Lodge: Old and New Questions About the Analysis of Cartels in New Zealand

By Dr Mark Berry¹

I. Introduction

Cartel cases under Part 2 of the Commerce Act 1986 (NZ) (“the Act”)² have, not surprisingly, dominated the restrictive trade practices enforcement landscape in New Zealand.  However, notwithstanding some 35 years of jurisprudence, uncertainty has continued to surround some of the key analytical concepts applying to cartels.

Recently the New Zealand Supreme Court was required for the first time to opine on a range of these concepts in Lodge Real Estate Ltd v Commerce Commission (“Lodge”).³  This paper reviews three of the issues addressed in this decision, namely:

• The legal test applying to the establishment of an “arrangement” under sections 27 and 30 of the Act, so far as it requires more than a mere expectation that at least one party will act (or not act) in a certain way;⁴
• Whether a discretion to depart from an arrangement provides a basis for concluding that there is no purpose or effect of fixing, controlling, or maintaining price under section 30 of the Act. This case involved facts which required only a consideration of the “controlling” of price limb; and
• Whether a “de minimis” exception applies under section 30 of the Act where the arrangement relates only to control over a small component of the overall price.

II. Background

For context, a brief outline of the facts in Lodge is necessary.  This case, like others before, exposes the risks that arise where competitors are required to respond to increased common costs,⁵ or where competitors contemporaneously seek to end common concessions (such as free add-on products or services) which may have arisen because of price matching resulting from past competition.⁶

This case centred upon Trade Me’s costs for advertising real estate.  Trade Me was at the relevant time the most prominent third-party website for the advertising of listed properties.  The other website for such advertising was realestate.co.nz.  This is collectively owned by the real estate agencies, but is not as comprehensive as Trade Me.

Until 2013, Trade Me provided a standard listing service to the real estate agencies on a basis which capped the amount payable by any agency in any given month.  Trade Me charged a base monthly amount plus a fee for each listing.  However, because the cap was set relatively low, all listings after the first five or six were effectively free.  Because of this low cost for the service, it was normal for agencies to absorb the Trade Me costs.

In 2013, Trade Me decided to radically change its approach to pricing.  A fee was proposed for each individual standard listing of $159 plus GST, with no cap.  The impact of this was profound.  In the case of Lodge, it faced an increase in annual Trade Me listing fees from $8,000-$9,000 to around $200,000-$220,000 if it continued to absorb these costs.⁷

Trade Me’s communications with the real estate agents suggested that these costs should be passed on to consumers, rather than be absorbed by these agencies.  However, this new cost created a competitive tension.  Critical to the success of these agencies is the winning of new listings.  If agents continued to absorb the Trade Me listing costs, this could well have been determinative of a vendor’s choice of agent where other agents were not prepared to accommodate this cost.  The risk of this competitive tension is apparent from the comparative costs of Trade Me listings and the average commission payable by vendors if the property is sold. The Trade Me fee of $159 plus GST to agents was a small amount considered alongside the average commission of $15,000.⁸

Various communications and meetings followed between the competitor agencies.  A default position emerged whereby the agencies would no longer absorb the Trade Me listing fee.  Rather, this listing fee was to be passed on to either the customer or the salesperson.⁹

III. The Legal Test Applying to “Arrangements”

The issue whether there was an “arrangement” between the agencies in Lodge was hotly contested.  The agencies accepted that there was a consensus among them that they would adopt a vendor funding model as a default position for the Trade Me listings in future.  However, they argued that each agency had independently come to this view before they met to discuss this development.  They further argued that while there was both consensus and expectation between them, there was no conditionality (one party agreeing to do X on condition that the other also did X) and no moral obligation on the part of any agency to do so.¹⁰

The Supreme Court upheld the finding that there was an “arrangement” in the relevant sense.  There was a consensus reached by the agencies which involved a commitment from each of them to adopt a vendor funded model for Trade Me listings and to remove existing listings in January 2014.  This created an expectation of a common course of conduct.¹¹

The matter that is, however, of particular interest here is the Supreme Court’s review of the legal test for establishing the existence of an “arrangement”.  As will be apparent from the above outline, the primary submission on appeal was that, for an “arrangement” to exist, the relevant parties had to have made a commitment to each other in the sense that they became subject to a “moral obligation” to act in the proscribed manner.

Prior to the decision of the Supreme Court in Lodge, the landscape on this subject was complicated.  High level judicial rule formulations can be open to different interpretations.  Such rule formulations in respect of what amounts to an “arrangement” under competition law is a prime example, given their dependency on the interaction of concepts of “expectation”, “moral obligation”, and “commitment”.  These are different concepts in ascending order of proof requirements, as will be further discussed below.

The leading New Zealand case on these concepts prior to Lodge was the decision of the Court of Appeal in Giltrap City Ltd v Commerce Commission (“Giltrap”).¹²  The majority decision delivered by Tipping J concluded:¹³

While the concept of moral obligation is helpful in that it will often reflect the effect of an arrangement or understanding under s 27, the flexible purpose of the section is such that it is best to focus the ultimate inquiry on the concepts of consensus and expectation.  A finding that there was a consensus giving rise to an expectation that the parties would act in a certain way necessarily involves communication among the parties of the assumption of a moral obligation.

Before there can be an arrangement under s 27 (or for that matter an understanding) there must be a consensus between those said to have entered into the arrangement.  Their minds must have met – they must have agreed – on the subject matter.  The consensus must engender an expectation that at least one person will act or refrain from acting in the manner the consensus envisages.  In other words, there must be an expectation that the consensus will be implemented in accordance with its terms.

This aspect of the majority’s decision involves ambiguity.  On the one hand, it appears to formulate a test based on an “ultimate inquiry” as to expectation.  However, on the other hand, this inquiry appears to be qualified on the basis that it also necessarily involves an assessment of the assumption of a “moral obligation”.

The minority decision of McGrath J in Giltrap, realising that the majority decision may potentially capture conscious parallelism given its apparent focus on expectation,¹⁴ commented as follows:¹⁵

As these concepts [consensus and expectation] carry the notion of a moral (or non-legal) obligation, that in my view should remain an important touchstone for determining whether there is an arrangement or understanding under s 27.

In most cases of apparently collusive behaviour the existence of moral obligations between parties will point strongly to the existence of an arrangement or understanding.  It seems to me that in the context of restrictive trade practices one cannot have an expectation to the necessary degree that another will perform an act unless the first person considers the other legally or morally obliged to do so.

Such background provided fertile ground for argument in Lodge.  The Supreme Court interpreted the majority in Giltrap to say that mere consensus giving rise to an expectation that parties would act in a certain way is not enough and that it is necessary also to establish the existence of a moral obligation between the parties.¹⁶  It followed that the Supreme Court did not consider that the majority decision in Giltrap differed in any material way from the decision of McGrath J.

However, the Supreme Court proceeded to express some unease with the concept of “moral obligation”.  It said that “calling such an obligation a ‘moral obligation’ introduces morality into a context where it adds nothing.”¹⁷  For this reason the Court preferred to “substitute ‘made a commitment’ in place of ‘assumed a moral obligation’.”¹⁸  The Court concluded:¹⁹

If there is a consensus or meeting of minds among competitors involving a commitment from one or more of them to act (or refrain from acting) in a certain way, that will constitute an arrangement (or understanding).  The commitment does not need to be legally binding but must be such that it gives rise to an expectation on the part of the other parties that those who made the commitment will act or refrain from acting in the manner the consensus envisages.

The Supreme Court has, therefore, resolved the uncertainty surrounding the Giltrap formulation.  At the heart of the test now lies the need objectively to establish a “commitment” rather than a “moral obligation” between the parties.²⁰  What impact will this new test likely have?  Will the “commitment” test involve a different standard of proof?

Both “commitment” and “moral obligation” are terms that are capable of wide and varied meanings.

The concept of “moral obligation” is suggestive of some duty that is owed, which ought to be performed, but which may not be legally binding.  It is a concept that embraces ethical imperatives based upon universal views of what might be good and right.²¹  “Moral obligation” is a concept that does not sit comfortably in the competition law context, and the Court’s reservations about it are understandable, notwithstanding its case law origins.²²  The issue of morality has no obvious place, and has not previously received appropriate scrutiny, in this context.

The replacement term “commitment” is, however, potentially open to different interpretations.  “Commitment” can, in the commercial law context, involve an agreement or pledge to do something in the future.²³  On this definition, “commitment” has a potentially expansive meaning.  At one end of the spectrum, it could extend to include a contractual obligation, although this interpretation is not open here because sections 27 and 30 of the Act refer to both “arrangements” and “contracts”.  The scheme of such provisions therefore indicates that the concepts differ and, as a result, an “arrangement” will involve “a consensual dealing lacking some of the essential elements that would otherwise make a contract”.²⁴  Nonetheless, “commitment” may on one interpretation require the establishment of an agreement which, while not strictly speaking legally enforceable, will otherwise display many of the characteristics of a contract.  In particular, for there to be a “commitment”, the critical terms of importance to the cartelists may need to be established.  For example, it may need to be demonstrated that the communications between the parties have resulted in an actual agreement as to future co-ordinated price conduct.

Another possible meaning to be given to the term, which is discernible from Australian case law, is that a “commitment” requires that at least one party will “assume an obligation” or give “an assurance or undertaking” to act (or not to act) in a certain way.²⁵  These definitional guidelines may be indicative of something less than the establishment of an actual agreement.

Accordingly, there is some uncertainty as to the standard of proof that the “commitment” test may impose.  Subtle distinctions exist between these various formulations.  There is, however, the potential that the “commitment” test will set a reasonably high bar, and one that is likely higher than that which applied to “moral obligation”.²⁶

IV. The Discretion to Depart from the Arrangement

The High Court in Lodge²⁷ accepted that there was an arrangement between real estate firms who were in competition with other.  However, the Commission’s case failed in the High Court on the one ground that the arrangement did not have the purpose, effect or likely effect of controlling (or providing for the controlling) of the price for real estate sales or advertising services provided by the real estate companies.  This conclusion was based on a finding that the vendor funding arrangement entered into between the real estate agents did not “prevent, in particular or desirable circumstances, the agency and/or agent bearing some portion or all of [the Trade Me funding] expense”²⁸ and that, therefore, the arrangement did “not interfere with the competitive setting of the price”.²⁹  The retention of the freedom of pricing discretion to absorb part or all of the Trade Me costs, despite the arrangement, formed the basis for the High Court’s decision that the arrangement did not have the effect of controlling, or providing for the controlling, of price.  It was also found that this meant that, objectively, the arrangement could not have the purpose of controlling the price.³⁰

This particular focus on the discretionary aspect of the vendor funding arrangements was not argued before the High Court.  The High Court did endeavour to provide details of the extent to which agencies had paid part or all the Trade Me fee, but there was no reliable information.  To the extent that there was data, it reflected a very limited scale of departure.  Over the months of February and August 2014, Lodge paid all the Trade Me fee on twelve occasions and part of this fee on three occasions.  Lodge had in the order of 1,500 listings per year.³¹  These numbers, therefore, reflected that the exercise of the discretion arose in less than one per cent of cases.

Intuitively, it is difficult against this background to comprehend the High Court’s adoption of this freedom to depart exemption.  Clearly, this approach left open the prospect of avoidance of the cartel provisions.  Sophisticated cartelists on this analysis needed only to build into their agreements a little room to deviate at the margins to escape liability.

The High Court’s decision on this issue was overturned by both the Court of Appeal and Supreme Court.  These appellate courts differed to some degree in their approach to this issue, although ultimately there is no conflict between them.  These decisions provide clarity on the test for “control” and expose the obvious errors in the High Court’s analysis.

The Court of Appeal noted that it may often be that within cartels there will be room for the exercise of some discretion over price.³²  The Court of Appeal’s central finding on the “control” of price issue was that “collusion on a start or offer price can be enough”.³³  It followed that “obviously the starting position of any vendor as to price will have some effect on the ultimate agreement on price.”³⁴

The Supreme Court accepted that agreement on the offer price may have the effect of controlling price.³⁵  However, the Court proceeded to describe relevant principles within a wider context and relied upon what it considered to be longstanding principles from earlier Australasian decisions from which it saw no reason to depart.  The Court cited with approval the following statement of principle from the Federal Court in ACCC v CC (NSW) Pty Ltd:³⁶

The word “control” is not defined in the [Trade Practices] Act.  Its natural or ordinary meaning is “to exercise restraint or direction over” (the Macquarie Dictionary) or “to exercise restraint or direction upon the free action of” (the Oxford English Dictionary) a person or thing.  There are degrees of control and there may be control although the “restraint” or “direction” is not total.  An arrangement or understanding has the effect of “controlling price” if it restrains a freedom that would otherwise exist as to a price to be charged.

The Supreme Court also found that the earlier decision of the High Court in Caltex applied this approach, and that there was no basis for distinguishing this case from the present.³⁷  In Commerce Commission v Caltex New Zealand Ltd (“Caltex”) three petrol retailers entered into an arrangement to abandon a practice of offering a free carwash to any person who purchased $20 or more of petrol.  The arrangement did not involve any agreement as to the prices to be charged for petrol or carwashes.  Nonetheless, the High Court found in Caltex that the arrangement had the effect of controlling the price of petrol because it restrained the free action of parties setting the price.³⁸  The Supreme Court in Lodge viewed the arrangement to discontinue the practice of absorbing the Trade Me listing costs to have clear similarities to the arrangement to discontinue the free carwash in Caltex.

The Supreme Court in Lodge then expanded upon its observations of the concept of freedom of action as follows:³⁹

As we see it, the fact that in any individual transaction an agency could decide to absorb the Trade Me listing does not mean there was complete freedom on the part of the agencies in relation to every transaction into which they entered.  On the contrary, agencies were restrained by the agreed position of adopting vendor funding as their default option.  The agencies could not simply defy the arrangement and adopt an agency funding model as their default position without failing to meet the agreed expectation that founded the arrangement.  If an agency decided to absorb the cost of the Trade Me listing as a default position or even on a regular basis, it would be cheating on the arrangement.  So, although allowed some freedom, the Hamilton agencies were not free to ignore the arrangement to adopt a vendor funding model as agreed.

Accordingly, the inquiry centres upon a comparative assessment of freedoms.  Without arrangements between competitors, rivals are left to set all aspects of price on a unilateral basis.  The emergence of arrangements between competitors relating to price creates a restraint on that freedom, amounting to the control of price for the purposes of section 30 of the Act.

Turning to the degree of restraint that is required it can, based on the Supreme Court’s decision in Lodge, be assumed that arrangements that involve commitments will likely be complied with in a material way.  While there may be some scope for freedom to depart from the arrangements, either because of acknowledged rights to do so or because of cheating, this does not create freedoms such that there is no underlying control of price.  It can be assumed that the commitments made under the arrangements will not be ignored.

A default position as to the start or offer price, or any other arrangements that may impact on the ultimate price, will be sufficient to establish control. Some freedom to deviate on the pricing arrangements does not provide a sound basis for concluding that the overall arrangements will not have the purpose or effect of controlling price.  There is an obvious logic to this approach.  Given that the underlying price fixing arrangements reached between competitors will not likely be ignored, the control of prices under these arrangements will impact on all consumers who are not beneficiaries of the occasions where there may be some deviation.  This is the very mischief that the deeming provisions of section 30 are designed to address.

V. The “de minimis” Component Exception

The analysis of collusion over a component of price involves some complexity.  Is control over a component alone enough to establish liability, or must liability be established in respect of the overall price?  Instinctively, where there is any price fixing between competitors (including the fixing of component parts), assumptions of potential liability may attach given the per se nature of the prohibition.  However, the law has a certain tolerance of conduct which may have only a de minimis impact.

As will be apparent from the above outline of facts, this issue arose directly in Lodge.  The impact of the vendor funding arrangement was that vendors would be required to pay the $159 (plus GST) listing fee for Trade Me advertising.  This fee needed to be assessed in two different settings.  If the sale process was successful, then this fee would be a small component of the price for the overall real estate services when considered alongside the average of commission of $15,000.  If, however, the sale process was unsuccessful then this amount would have constituted a significant part of the advertising costs which would still need to be met by the vendor.

This issue also arose in a recent Australian decision, ACCC v Olex Australia Pty Ltd (“Olex”).⁴⁰  In that case the alleged cartel related to suppliers of electrical cable agreeing to fixed cutting fees of $85 for each cut. These fees were a “modest component” of the overall price of electrical cable.⁴¹  Unlike in Lodge, this component was not a transparent separate fee to be paid by customers.  Customers bought by reference to the overall price/service package.

A comparative review of the reasoning and analysis in these two cases is informative.

In Lodge the component aspect of the price related only to the situation where the sale was successful.  In cases where there was no sale, the vendor was left to pay the Trade Me fee.  This was found to be all or a substantial part of amount payable to the agency.⁴²  Therefore, de minimis component analysis did not apply to this situation.

Turning to the analysis of the successful sale situation, where the Trade Me fee was a part of the overall payment made to the agencies, the Supreme Court rejected the approach taken by the Court of Appeal that “price” under section 30 extended as a matter of general principle to a component part of a price.  Rather, the position was taken that “the correct position is that price includes a component of the price unless that component is insignificant in competition terms”.⁴³  It followed, therefore, that the Supreme Court accepted that “there will be cases where the component of the overall price that is affected by the arrangement is so insignificant that it cannot have the effect of controlling the overall price, assuming that the overall price is otherwise determined by market forces”.⁴⁴  The Court added that: “As we see it, the correct position is that price includes a component of price unless that component is insignificant in competition terms”.⁴⁵

The Court addressed the competitive significance of this component fee in two ways.  Firstly, the Court addressed the competitive significance of this fee in the context of competition between the agencies for new listings.  The Court found this to be an important factor in the competition between agencies for new listings because if one agency had broken ranks and continued to absorb the Trade Me fee, then other agencies would have risked losing listings if they did not follow.⁴⁶  Therefore, this arrangement was of competitive significance because it was instrumental in the competitive process for new listings.

The Court also accepted that competitive significance attached to the cost to the agencies of absorbing the fee.  While in the case of individual sales the cost to agencies was $159 (plus GST), the additional annual cost to agencies of absorbing this fee would have been substantial, and  in the region of $200,000-$220,000 for Lodge.⁴⁷  This finding was coloured to some extent by inconsistencies that were seen to exist in the arguments raised by the appellant, namely that the new fee was so high that it was a natural reaction for the agencies not to absorb it given the cost, yet section 30 could not apply because it was such a small component of the overall price.⁴⁸  The competitive significance of this factor is less apparent.  On one view, it simply relates to the profitability of agencies rather than the competitive process.

The Court concluded:⁴⁹

We are satisfied, although the arrangement related to a mathematically small component of the overall charges by the Hamilton agencies to customers who successfully sold their properties, it was nevertheless a sufficiently significant competent of the overall price to bring the arrangement within the ambit of s 30.

The observations of the Federal Court in Olex on the issue of the alleged price fixing for cable cutting fees are dicta, because the Court found that there was no basis for this allegation.⁵⁰  Further, the decision in Olex addressed this point in a different context.  The cartel provisions of section 44ZZRD of the Competition and Consumer Act 2010 (Cth) did not apply because the cutting of the cable was not a “service” within the meaning of that Act.  Rather, the issue was addressed as an alternative pleading, namely that the cutting fee provision had the purpose and/or likely effect of controlling the price of electrical cable.⁵¹  Notwithstanding this, the analysis in Olex was considered by the Supreme Court in Lodge to be of relevance to the interpretation of New Zealand’s cartel provisions.  In particular, Olex was considered important for the following statement of principle:⁵²

Generally, more needs to be shown than merely that a provision has the likely effect of controlling a component of the price.  It must have the likely effect of controlling the overall price, ie be a materially significant proportion of the price.

On the facts the Court found no evidence of discussions between competitors of a commitment to exercise control over the price of cut cable.  The cutting fees were a modest component of the overall price of cable and “there was no commercially realistic ability to control the price of electrical cable by controlling the price charged for cutting services”.⁵³  It was found that customers bought by reference to the overall price and that this was dependent on market forces and competitive constraints.⁵⁴

Lodge and Olex are at one in holding that cartel conduct in respect of a component of an overall price does not establish liability without more.  Significance needs to attach to the component of price.  However, the two decisions expressed and applied this significance factor in a different way, perhaps because of the different facts before them.

The Olex formulation involves an inquiry only into the price materiality of the component, with reference to the overall price.  The focus is upon the ultimate price to the consumer.  Is there evidence that the component is of such proportionate significance as to impact on the overall price?  This is a narrow price-based inquiry, the resolution of which will depend on the facts of each case.  A consistent application of this test may not always be apparent.

In contrast, the Lodge formulation is potentially more expansive.  The test is expressed in wider terms and involves an inquiry into whether the component is insignificant in competition terms.  The manner in which the Supreme Court applied its competitive significance test in Lodge suggests that a wider inquiry than just a direct price-based test may come into play.  This is reflected in the inquiries into the competitive significance of absorbing the Trade Me fee in relation to the winning of new listings, and the cost burden upon the agencies of absorbing the fee.  The direct focus of these inquiries is upon the competitive materiality of the vendor funding arrangements to the agencies in their capacity as competitors.

Ultimately, analysis of the potential competitive impact of this fee on the agencies leads to the conclusion that the price to consumers will likely be impacted.  Where, as here, competitors stifle rivalry for new listings and pass on prices to maintain profits, it can be expected that the price to consumers will increase. However, the analysis in Lodge stops short of undertaking a price-based inquiry into the materiality of this outcome.  Adopting the Olex approach, the issue would be whether the Trade Me fee component of $159 (plus GST) was a materially significant proportion of the overall amount, which was $15,000 on average.  This question was not asked or answered in the Lodge decision.

It is suggested that the price-based test in Olex is to be preferred because of its focus on the materiality of the component to consumers impacted by the price fixing.  Section 30 is primarily concerned with impacts on consumers rather than, in this case, the suppliers of services in question.

VI. Concluding Comments

The Supreme Court’s decision in Lodge has settled a past uncertainty and opined on two new issues.  However, questions remain about the impact that this decision may have.

It is now settled that proof of a mere expectation is not enough to establish an “arrangement” under sections 27 and 30 of the Act.  More is required, namely the establishment of a “commitment” to the cartel by competitors (rather than a “moral obligation”).  However, questions remain as to the standard of proof that may apply to this new test.  There is the prospect that the “commitment” test may set a reasonably high bar.⁵⁵

Clarity has been provided on the issue whether an agreement between the cartelists to allow some discretion to depart from the cartel means that there is no control over price for the purposes of section 30.  Agreed positions between cartelists will likely be materially honoured, and some freedom to depart from these arrangements will not provide an exemption from the application of the price fixing laws.  A material impact of the cartel on consumers will remain.

A common position has emerged on both sides of the Tasman regarding whether control of a component of a price may contravene section 30 of the Act.  Such conduct does not contravene this section without more.  However, different approaches are discernible about what is required for there to be something more.  Lodge provides a wide formulation that requires that the component of the overall price must be competitively significant.  A narrower price-based test appears in Olex: would the component likely effect the controlling of the overall price, because of the materiality of the proportion of the component price.  Future jurisprudence will better inform upon this perceived difference in approach.  A focus upon the overall price to consumers, as occurred under Olex, is preferable.  Further consideration of the Lodge test leaves open the potential for convergence.

Click here for a PDF version of the article

1 Barrister, Auckland.  Comments and questions are welcome to mark.berry@mblaw.co.nz.

2 For an outline of New Zealand cartel case law developments up to 2017, see Chris Noonan, Competition Law in New Zealand, ch 6 and 1102-13.  For coverage of the current cartel provisions introduced under the Commerce (Cartels and Other Matters) Amendment Act 2017, see Gault on Commercial Law, CA30.01AA – CA30C.02.  References to section 30 in this paper now extend to new sections 30 and 30A of the Act.

3 [2020] NZSC 25.

4 The Court confined its analysis to the “arrangement” element of these provisions based upon the pleadings.  The Court noted that the same principles were also relevant to the “understanding” element of these provisions.  It further observed that the concept of “understanding” potentially has a less restrictive meaning than “arrangement”.  However, as this matter was not before the Court, the issue was not taken further: id, at [30].

5 This issue arose, for example, in the air cargo litigation in respect of understandings reached between competing airlines in respect of common fuel and security surcharges: see Commerce Commission v Air New Zealand Ltd [2013] NZHC 1414.

6 A case in point is the carwash case where petrol retailers pursued a co-ordinated strategy to discontinue the offer of free carwashes to purchasers of fuel for $20 or more: see Commerce Commission v Caltex New Zealand Ltd (1999) 9 TCLR 305.

7 Lodge, supra n 3, at [16].

8 Id, at [157].

9 As it happened, Trade Me’s new pricing policy did not survive.   The default position adopted by the agencies resulted in a decrease of listings which eventually caused Trade Me in 2014 to reintroduce a capped monthly fee of $999 for agencies in the regions and $1,399 for agencies in the metropolitan regions: id, at [23].

10 Id, at [60].

11 Id, at [109].

12 [2004] 1 NZLR 608.

13 Id, at [15] and [17].

14 See Matt Sumpter, New Zealand Competition Law and Policy, at 104-05, Noonan, supra n 2, at 301 and Paul Scott, “Going it Alone: Arrangements and Understandings Under the Commerce Act 1986” (2010) 16 NZBLQ 98, at 123-26.

15 Supra n 12, at [66] and [70].

16 Supra n 3, at [53] – [54].

17 Id, at [54].  The Court of Appeal in Lodge also expressed similar reservations.  It said: “We have concerns that if moral obligation is elevated to a stand-alone requirement, the analysis of arrangements would get bogged down in moral assessments, which are by their nature unpredictable, and in a commercial context, incapable of precise assessment”: Commerce Commission v Lodge Real Estate Ltd [2019] 2 NZLR 168, at [67].

18 Id, at [54].

19 Id, at [58].  The Court of Appeal in Lodge also described the test in this way and concluded that “there has to be an element of conditionality in an understanding, that is the parties recognise that they will commit to a course of future conduct on the basis that others are making the or a similar commitment and act in accordance with that commitment”: supra n 17, at [68].  This approach is also consistent with Australian case law which has taken the position that a mere expectation is not enough and that something more is required to be established in the sense that at least one party will “assume an obligation” or give an “assurance” or “undertaking” that it will act in a certain way: see, e.g., ACCC v CC (NSW) Pty Ltd [1999] FCA 954, at 141 and Rural Press Ltd v ACCC [2002] FACFC 213, at 79.

20 The Supreme Court in Lodge also noted that the assessment is objective, and assertions of subjective intentions will provide little assistance: id, at [50].

21 See, e.g., Black’s Law Dictionary (11^th ed, 2019), at 1293.

22 Australasian case law has routinely traced the origins of this test to the decision of Willmer LJ in British Basic Slag Ltd v Registrar of Restrictive Trading Agreements [1963] 1 WLR 727, at 739.

23 Supra n 21, at 340.

24 ACCC v Leahy Petroleum Pty Ltd [2007] FCA 794, at 26.

25 See, e.g., ACCC V Australian Egg Corp Ltd [2017] FCAFC 152, at 95 and Leahy Petroleum, id, at 35.

26 Other commentaries also are suggestive that the “commitment” requirement will raise enforcement obligations: see John Land, “Clarifying New Zealand Competition Law: Establishing ‘Arrangements’ Between Competitors, the ‘Controlling’ of Price, and Anti-Competitive ‘Purpose’ after Lodge” 25 NZBLQ 255, at 260 and Noonan, supra n 2, at 302.

27 Commerce Commission v Lodge Real Estate Ltd [2017] NZHC 1497.

28 Id, at [215].

29 Id, at [227].

30 Id, at [231] – [233].

31 Id, at [228] – [229].

32 Supra n 17, at [87] and [91].

33 Id, at [83] and [87].  Some reliance for this position was placed on the following decisions: Dole Food Company Inc v European Commission [2013] 4 CMLR 31, Balmoral Tanks Ltd v Competition and Markets Authority [2017] CAT 23 and Plymouth Dealers’ Assoc of Northern California v US 279 F 2d 128 (9^th Cir 1960).

34 Id, at [90].

35 Supra n 3, at [136].

36 Id, at [141].

37 Supra n 3, at [143] and [147].

38 Caltex, supra n 6, at 311.

39 Supra n 3, at [148].

40 [2017] FCA 222.

41 Id, at 655.

42 Supra n 3, at [157].

43 Id, at [156].

44 Id, at [155].

45 Id, at [156].

46 Id, at [158].

47 Id, at [159].

48 Id.

49 Id, at [161].

50 Supra n 40, at 651.

51 Id, at 655.

52 Id, at 657.

53 Id, at 655.

54 Id, at 655-58.

55 This observation stands in contrast with the High Court’s observation in the first phase of Lodge that the “bar is determinedly not set high” for the establishment of an “arrangement”.  However, this statement was made on the basis that only an expectation was required to be established: supra n 27, at [21] and [177].