authorisation of Acquisition of Macquarie Generation by AGL Energy Limited [2014] ACompT 1 (MacGen). The issue in MacGen was whether the use of government revenue generated from the privatisation of MacGen was a public benefit. It concerned revenue derived from a change in ownership of assets from the government to the private sector not, as in this case, a transfer of surplus from a merged entity to the government.
864 In any event, to the extent that the “tax benefit” in this case might by characterised as a benefit, it would best be understood as a “forced pass through” to the public of some of the cost efficiencies likely to arise from the Proposed Acquisition.