1 The “by object” restriction in competition law is commonly understood as a restriction on conduct considered so inherently harmful to competition that the competition authority need not consider the effects of the conduct before concluding that the conduct was anti-competitive. In this matter, warehousers operating in the same complex discussed price increases implemented by the largest operator in that complex. It is undisputed that the warehousers who are parties to this case have a relatively small market share and low market power.

2 Whether this discussion between the warehousers can be, and should be, deemed a “by object” restriction of competition depends on when and how the “by object” restriction applies, and more specifically, the extent to which a competition authority must examine the economic context within which the alleged infringement occurred, before it can conclude that the conduct was, by its very nature, anti-competitive.

3 This is the first occasion on which the High Court has been called upon to consider an appeal under the Competition Act 2004 (“Competition Act”). The appeal arises from a finding by the Competition Appeal Board (“CAB”) that the Competition and Consumer Commission of Singapore (“CCS”) had not proved that an exchange of pricing information between a group of warehouse operators amounted to a restriction of competition “by object” under s 34 of the Competition Act. The CCS now challenges that finding as wrong in law. For the reasons which I set out below, I agree with the CCS, and allow the appeal.

4 Before turning to the facts and the issues, I set out the structure and respective roles of the three bodies whose decisions are engaged in this appeal: the CCS, the CAB, and this court. Understanding the roles and powers of these bodies and their respective limits within the broader landscape of competition law in Singapore is helpful to appreciate how the questions arising on appeal must be resolved.

5 The CCS is a statutory body established under the Competition Act to administer and enforce three key prohibitions in competition law: “Agreements, etc., preventing, restricting or distorting competition” under s 34 of the Competition Act; “Abuse of dominant position” under s 47 of the Competition Act; and “Mergers” which have resulted or may be expected to result in a substantially lessening of competition under s 54 of the Competition Act. In particular and relevant to this appeal, the prohibition under s 34 is further understood to fall into two broad categories: where the agreement, decision by associations or concerted practice has as its object the prevention, restriction or distortion of competition within Singapore (ie, “by object” limb), and where the agreement, decision by associations or concerted practice has as its effect the same (ie, “by effect” limb).

6 Vested with both investigative and adjudicative powers, the CCS may conduct investigations into suspected infringements under s 62 of the Competition Act. Upon completing such investigations, it may, under s 68 of the Competition Act, issue decisions finding infringements, direct infringing parties to cease the offending conduct, and impose financial penalties of up to 10% of the infringing undertaking’s turnover in Singapore for each year of infringement, for a maximum of three years, amongst other powers.

7 The CCS thus combines the roles of fact-finder, prosecutor, and first-instance decision-maker. As a counterbalance, the Competition Act provides for an independent appellate body in the CAB. The CAB is a specialist tribunal established under Part 4 of the Competition Act and constituted by persons appointed “on the basis of their ability and experience in industry, commerce or administration, or their professional qualifications or their suitability otherwise for appointment”: s 72(1) of the Competition Act. It has the power to conduct a full merits review. Under s 73(8) of the Competition Act, it may confirm or set aside the decision which is the subject of appeal. It also has the powers to find facts as it is endowed with the power to “give such direction, or take such other step, as the [CCS] could itself have given or taken” (s 73(8)(c) of the Competition Act), and “make any other decision which the [CCS] could itself have made” (s 73(8)(d) of the Competition Act). The CAB may therefore re-examine the evidence to reach its own factual conclusions and form its own view as to the economic analysis.

8 Under s 74 of the Competition Act, a further appeal lies from the CAB to the General Division of the High Court, and thereafter to the Court of Appeal under s 74(4). This appeal is only on points of law or the quantum of a financial penalty: s 74(1) of the Competition Act. This court’s jurisdiction accordingly does not extend to re-examining the CAB’s findings of fact or its economic assessments, but is confined to ensuring that CAB acted lawfully within the correct legal boundaries. Where the CAB has done so, its conclusions are not open to challenge in this court merely because this court might have weighed the evidence differently.

9 I set out the background facts to this appeal, as found by the CCS and to the extent accepted by the CAB. As I noted in the preceding paragraph, this court is limited to deciding on appeals on points of law, and has no power to review the CAB’s findings of fact.

10 The first respondent, CNL Logistics Solutions Pte Ltd (“CNL”) and second respondent, Gilmon Transportation & Warehousing Pte Ltd (“Gilmon”), provide warehousing services at Keppel Distripark, a multi-tenanted cargo distribution complex in Singapore. ¹ As warehouse operators, they provide storage and cargo handling services, including receiving shipments, temporary storage, unpacking of consolidated cargo, and arranging onward delivery to consignees. ²

11 Keppel Distripark is a free trade zone. ³ Duties and Goods and Services Tax are not levied on cargo stored within it – they fall due only when goods are brought out for local sale or consumption, or consumed within the zone itself. ⁴ At the material time, approximately 26 warehouse operators operated at Keppel Distripark. ⁵

12 On 15 June 2017 at approximately 4.41pm, Hup Soon Cheong Pte Ltd (“HSC”), the largest warehouse operator within Keppel Distripark, notified a customer that it would be imposing a new levy on import cargo stored within the free trade zone (“FTZ Surcharge”). ⁶ It posted the same notice at its warehouse office shortly thereafter. ⁷ HSC was the first warehouse operator to announce the FTZ Surcharge. ⁸ HSC’s sister company, Capital Logistics Services Pte Ltd (“CLS”), which was also one of the larger operators at Keppel Distripark, followed suit the same day. ⁹

13 Other warehouse operators followed in short order. ¹⁰ Among them were the respondents and two additional operators, Penanshin (PSA KD) Pte Ltd (“Penanshin”) and Mac-Nels (KD) Terminal Pte Ltd (“Mac-Nels”) (collectively, the “Undertakings”). ¹¹ The Undertakings held approximately 16.09% of the warehouse operator market at Keppel Distripark, ¹² with the respondents accounting for approximately 5.75% of that market. ¹³

14 On 17 November 2022, the CCS issued its Infringement Decision (“ID”) against the Undertakings. The CCS’s case proceeded on two alternative bases.

15 The CCS’s primary case was that a meeting took place at around 11.30am on 15 June 2017 between representatives of the respondents and Penanshin (“15 June 2017 Meeting”). ¹⁴ The CCS found on a balance of probabilities that at that meeting, CNL’s Director and General Manager, Vasu s/o Achutan (“Mr Vasu”) and Gilmon’s Managing Director, Teo Siang Siak (“Mr Teo”), met with Penanshin’s Container Freight Station Manager, Mohamed Yasrin bin Mohamed Yasil (“Mr Yasrin”). ¹⁵ The CCS found that Mr Vasu and Mr Teo told Mr Yasrin that certain warehouse operators, including the respondents, intended to adopt the FTZ Surcharge, and invited Penanshin to do likewise. ¹⁶ They also asked Mr Yasrin to ask if Mac-Nels would join in on adopting the FTZ Surcharge. ¹⁷

16 Alternatively, the CCS found that even if the 15 June 2017 Meeting did not take place, there had been an exchange of commercially sensitive pricing information between the Undertakings prior to their implementation of the FTZ Surcharge. ¹⁸ The CCS identified the following communications as establishing this (“Communications”):

(a) At 7.47pm on 15 June 2017, Thomas Chua, Gilmon’s Assistant General Manager (“Mr Chua”), told Mr Yasrin that Gilmon intended to impose the FTZ Surcharge. He then forwarded to Mr Yasrin, via WhatsApp, the FTZ Surcharge notices that HSC and CLS had issued. ¹⁹

(b) Mr Vasu called Mr Yasrin and asked him whether Penanshin was going to impose the FTZ Surcharge. ²⁰

(c) At 9.57pm on 15 June 2017, Mr Yasrin sent a WhatsApp message to Mac-Nels’ director, Matthew Er Yeong Yang, (“Mr Er”), in the following terms: ²¹

The following morning, at 7.28am on 16 June 2017, Mr Er replied: “Yes I will follow.” ²²

(d) Between 7.36am and 7.42am on 16 June 2017, Mr Yasrin forwarded to Mr Er, via WhatsApp, copies of the FTZ Surcharge notices issued by HSC and CLS. ²³

(e) At 7.40am and 7.42am on 16 June 2017, Mr Yasrin sent separate but identical WhatsApp messages to Mr Vasu and Mr Chua: ²⁴

(f) Mr Vasu replied “Thx bro”. Mr Chua replied “Thanks”. ²⁵

17 On the basis of the 15 June 2017 Meeting and the Communications, the CCS found that the Undertakings had engaged in “Price Fixing Conduct”: ²⁶ they had contacted competitors directly to disclose their future pricing intentions, solicited confirmation that those competitors would follow suit, and received that confirmation. The CCS found on a balance of probabilities that this amounted to an agreement and/or concerted practice to coordinate the imposition of the FTZ Surcharge, with the object of preventing, restricting or distorting competition in the market for warehousing services at Keppel Distripark, in breach of s 34 of the Competition Act. ²⁷

18 The CCS investigated 11 warehouse operators at Keppel Distripark in connection with the FTZ Surcharge. Of these, only the Undertakings were found to have engaged in infringing conduct. ²⁸ The CCS stated in the ID that it did not find evidence of the other operators’ participation in the “Price Fixing Conduct”. ²⁹

19 The respondents appealed to the CAB against the Infringement Decision. The CAB allowed the appeal in full. Its decision rested on three findings, the first two of which the CCS does not challenge on this appeal.

20 First, the CAB found that the CCS had not proven on a balance of probabilities that the 15 June 2017 Meeting took place. ³⁰ The direct evidence amounted to nothing more than a “he said, she said” situation. ³¹ The CCS’s case rested primarily on Penanshin’s Leniency Statement and Mr Yasrin’s subsequent interviews. The CAB found that these were undermined by three considerations:

(a) Mr Yasrin’s earliest interview on 19 November 2019 – given before Penanshin submitted its leniency application – made no mention of any meeting, and identified Mr Chua rather than Mr Teo as the Gilmon representative who had communicated with him. ³²

(b) The discrepancies in Mr Yasrin’s account were never put to him for clarification in later interviews, and his testimony was never tested by cross-examination. ³³

(c) Penanshin had an economic incentive, as a leniency applicant, to shift responsibility for initiating the conduct onto the respondents, since an initiator of a cartel is ineligible for full immunity. ³⁴

21 The CAB further noted that Mr Yasrin’s own admission that he had “called the other parties … to get their notices” was more consistent with him having been the initiator of contact than with him having been approached at a meeting by the respondents. ³⁵ The supporting evidence relied upon by the CCS – Mr Yasrin’s WhatsApp messages to Mr Er, Mr Vasu, and Mr Chua, set out above at [16] – did not resolve this ambiguity in the CCS’s favour. ³⁶ The word “follow” in Mr Yasrin’s message could mean following the general body of Keppel Distripark operators who had announced the surcharge after HSC’s public announcement or following a compact proposed by the respondents at the 15 June 2017 Meeting. ³⁷

22 Second, the CAB found that on CCS’s alternative case, the Communications also did not constitute an agreement. The CCS’s alternative case was also predicated on the 15 June 2017 Meeting having taken place. Having found that the meeting was not proven, and there being no other evidence of any offer from the respondents capable of founding a consensus, the case for an “agreement” fell away entirely. ³⁸

23 Third, the CAB held that while the Communications constituted a concerted practice, ³⁹ the CCS had not proven that this concerted practice amounts to a “by object” infringement of s 34 of the Competition Act. This is the finding CCS now challenges on this appeal.

24 On this third point, the CAB first rejected the CCS’s characterisation of the Communications as “Price Fixing Conduct”. ⁴⁰ The CAB found that the CCS’s definition of price fixing – encompassing any conduct that “facilitate[s] the coordination of future pricing conduct between competitors” – was too broad. ⁴¹ Price fixing requires an agreement between competitors to fix prices, which, in the CAB’s view, would be distinct from information sharing. ⁴² To conflate them would be to import, without justification, the presumption of harm that attaches to price fixing into every case involving the disclosure of price-related information between competitors. ⁴³

25 The CAB then determined the legal test for “by object” infringements. Conduct falls within the “by object” limb, only where it is “regarded, by their very nature, as being harmful to the proper functioning of normal competition” or where “such coordination reveals in itself a sufficient degree of harm to competition”. ⁴⁴ This test must be applied restrictively, otherwise, the competition authority would, in effect, be exempted from the obligation to prove the actual effects on the market. ⁴⁵ For information-sharing conduct specifically, the CAB held that the surrounding market context – including market structure, the number and size of market participants, the nature of the information exchanged, and the economic conditions prevailing in the relevant market – must be adequately examined before a conclusion can be reached that the conduct is so obviously injurious to competition as to warrant treatment as a “by object” infringement. ⁴⁶

26 The CAB found that the CCS had failed to conduct that examination. The CCS had dismissed the respondents’ small market shares as irrelevant on the basis that price fixing always has an appreciable adverse effect on competition regardless of market share. ⁴⁷ But, in the CAB’s view, that reasoning was available only once the conduct had been properly characterised as price fixing – which it was not. ⁴⁸ The respondents’ conduct was an exchange of pricing information, hence the CCS was required to engage with the market context. ⁴⁹ The CAB noted that such observations about market structure as appeared in the ID were scattered across sections dealing with other issues and did not constitute a coherent analysis capable of grounding a finding of a “by object” restriction of competition. ⁵⁰ Against the backdrop of a market with approximately 26 operators dominated by HSC and CLS, in which the respondents together held only 5.75%, the CCS had offered no adequate explanation of why the Communications were, by their nature, so likely to harm competition as to dispense with any examination of effects. ⁵¹ The CAB accordingly allowed the appeal and set aside the ID.

27 The CCS now appeals to this court. I briefly outline the parties’ arguments here. Where necessary, I will examine them in greater detail in the course of my analysis below.

28 The CCS’s position is that the Communications constituted a concerted practice with the object of restricting competition, in violation of s 34 of the Competition Act. The CCS argues that the CAB erred in setting the legal threshold for “by object” infringements too high, and that the CAB imposed a requirement for a more searching examination of the economic context than the law demands. In doing so, the CAB had, in the CCS’s view, effectively imported an effects-based analysis into what should have been a straightforward “by object” inquiry. ⁵²

29 In brief, the CCS makes four submissions:

(a) The CAB was wrong to treat the “by object” category as confined to conduct that is “well-established, egregious and obviously harmful” ⁵³ – the test asks only whether the conduct reveals a sufficient degree of harm to competition, ⁵⁴ and the lack of precedent for finding the exchange of pricing information as a “by object” infringement should not be a bar. ⁵⁵

(b) The CAB drew an impermissible distinction between information sharing and price fixing that has no basis in law – the overarching legal test applies uniformly to all forms of coordination. ⁵⁶

(c) The examination of economic context is no more than a “basic reality check”, and where the anti-competitive object is easy to perceive, this examination should be limited to “what is strictly necessary to confirm or cast doubt on the harmful nature and anti-competitive object revealed by the analysis of the content and objectives of the practice in question”. ⁵⁷

(d) On the facts, the CCS conducted precisely that examination in the ID – identifying the price-sensitivity of the market, the risk of customer switching, and the way the Communications reduced each Undertaking’s uncertainty – and the economic context disclosed nothing that precluded a “by object” finding. ⁵⁸

30 The respondents’ case is that the CAB identified the correct legal test, applied it correctly, and reached a conclusion well within the range available to a properly instructed specialist tribunal. The respondents argue that the CCS has not identified any error of law. In the respondents’ view, the CCS is disagreeing with the CAB’s factual and evaluative conclusions – conclusions this court has no jurisdiction to re-examine.

31 The respondents make three submissions:

(a) There is no genuine legal controversy – both parties and the CAB accept that the same seminal case applies, and a disagreement about how an agreed test was applied to the facts is not a point of law. ⁵⁹

(b) The CCS has misconstrued the CAB’s decision in order to manufacture apparent points of law – the CAB did not treat the “by object” category as closed, ⁶⁰ did not apply a different legal test to information sharing, ⁶¹ and did not require a full effects analysis. ⁶² What it required was an adequate examination of economic context, which the agreed-upon precedent mandates, and which the CCS failed to conduct. ⁶³

(c) On the facts, the Communications were a one-off exchange of information between small operators holding a combined market share of 5.75%, following a public price announcement by the two largest players in the market ⁶⁴ – conduct that cannot, without more, be characterised as “so obviously injurious to competition” as to dispense with any examination of effects. ⁶⁵

32 Before identifying the substantive issues on appeal, I address the respondents’ argument that this appeal by the CCS does not raise a question of law such that this court’s jurisdiction under s 74 of the Competition Act is not properly invoked. I am satisfied that it does.

33 In Ng Eng Ghee v Mamata Kapildev Dave [2009] 3 SLR(R) 109 (“Ng Eng Ghee”) at [102], the Court of Appeal held that “[t]he question of what a statutory term means is clearly an issue of law and a misinterpretation of the term would constitute an ex facie error of law open to appeal.” The CCS’s challenge centres on the interpretation of the statutory terms “by object” and “concerted practice” – specifically, whether the CAB applied the correct legal test when assessing whether the Communications constituted a concerted practice which had as its object the prevention, restriction or distortion of competition. This is a question of law.

34 This question is not resolved, as the respondents contend, merely because both parties cite the same authority. ⁶⁶ Parties may agree on the applicable precedents while disagreeing fundamentally on what those precedents require. To that end, I note that both parties invoked Ng Eng Ghee in support of competing interpretations of what constitutes a question of law. It is unnecessary to resolve that broader debate here, as the present case clearly raises questions of law even on the narrowest interpretation of what constitutes a question of law.

35 The appellant put several questions before me as part of their appeal. I identify the following issues for determination:

(a) The applicable law on concerted practices, and whether the CAB was correct in finding that the Communications constituted a concerted practice;

(b) The applicable law on “by object” restrictions of competition under s 34 of the Competition Act, and the role of economic context in determining “by object” restrictions of competition; and

(c) Whether the CAB was correct in finding that the conduct did not amount to a “by object” restriction, including whether the CAB was correct in holding that the CCS had paid insufficient regard to the relevant economic context.

36 Section 34(1) of the Competition Act provides:

37 Section 34(2)(a) provides that for the purposes of s 34(1) of the Competition Act, agreements, decisions or concerted practices may, in particular, have the object or effect of preventing, restricting or distorting competition if they “directly or indirectly fix purchase or selling prices or any other trading conditions”.

38 Additionally, in interpreting the prohibition set out in s 34(1) of the Competition Act, United Kingdom (“UK”) and European Union (“EU”) law is persuasive. It was held in Pang’s Motor Trading v Competition Commission of Singapore [2014] SGCAB 1 at [33] that this is because s 34 of the Competition Act was modelled on the relevant UK and EU instruments (which I set out below). This is confirmed by the second reading of the Competition Bill: Singapore Parliamentary Debates, Official Report (19 October 2004) vol 78; at col 863–864 (Dr Vivian Balakrishnan, Senior Minister of State for Trade and Industry).

39 The parties (and the CAB) all purport to follow the approach in the EU and UK. I therefore find it useful to consider the relevant foreign instruments, and survey the voluminous case law on the subject. This is consistent with, as I have stated above, the value of EU and UK jurisprudence on this subject. Where relevant, I shall also consider the Opinions of the Advocate Generals in the EU cases, as these often inform and guide the courts’ reasoning. Not all the cases considered in this judgment were cited by the parties in their submissions, whether before me or before the CAB. I undertake this survey to provide the full context behind the principles to be expounded in this judgment.

40  Thus, insofar as the statutory language does not deviate from their EU and UK origins, I draw upon such foreign authorities in the analysis that follows.

41 As I have stated above at [20]–[21], the CAB found as fact that there was no agreement between the undertakings, and no appeal lies against this finding of fact.

42 Whether there was a concerted practice (as the conduct in this case was characterised by both the CCS and CAB) is an anterior question to whether parties’ conduct constitutes a “by object” restriction of competition. The appellant’s submission is that the CAB correctly held that the Communications constituted a concerted practice. The respondents submit that the CAB did not find that there was a concerted practice. Rather, the respondents’ position is that the CAB “merely assumed the hypothesis” that there was a concerted practice, and that since the CAB found that there was no “by object” restriction, it did not make a finding on whether there was a concerted practice. ⁶⁷

43 The respondents’ submission is inaccurate. The CAB considered the issue of whether there was a “by object” restriction after finding that the conduct constituted a concerted practice. I do not see any indication, explicit or implicit, that the entire second stage of the CAB’s analysis was purely hypothetical. For completeness, I reproduce [79] of the Decision below:

44 Nonetheless, insofar as the respondents are challenging a finding that there was a concerted practice, I consider the applicable law on concerted practices, and whether the CAB properly directed itself in this regard.

45 The CAB defined concerted practices as situations, other than agreements, where undertakings “knowingly substitute the risks of competition with some form of practical cooperation”. ⁶⁸ The CAB set out the test for finding a concerted practice as follows: ⁶⁹

46 The CAB went on to reject the CCS’s submission that a concerted practice could be found simply because competitors had communicated with each other. Such communications, whether direct, indirect, unilateral, or bilateral, were necessary but not sufficient. In the CAB’s view, there must have been subsequent market conduct causally connected to the undertakings’ prior communications with each other. ⁷⁰

47 I shall consider the applicable case law below. I find that the CAB correctly directed itself on the law, and as such, I see no reason to disturb its finding that there was a concerted practice in the present matter.

48 The definition of a concerted practice as it is understood in EU jurisprudence is set out in Coöperatieve Vereniging “Suiker Unie” UA v European Commission, Joined Cases C-40/73 to C-48/73, C-50/73, C-54/73 to C-56/73, C-111/73, C-113/73 and C-114/73, ECLI:EU:C:1975:174 (“Suiker Unie”) at [26] (cited in the Decision at [30]):

49 Gerhard Züchner v Bayerische Vereinsbank AG, Case 172/80, ECLI:EU:C:1981:178 (“Gerhard Züchner”) concerned a general service charge imposed by most European banks. The Court of Justice of the European Communities adopted the definition and approach from Suiker Unie. A finding of a concerted practice did not require an actual plan, so long as there was cooperation that affected or otherwise undermined the requirement that undertakings act independently. While undertakings could adapt to each other’s actions (present or past), “direct or indirect contract” (in the sense of undertakings working in concert) that affected their independence would be prohibited. As for the service charge at issue, although it could be justified based on the costs incurred by the banks in making international transfers, there could nonetheless be a concerted practice: Gerhard Züchner at [16]–[17]. This would be the finding if there were the requisite cooperation and coordination between undertakings to impose the charge: Gerhard Züchner at [22]. I take from Gerhard Züchner the point that conduct can amount to a concerted practice even if there is a potential alternative justification for the conduct.

50 I turn next to A. Ahlström Osakeyhtiö v European Commission, Joined Cases C-89/85, C-104/85, C-114/85, C-116/85, C-117/85 and C-125/85 to C-129/85, ECLI:EU:C:1993:120 (“A. Ahlström Osakeyhtiö”). The case involved undertakings and trade associations involved in the manufacture of wood pulp making quarterly pricing announcements in the trade press. The Court of Justice found that even though the pricing announcements were made very close in time to each other, this was not due to a concerted practice. Instead, the structure of the market resulted in a high degree of transparency. For example, buyers were always in contact with multiple sellers, and would inform sellers of the prices announced by other sellers. There were also relatively few buyers, and the buyers would exchange information about the quotes they received. Several sellers also bought materials from each other, and consequently would have been aware of each other’s pricing: A. Ahlström Osakeyhtiö at [81]–[88].

51 Furthermore, announcements did not reduce uncertainty as to competitors’ future pricing or attitude, as at the time of the announcements, each undertaking would not know of what the future pricing would be: A. Ahlström Osakeyhtiö at [64].

52 In relation to parallelism in the prices announced in A. Ahlström Osakeyhtiö, it was held that the existence of a concerted practice was not the only plausible reason for such conduct. It is worth setting out [126] of A. Ahlström Osakeyhtiö in full:

53 I take from this paragraph the point that parallelism is not itself evidence of a concerted practice. If there is a plausible alternative explanation, then there should not automatically be a finding of a concerted practice.

54 A. Ahlström Osakeyhtiö may be compared with John Deere Ltd v European Commission, Case C-7/95 P, ECLI:EU:C:1998:256 (“John Deere”). In the latter case, tractor manufacturers and importers in the UK cooperated on an information system that related to sales data. The Court of Justice found that the exchange of information constituted a concerted practice, as it allowed undertakings to know about their competitors’ market positions and strategies, thereby reducing uncertainty as to each other’s conduct. In coming to this conclusion, the Court of Justice had regard to the nature of the information exchanged, frequency of the exchange, and to whom it was disclosed: John Deere at [88]–[89]. A. Ahlström Osakeyhtiö was distinguished in that case on the basis that the price announcements in A. Ahlström Osakeyhtiö were to consumers, whereas the information exchange in John Deere was only among sellers.

55 Anic Partecipazioni v European Commission, Case C-49/92 P, EU:C:1999:356 (“Anic”) was referenced in the Decision, ⁷¹ as well as by the appellant. ⁷² The first part of Anic relevant to the present appeal is the pronouncement that the reference to “undertakings, decisions… or concerted practices” is intended to cover all forms of collusion. This is to ensure that all forms of collusion are prohibited: Anic at [108]. Furthermore, even if the agreements, decisions, and concerted practices are meant to catch different forms of collusion, the same course of conduct that constitutes a “by object” restriction can have elements of both an agreement and concerted practice: Anic at [112]–[114]. There would be no strict need to categorise conduct as either agreements or concerted practices. The competition authority would be entitled to characterise certain forms of conduct as “principally” agreements and others concerted practices: Anic at [132].

56 The other relevant pronouncement in Anic is the presumption of causal connection. This was not the only presumption discussed in Anic. Anic also set out a presumption that where an undertaking attends a price-fixing meeting, it is presumed that it participated in the cartel, unless the undertaking can adduce contrary evidence. In light of the CAB’s finding that no such meeting took place in the present matter, that presumption is irrelevant. However, I refer to the relevant presumption as the “presumption of causal connection” for clarity.

57 The presumption of causal connection is to the effect that where undertakings take part in a prohibited concerted practice and remain on the market, it is presumed that they had taken account of the information. This is a rebuttable presumption, where the undertakings investigated may provide proof that they did not in fact take into account the information exchanged: Anic at [121]. The CAB referred directly to this presumption, and the relevant parts of Anic. ⁷³

58 In Tate & Lyle plc, British Sugar plc and Napier Brown & Co. Ltd v European Commission, Joined Cases T-202/98, T-204/98 and T-207/98, ECLI:EU:T:2001:185 (“Tate & Lyle”, upheld on appeal in ECLI:EU:C:2004:255), the Court of First Instance dealt with a dominant undertaking in the British sugar market ending a price war, and informing its competitors of its future pricing intentions: Tate & Lyle at [3]–[11]. The Court of First Instance held that a concerted practice can be found even if information or intentions are shared unilaterally and without reciprocity. The mere receipt of information could constitute a concerted practice: Tate & Lyle at [54]–[58].

59 It was also held in Tate & Lyle that the fact that future prices had already been communicated to customers did not necessarily mean there could not be a concerted practice. In that case, the dominant undertaking had informed customers on an individual basis of its prices. That did not mean the prices were readily-accessible and objective market data (which, if shared, would likely not constitute a concerted practice). The meetings took place before the prices were announced, and led to a “climate of mutual certainty as to their future pricing policies”: Tate & Lyle at [60]. The Court of First Instance also rejected one undertaking’s argument that there was no anti-competitive intention, based on its position as both a customer as well as a competitor of the other undertakings. The rejection was because the undertaking in question did not make its lack of anti-competitive intention clear to the other undertakings, and because it still received confidential information of its competitor’s pricing intentions which “each participant could not fail to take into account… in order to determine the market policy which it intended to pursue”: Tate & Lyle at [62]–[67].

60 I consider next JJB Sports plc and Allsports Limited v Office of Fair Trading [2004] CAT 17 (“JJB Sports”). This was a case cited in the ID, but not by the appellant in its submissions. In JJB Sports, the Competition Appeal Tribunal cited Suiker Unie and Tate & Lyle, among other cases, as setting out the relevant law on concerted practices. There are three points worth noting from JJB Sports:

(a) First, the Competition Appeal Tribunal accepted that it is possible for multiple undertakings to be sanctioned for taking part in a concerted practice, even if parties only ever take part in bilateral communications. In such circumstances, it would not be necessary to decide there was a single concerted practice, or multiple concerted practices. I take it that such findings are necessarily fact-sensitive. The example given in JJB Sports at [643] was as follows:

(b) Second, the Competition Appeal Tribunal cited Anic for the proposition that a competition authority need not characterise an infringement as either an agreement or a concerted practice. All that is needed is for the conduct in question to constitute one or the other: JJB Sports at [644].

(c) Third, the Competition Appeal Tribunal stated that where an undertaking attended a private meeting where pricing intentions were discussed, there would be a concerted practice among the attendees even if information was only shared unilaterally by one undertaking: JJB Sports at [873].

61 Anic was cited in T-Mobile Netherlands BV v Raad van bestuur van de Nederlandse Mededingingsautoriteit, Case C-8/08, ECLI:EU:C:2009:343 (“T-Mobile”) for the proposition that the listing of agreements, decisions, and concerted practices in the Treaty of the Functioning of the European Union (“TFEU”) was intended to catch all forms of collusion. These categories of conduct were distinguishable not in kind, but “only by their intensity and the forms in which they manifest themselves”: T-Mobile at [23]. This meant that the criteria for finding a “by object” restriction is the same regardless of how the conduct is characterised. Whether conduct amounted to a concerted practice would be determined in light of the idea that undertakings must determine their strategies independently: T-Mobile at [32].

62 The above does not mean that undertakings must build walls round themselves. There is a distinction between the right of undertakings to adapt themselves to their competitors’ existing or anticipated conduct, which is permitted, and direct and indirect conduct which may influence conduct or disclose intentions, which is prohibited: T-Mobile at [33]. Furthermore, where there is such prohibited contact, the presumption of causal connection would apply even if there was only a single meeting or interaction.

63 I consider next Balmoral Tanks Ltd v Competition and Markets Authority [2017] CAT 23 (“Balmoral Tanks”) which also extensively discusses the legal principles on “by object” restrictions. It was cited by the appellant in support of the proposition that a single instance of contact can amount to a concerted practice. That is not controversial. However, Balmoral Tanks contains three other relevant points on concerted practices:

(a) First, in determining whether a concerted practice constituted a “by object” restriction, there is no need to consider the effects of the concerted practice. However, there is a need to show that information exchanged had an effect on parties. In other words, it must be shown (in the context of the presumption of causal connection) that the undertakings concerned conducted themselves on the basis of the information exchanged: Balmoral Tanks at [119].

(b) Second, the basis for sanctioning and prohibiting concerted practices is that competitors should not, as a general rule, know how each other will behave. This is why conduct that causes a reduction in uncertainty will be deemed a concerted practice so long as there is some reciprocity. A similar point was made in Lexon (UK) Limited v Competition and Markets Authority [2021] CAT 5 (“Lexon”) at [162], that even if the information exchanged is publicly available, there can still be a concerted practice if it provides “valuable reassurance” or the like. An example would be an exchange of information accompanied by opinions that would not normally be made available to the public.

(c) Third, that the strictness of this area of the law (in particular the presumption of causal connection) is because “it is hard to think of any legitimate reason why competitors should sit together and discuss prices at all”: Balmoral Tanks at [41]. This is consistent with the reference in A. Ahlström Osakeyhtiö to whether there is a plausible alternative explanation for the conduct.

64 It is also worth noting that the Competition Appeal Tribunal in Balmoral Tanks considered whether the market was one where a single exchange of information could constitute a “by object” restriction. It was held that the meeting in question gave an indication that a new competitor was not intending to push prices down, and because all the other suppliers were used to fixing prices and exchanging information. This meant that a one-off exchange of information could constitute a “by object” restriction in that particular market: Balmoral Tanks at [101]–[106]. While Balmoral Tanks indicates that a single instance of contact can result in a concerted practice and a “by object” restriction, it did not hold that a single instance of contact will necessarily amount to such.

65 I finally consider Phones 4U Ltd (in administration) v EE Ltd (Competition and Markets Authority, intervening) [2025] EWCA Civ 869 (“Phones 4U”). I find the case very helpful in setting out the applicable principles on concerted practices, and the presumption of causal connection from Anic.

66 The English Court of Appeal accepted that the provision of sensitive business information, such as an exchange of future price increases, could constitute a concerted practice and a “by object” restriction. However, the concept of concerted practice implies some subsequent market conduct, as well as reciprocity. Referring to Eturas UAB v Lietuvos Respublikos konkurencijos taryba, Case C-74/14, ECLI:EU:C:2016:42 (“Eturas”) and the Opinion of Advocate General Szpunar in Eturas, ECLI:EU:C:2015:493, the English Court of Appeal also stated that what constitutes reciprocity or tacit approval would naturally depend on the circumstances. A lack of opposition in some cases of illicit communication will constitute reciprocity, because the context may be such that silence will be appreciated as approval. In particular, the unilateral disclosure (and concomitant receipt) of information can be a concerted practice if one party requests or accepts it: Phones 4U at [112], citing Cimenteries CBR v European Commission, Joined Cases T-25/95 etc, ECLI:EU:T:2000:77.

67 The above would not necessarily or wholly apply to cases of unanticipated “one way” communication of confidential information: Phones 4U at [145]. However, it was noted that the recipient in such cases would face evidential difficulties in showing that their silence was not tacit approval (since they would have likely benefitted from reduced uncertainty), due to the presumption of causal connection. In relation to the presumption of causal connection, the English Court of Appeal clarified that the presumption is not rebuttable only by public distancing (referring to Eturas).

68 Phones 4U concerned the decision of several mobile network operators (“MNOs”) in the UK to stop supplying consumer connections to mobile networks through indirect retailers (in this case, Phones 4U Limited (“Phones 4U”)). Instead, they would focus on selling to consumers directly: Phones 4U at [6]–[12]. The context of the cessation of supply was that the MNOs had a “strategic desire to increase direct distribution and to reduce their reliance on indirect channels”: Phones 4U at [21]. Another indirect retailer had also undergone a merger, making it more attractive to enter exclusive arrangements with them instead of with Phones 4U. It was also found that MNOs had a relative lack of insight into each other’s plans: Phones 4U at [22].

69 It was alleged that the MNOs had decided to stop supplying to Phones 4U due to a concerted practice. The alleged concerted practice took the form of, among other things, the chief executive officers of O2 and EE meeting for lunch wherein the former informed the latter of his company’s intention to reduce supply to indirect retailers, and certain discussions between senior executives of various MNOs: Phones 4U at [13] and [42]–[51].

70 In the light of the propositions above, the English Court of Appeal upheld the English High Court’s decision that there was no concerted practice in that case. In relation to the lunch meeting, the information given was (a) provisional and (b) was limited and vague. This was sufficient evidence that the information was not taken into consideration, thus rebutting the presumption of causal connection: Phones 4U at [176]. Another alleged instance of a concerted practice was found not to be so, as the decision to stop supplying to Phones 4U was made several months later, “after significant developments in the UK market and extensive and detailed negotiation between Vodafone UK and [Phones 4U]”: Phones 4U at [269].

71 From the case law considered above, it is reasonably clear that the definition of concerted practices has been accepted to be that as set out in Suiker Unie. Gerhard Züchner further clarifies that conduct can amount to a concerted practice even if there is a legitimate alternative explanation, such as costs. There may also be a concerted practice even if there is only a single instance of contact or communication.

72 This does not mean the CAB was wrong to conclude in the Decision at [72] onwards that a one-off communication cannot itself constitute a concerted practice. Read in context, the thrust of the Decision is not that a single instance of communication can never be the basis of a concerted practice. The CAB directly referenced T-Mobile for the proposition that a single meeting can be the basis of a concerted practice. ⁷⁴ The point of that section of the Decision is that the wrong is not in the communication itself. That is correct, and supported by the cases I have considered above, particularly Phones 4U. Contact or communication simpliciter will not necessarily mean there has been a concerted practice. Rather, the information exchanged, and the manner in which this is done, must be of the sort that would reduce uncertainty between competitors, and must include some element of reciprocity.

73 Furthermore, a reduction of uncertainty is a necessary but not sufficient requirement for finding a concerted practice. The existence of a plausible alternative explanation (that the conduct is not anti-competitive) will militate against a finding that there was a concerted practice. This may explain why the price announcements in A. Ahlström Osakeyhtiö did not constitute a concerted practice. Even if there was reduced uncertainty, the announcements also gave transparency to customers, which meant it could not be described as anti-competitive.

74 However, the sin is not in the fact of communication or receipt thereof. Concerted practices indicate a need for some reciprocity. Though it does not arise for my decision in this appeal, I find it unlikely that, for example, a business that receives information from a competitor without asking for it would necessarily or automatically be deemed as having participated in a concerted practice simply because it remained on the market. As seen from Phones 4U, there must still be some reciprocity or tacit approval (though such reciprocity or tacit approval may be presumed through silence).

75 In the course of the Decision, the CAB appeared to disagree with the ID on two other points:

(a) First, the CAB was of the view that the ID did not “clearly articulate the definitional boundaries” between agreements and concerted practices. ⁷⁵

(b) Second, the CAB made no explicit finding on whether there could be a concerted practice to fix pricing (as opposed to an agreement to fix prices).

76 The appellant appears to disagree in both respects, submitting that “[a] bright line distinction between ‘agreements’ and ‘concerted practices’ cannot be drawn”, ⁷⁶ and that it is possible to find a concerted practice to fix prices. ⁷⁷ Though they are not directly relevant to the disposal of the present appeal, I consider them for completeness, and because they are worth clarifying for future investigations by the CCS.

77 In relation to the first point, I agree with the appellant that there is no strict need to characterise conduct as either an agreement or a concerted practice. This has been the approach in cases such as Anic, JJB Sports, and T-Mobile. Agreements and concerted practices are different by degree, rather than by type. Once conduct falls into any one of these categories, the legal tests for whether there is a “by object” or “by effect” restriction of competition are the same, applying equal to all three categories of conduct.

78 A competition authority may therefore label conduct however it wishes, so long as it can prove that the requirements for the relevant category have been met. In turn, the undertakings investigated are entitled to adduce evidence or submissions to the effect that the competition authority has not proved that the conduct in question featured an agreement, decision, or concerted practice, as the case may be. This is a question of ensuring that the undertaking knows the case it has to answer, and can provide evidence to challenge the authority’s case.

79 In relation to the second point, the appellant correctly acknowledges that the Decision did not contain an explicit rejection of the possibility of a concerted practice to fix prices. However, the appellant challenges the Decision insofar as it contains an implied rejection of such a concerted practice. ⁷⁸ As I understand it, the respondents do not challenge this finding. As I have received submissions on this point, and the appellant has asked for doctrinal clarity, I discuss it briefly.

80 In short, it is very possible for a concerted practice to have an anti-competitive object if it will “directly or indirectly fix purchase or selling prices or any other trading conditions”: T-Mobile at [37].

81 I do not think that the CAB held that a concerted practice to fix prices could never exist. All that the Decision stated was that the definition of “price fixing” that the CCS proposed to the CAB was too wide.

82 As the respondents correctly submit, ⁷⁹ the CAB acknowledged that sharing price-related information can be as serious as price fixing (in the colloquial understanding of the term). For its part, the appellant has stated that price fixing has no immutable definition, ⁸⁰ and that the analysis is fact-specific, considering the harm in the individual case. ⁸¹

83 It seems to me, therefore, that all involved have approached the matter through the lens of analysing the seriousness of the conduct, and the harm caused. I do not think that the particular label (ie, whether the end of the concerted practice is to fix prices, to exchange information, or something else) is determinative. Categorisation can be useful, such as for determining in subsequent cases whether there is sufficient experience such that conduct can be deemed a “by object” restriction, and whether a particular case is analogous to previous ones. However, when analysing the particular case, the competition authority must ensure that it is not adopting the formalistic approach warned-against in the cases discussed in the next section of this judgment. It is not so simple as categorising conduct (whether an agreement, concerted practice, or the like) as price fixing and going on to determine the appropriate sanction. There is still a need to consider if the conduct in question, however characterised, is sufficiently serious (as I explain in the next section).

84 Thus, a concerted practice may be constituted by conduct that is not price fixing in the classic sense but is related or analogous to it. I do not think the respondents are challenging this, nor that the CAB had shut the door to this.

85 As I have stated above at [43], the CAB held that there was a concerted practice, though it did not state this clearly. I see no reason to depart from the CAB’s holding on this point.

86 The CAB held that the Communications were the basis for the concerted practice in the present appeal. This was characterised in the ID, ⁸² and discussed in the Decision, ⁸³ as a single or one-off communication. The appellant’s and respondents’ submissions also considered the possibility of single instances of contact constituting a concerted practice (and in turn “by object” restrictions). ⁸⁴

87 There is no real controversy that a single instance of contact or communication may amount to a concerted practice. Balmoral Tanks and T-Mobile both involved information exchanges on a single occasion. Both of these resulted in a finding that there was a concerted practice, constituting a “by object” restriction of competition.

88 Given that a single instance of information exchange can amount to a concerted practice, the analysis then focuses on whether the information exchanged is of the sort that would reduce uncertainty among undertakings, or result in practical cooperation. The uncertainty can go either to the fact of price movements, or to the details as to its implementation.

89 The Decision correctly identified that the information exchanged in the present appeal resulted in practical cooperation between the undertakings investigated. The Communications (as defined in the ID) related to various undertakings’ intention to raise prices. As stated in the ID, ⁸⁵ and acknowledged in the Decision, ⁸⁶ the respondents then used this to convince customers to agree to the implementation of the FTZ surcharge.

90 Contrary to the appellant’s submission, ⁸⁷ the CAB was correct to say that such conduct would constitute a concerted practice only to the extent that it can be shown that undertakings had taken into account the information exchanged. As I have stated above, the fact of communication is not in itself prohibited. In an information exchange, it is the reciprocity or subsequent use of the information that gives rise to the concerted practice. That is not to say that a competition authority must prove this by evidence. The CAB made no such pronouncement. The CAB correctly identified the presumption of causal connection, and that the presumption is an alternative method of proof. ⁸⁸ That is why the CAB then went on to consider whether the conduct that is the subject of the present appeal constitutes a “by object” restriction.

91 As I have stated above, where undertakings participate in a concerted practice, and remain active on the market thereafter, they are presumed to have taken into account the information exchanged. The presumption of causal connection also presumes that there is a causal link between the exchange of information, and undertakings’ subsequent conduct. This is a rebuttable presumption. If undertakings can show that the information exchanged played no part in their future conduct, then they cannot be sanctioned.

92 In the present case, the respondents took part in a concerted practice. They remained active on the market thereafter. It is therefore presumed that the respondents took the information exchanged into account in their future actions and strategy. The CAB in any event found that the respondents had subsequently used the information exchanged to convince customers to agree to the implementation of the FTZ Surcharge. ⁸⁹ The CAB therefore considered the possibility of proving by presumption a causal connection between the exchange of information and undertakings’ subsequent conduct. The CAB also found that such a connection was established through evidence.

93 The CAB thus directed itself correctly on the law on concerted practices. It consequently held that there was a concerted practice between the respondents. Insofar as the appellant or respondents are of the view that there was no such holding, or that it was incorrect, I disagree, and decline to disturb the CAB’s decision and findings on the existence of a concerted practice.

94 I now discuss the law on “by object” restrictions of competition. The parties agree, ⁹⁰ as the CAB held, ⁹¹ that the applicable test for identifying a “by object” restriction is set out in Groupement des cartes bancaires (CB) v European Commission, Case C-67/13 P, ECLI:EU:C:2014:2204 (“Cartes Bancaires”). I reproduce the relevant paragraphs in full:

95 The extract above gives some indication of the appropriate steps to take to identify a “by object” restriction of competition. However, it does not define what a “by object” restriction fundamentally is, beyond suggesting that they cover “certain types of coordination between undertakings [which] reveal a sufficient degree of harm to competition”: Cartes Bancaires at [49]. At best, a “by object” restriction is framed negatively in Cartes Bancaires, that it refers to conduct that is sanctioned for being anti-competitive without need to determine its precise economic or anti-competitive effects.

96 This definitional gap is the source of parties’ disagreement on appeal. The CCS’s case is that the test asks only whether the conduct reveals, by its very nature, a sufficient degree of harm to competition. Once that question is answered affirmatively, no examination of effects is required. The respondents’ case, and the CAB’s holding, is that the examination of economic context required by Cartes Bancaires must be sufficiently searching to demonstrate that the specific conduct, in the specific market, was so obviously harmful to competition that the competition authority could dispense with any effects analysis, and that the CCS failed to conduct that examination. ⁹²

97 As alluded to above at [29(a)], the CAB stated that only the “most well-established, egregious and obviously harmful” types of conduct may be classified as “by object” restrictions. ⁹³ This goes to conduct that is deemed “by their very nature, as being harmful to the proper functioning of normal competition”, or cases where the conduct itself reveals a “sufficient degree of harm” (quoting Cartes Bancaires at [50] and [57]). ⁹⁴

98 The appellant has characterised a “by object” restriction of competition as referring to the “objective purpose” or “objective aims” of the conduct. ⁹⁵ In other words, this is conduct “where the facts point objectively to the purpose of the conduct as being anti-competitive/restrictive of competition”. ⁹⁶

99 The respondents have not directly addressed the nature of a “by object” restriction of competition. However, the respondents have appeared to accept that it goes to conduct that reveals a “sufficient degree of harm to competition”. ⁹⁷ The respondents also appear to submit (or to accept the CAB’s determination, insofar as this reflects the CAB’s determination) that the difference between analysing the consequence or effects of conduct on the market in “by object” and “by effect” restrictions is one of degree. ⁹⁸

100 With this issue in mind, I shall first consider the nature of a “by object” restriction. I shall then consider the role of the economic context as raised in Cartes Bancaires at [53].

101 The first issue in this section is the proper scope of the prohibition on “by object” restrictions of competition.

102 The relevant provision in the EU is Article 101(1) of the TFEU. It states:

103 The scope of Article 101(1) is limited by its reference to “trade between Member States” and to “the internal market”. That is obviously not applicable to s 34 of the Competition Act. Where EU case law is considered, I shall limit myself to identifying the nature of the “object” in a “by object” restriction of competition.

104 The relevant provision in the UK is s 2(1) of the Competition Act 1998 (“UK Competition Act”). It states:

105 I do not propose to rehearse the entire corpus of law on “by object” restrictions of competition. However, understanding the nature of “by object” restrictions and the scope of the prohibition requires close attention to the precise facts of each case, the principles that the relevant court or tribunal declared were applicable, and the way in which these were applied to the facts.

106 A convenient place to begin is Société Technique Minière v Maschinenbau Ulm GmbH, Case C-56/65, ECLI:EU:C:1966:38 (“STM”). STM stands for the proposition that findings of a “by object” restriction and a “by effect” restriction were alternative, and not cumulative, requirements for sanction by the European Commission. This was based on the word “or” in the present equivalent of Article 101(1) of the TFEU: STM at p 249. Factors relevant to this analysis were the nature and quantity of the products covered, the position and importance of the parties in the relevant market, whether there was an isolated agreement or series of agreements, and the severity of the agreement (in the sense of whether there were other opportunities for competitors to access the relevant market): STM at p 250. The case concerned the grant of an exclusive right to sell certain public works equipment from a particular supplier within the French market. In that context, it was held that an exclusive right of sale did not necessarily amount to a “by object” restriction, but the terms of the agreement in question might give rise to a “by object” restriction of competition: STM at p 251.

107 In Compagnie Royale Asturienne des Mines SA v European Commission, Joined Cases 29/83 and 30/83, ECLI:EU:C:1984:130 (“Royal Asturienne”), the Court of Justice held that finding a “by object” restriction of competition involved “examining the aims pursued by the agreement as such, in the light of the economic context in which the agreement is to be applied” (at [26]). Royal Asturienne involved a prohibition on the import and resale of zinc products into Germany (where prices were much higher than other jurisdictions), and a contractual requirement compelling resale outside Europe. It was held that the abovementioned restrictions were designed to prevent intranational re-export of goods, to maintain the system of dual or differentiated pricing. In particular, I note that the language used by the European court was that the “conclusion [could not] be avoided” that the conduct was anti-competitive in its design: Royal Asturienne at [28].

108 Gøttrup-Klim e.a. Grovvareforeninger v Dansk Landbrugs Grovvareselskab AmbA, Case C-250/92, ECLI:EU:C:1994:413 (“Gøttrup-Klim”) concerned a Danish cooperative that aimed to provide members with farm supplies at lower prices. The cooperative expelled certain members who joined competing cooperatives: Gøttrup-Klim at [3]–[15]. It was held that the cooperative’s ban on dual membership was not prohibited under (what is now) Article 101(1) of the TFEU. This was because undertakings acting together through the cooperative served as a counterweight to the contractual powers of the suppliers, and made for more effective competition. In this context, dual membership would undermine the cooperative’s ability to act as a counterweight. It was thus held that the ban on dual membership was not necessarily a restriction of competition, and might even have pro-competitive effects: Gøttrup-Klim [28]–[34].

109 I should note at this juncture that despite STM drawing a clear distinction between “by object” and “by effect” restrictions on competition, the European cases did not always indicate whether the conduct in question was sanctioned or investigated as a “by object” or a “by effect” restriction. In fact, the analysis in Gøttrup-Klim was ostensibly on the “object or effect” of the ban (at [29]), and STM does not appear to have been considered by the European court in that case. This may differ from the later cases considered in this judgment. All that can be said at this juncture is that Gøttrup-Klim appears to be a case where conduct was not sanctioned where it was not clearly anti-competitive in nature or purpose. Nonetheless, cases such as Gøttrup-Klim remain useful in determining the type of conduct that could be sanctioned, and the analysis that the European Commission was required to undertake before sanctions could be imposed.

110 Two such cases are A. Ahlström Osakeyhtiö, and John Deere, which I have discussed in the preceding section. Both cases involved undertakings and trade associations investigated by the European Commission for information exchanges. In the former case, this took the form of quarterly pricing announcements in the trade press. In the latter case, this took the form of an information system relating to sales data. These two cases indicate that it is not enough to identify conduct as falling within certain pre-existing categories of sanctionable conduct (such as information exchanges). Rather, the competition authority must then establish that the conduct as implemented is of the sort that would restrict competition or reduce uncertainty between competitors.

111 The information exchange in A. Ahlström Osakeyhtiö was not considered “in itself” an infringement of (what is now) Article 101(1) of the TFEU (at [65]). I take this to mean that the Court of Justice also found that the conduct in question was not a “by object” restriction. This was on the basis that the pricing announcements did not reduce uncertainty as to competitor’s future conduct (ie, competitors would not have any insight into future pricing decisions): A. Ahlström Osakeyhtiö at [60]–[65]. In contrast, the information exchange in John Deere was an infringement of EU competition law. This was because the information exchange gave sellers information on their competitors’ sales data, and the information was not shared with non-participating sellers or customers: John Deere at [88]–[91].

112 The next case to be considered is Competition Authority v Beef Industry Development Society Ltd and Barry Brothers (Carrigmore) Meats Ltd, Case C-209/07, ECLI:EU:C:2008:643 (“Beef Industry”). The judgment in Beef Industry indicates that the “obviousness” of the alleged anti-competitive object plays some role in analysing whether conduct constitutes a “by object” restriction.

113 That case involved an overcapacity in the Irish beef industry. As a result, the 10 largest undertakings formed an association for the purpose of reducing the capacity for beef processing. The agreements involved, among other mechanisms, compensating undertakings to leave the industry: Beef Industry at [3]–[8]. In examining the agreement, the Court of Justice held that the object was to “change, appreciably, the structure of the market through a mechanism intended to encourage the withdrawal of competitors”: Beef Industry at [31]. Furthermore, agreements to put certain processing plants out of use for several years were “obviously intended” to prevent the entry of new competitors in the market: Beef Industry at [38].

114 A more nuanced approach was raised in the Opinion of Advocate General Trstenjak in Beef Industry, ECLI:EU:C:2008:467, (“Beef Industry Opinion”). The Beef Industry Opinion was cited favourably at several points in Beef Industry, and has been cited in several of the English decisions surveyed in this judgment. In brief, the Advocate General was of the view that if an agreement constituted a “by object” restriction of competition, it did not matter whether there were anti-competitive effects: Beef Industry Opinion at [38]. Although an anti-competitive object could be found in cases where a restriction of competition is a necessary consequence of an agreement or conduct, this did not mean parties’ intentions were irrelevant. Rather, rational economic operators are taken to have intended the ordinary consequences of their acts: Beef Industry Opinion at [44]–[46]. The Advocate General also rejected the notion that “by object” restrictions sanction only the most obvious cases of anti-competitive behaviour: Beef Industry Opinion at [97].

115 The appellant relied, ⁹⁹ as did the CAB, ¹⁰⁰ on T-Mobile. That case concerned a meeting between the five mobile telecommunications network operators in the Netherlands. There was a finding of a concerted practice in relation to, among other things, the reduction of standard dealer remuneration for certain subscriptions: T-Mobile at [9]–[12].

116 Once more, the Court of Justice took the view that a “by object” restriction may be found without regard for the actual or potential anti-competitive effects of the conduct investigated. All that was needed for the conduct to have the “potential to have a negative impact on competition”, taking into account, amongst other things, the nature of the products, and number of undertakings involved: T-Mobile at [31]–[33]. The Court of Justice considered that even if market conditions meant that dealer remuneration would nonetheless have been reduced without the concerted practice in question, there could still have been competition in relation to the timing, extent, and details of such modifications: T-Mobile at [40]–[41] and [68]–[69]. T-Mobile, and the Opinion of Advocate General Kokott in T-Mobile, ECLI:EU:C:2009:110 (“T-Mobile Opinion”), viewed a “by object” restriction as going to the “potential” or “capacity” to restrict competition. I return to this point later in this judgment, but it suffices to state at this juncture that such an approach appears somewhat out of step with the cases that came before and after.

117 A further case cited by the appellant is Expedia Inc v Autorité de la concurrence, Case C-226/11, ECLI:EU:C:2012:795 (“Expedia”). ¹⁰¹ The case concerned, in material part, a challenge over the binding effect (if any) of a now-repealed Commission Notice. That Commission Notice stated that if an agreement was between undertakings with a market share of less than 10% in aggregate, the European Commission would not sanction it as a “by object” restriction of competition: Expedia at [4] and [13].

118 Expedia dealt primarily with whether agreements between small undertakings could result in an appreciable effect on competition. As the appellant has correctly submitted, ¹⁰² the concept of “appreciability” is not found in the Competition Act. It is not contained in the European or UK statutory instruments either. The concept of “appreciability” is derived from case law, and is part of the CCS’s enforcement prerogative. What is important to note at this stage is that the Opinion of Advocate General Kokott in Expedia, ECLI:EU:C:2012:544 (“Expedia Opinion”) stated at [50] that market share was irrelevant in finding a “by object” restriction. However, the judgment itself stated that conduct would not be prohibited if it had an “insignificant effect on the market”: Expedia at [16].

119 With that, Cartes Bancaires falls to be considered. As it sets out the agreed test for identifying a “by object” restriction of competition, it is important not just for how it guides the cases that followed, but how it treated the cases that came before (even implicitly). Cartes Bancaires concerned an association of French banks that aimed to increase the interoperability of bank cards issued by their members. The association implemented a fee system whereby a levy was placed on inactive members. The European Commission deemed this a “by object” restriction of competition: Cartes Bancaires at [3]–[5].

120 The Opinion of Advocate General Wahl in Cartes Bancaires, ECLI:EU:C:2014:1958 (“Cartes Bancaires Opinion”) should be considered, before the judgment itself. Advocate General Wahl laid out several key principles at [37]–[62], derived from the pre-existing caselaw:

(a) First, that “by object” and “by effect” analyses are alternate ways to sanction and prohibit conduct;

(b) Second, that the examination of the agreement in question and the surrounding context must be detailed, but this is distinct from considering the effects of the agreement;

(c) Third, that the surrounding context can only be used to negate an initial finding of a “by object” restriction, and cannot be invoked to prohibit conduct that does not appear to constitute a “by object” restriction;

(d) Fourth, that in light of the broad and precautionary (in the sense of preventing subsequent conduct that is not anti-competitive) result of finding that conduct is a “by object” restriction, such findings should only be made where the conduct in question has been “constantly prohibited” as shown by “experience based on economic analysis”; and

(e) Fifth, the object of the conduct must reveal a “sufficient degree of harm” to be sanctionable.

121 The Court of Justice in Cartes Bancaires broadly accepted the Cartes Bancaires Opinion. Significantly, the Court of Justice characterised “by object” restrictions as cases where conduct has been shown to be “so likely to have negative effects, in particular on the price, quantity or quality of the goods and services, that it may be considered redundant… to prove that they have actual effects on the market”: Cartes Bancaires at [51]. The result was the test set out at [94] of this judgment. Although this appears as if the Court of Justice had in mind a positive definition of a “by object” restriction, the focus in [51] of Cartes Bancaires on the “price, quantity or quality of the goods and services” and the “poor allocation of resources to the detriment, in particular, of consumers” may be somewhat out of sorts with the focus in [53] of the “sufficient degree of harm to competition”. The former goes to consumer welfare (which may include the quality of goods etc), whereas the latter contemplates the conditions of competition itself. This is not one and the same. It is possible for competitors to collude in a manner that benefits consumers (such as by agreeing to undercut other competitors), while also being harmful to competition.

122 The language of “so likely to have negative effects” and “sufficient degree of harm” in [53] was also a significant departure from the position in T-Mobile at [31] that all that is needed is the “potential to have a negative impact on competition”.

123 With the test in mind, the Court of Justice held that the fee system was not shown to be a “by object” restriction of competition. Even if the fee system could be anti-competitive, it had not been shown that there was a sufficient degree of harm. It was held that the fee system was intended to combat free-riding, where certain banks benefited from other banks’ development of a combined payment system. This was not by nature harmful to competition: Cartes Bancaires at [69]–[75].

124 Cartes Bancaires was considered and applied in Dole Food Company, Inc., and Dole Fresh Fruit Europe v European Commission, Case C-286/13 P, ECLI:EU:C:2015:184 (“Dole”). The case concerned, at a high level, banana suppliers who exchanged information relating to their weekly prices. They also shared their quotation prices once they were announced: Dole at [8]–[17].

125 Similar to Cartes Bancaires, “by object” restrictions of competition were characterised as cases where negative effects are so likely that a full “by effect” analysis is unnecessary. The search for a “by object” restriction was distilled into consideration of the “objectives and the economic and legal context”: Dole at [115]–[117]. Importantly, it was observed that competition rules (in the EU context) are “designed to protect not only the immediate interests of individual competitors or consumers but also to protect the structure of the market and thus competition as such”: Dole at [125]. Though not an explicit finding made by the Court of Justice, I note that the Opinion of Advocate General Kokott in Dole, ECLI:EU:C:2014:2437 at [128] indicated that a finding that an information exchange constituted a “by object” restriction did not require that the relevant market be oligopolistic. The only general principle was that the “supply must not be fragmented”.

126 I turn next to ING Pensii v Consiliul Concurenței, Case C-172/14, ECLI:EU:C:2015:484 (“ING Pensii”), which concerned a client-sharing arrangement by pension fund managers. Cartes Bancaires was considered and applied at [33]–[34], to the effect that once conduct is identified as a potential “by object” restriction of competition, the factors in Cartes Bancaires (such as the economic context) should be considered. The analysis is to determine whether the conduct is capable of constituting a “by object” restriction, and not whether it is capable of affecting competition.

127  In ING Pensii, there were new national rules requiring individuals to sign up for private pension funds, and it was anticipated that many would (mistakenly) sign up to several funds rather than just one. The client-sharing arrangement was entered into to avoid having clients (and the consequent profits) allocated by the Romanian government, which is what would have happened under the new national rules. The client-sharing arrangement was a “by object” restriction, being designed to strengthen participating undertakings to the disadvantage of non-participating undertakings: ING Pensii at [35]–[38].

128 It was held in Toshiba Corporation v European Commission, Case C 373/14 P, ECLI:EU:C:2016:26 (“Toshiba”) that in cases where conduct has previously been found to be a “by object” restriction, the consideration of the economic context can be limited to the extent “strictly necessary” to find there has indeed been a “by object” restriction: Toshiba at [28]–[29]. The case involved a market sharing agreement whereby the undertakings involved refrained from selling their products in each other’s markets. Market sharing was a type of conduct that was commonly or clearly a “by object” restriction, which meant that the Court of Justice took no issue with there being only a brief consideration of the economic context.

129 The next case I consider is FSL Holdings NV v European Commission, Case C-469/15 P, ECLI:EU:C:2017:308 (“FSL Holdings”). FSL Holdings concerned a cartel relating to the exchange of price-related information for the market for bananas in southern Europe, along the same lines as in Dole: Opinion of Advocate General Kokott in FSL Holdings, ECLI:EU:C:2016:884 (“FSL Holdings Opinion”) at [12]–[15].

130 As in Toshiba, it was held that where an anti-competitive object is particularly serious (in that case concerning a price-fixing cartel), the consideration of economic context can be a limited exercise. In relation to information exchanges specifically, the FSL Holdings Opinion further stated that undertakings’ small size and market share will not necessarily mean that the conduct is not anti-competitive. This is because, according to the FSL Holdings Opinion at [107], conduct can constitute a “by object” restriction of competition independent of the conduct’s effect.

131 Gazdasági Versenyhivatal v Budapest Bank Nyrt, Case C-228/18, ECLI:EU:C:2020:265 (“Budapest Bank”) concerned an agreement in the Hungarian banking industry, for uniform interchange fees for credit card payments (at [4]–[10]). The relevant point at this stage of the analysis is that it was held in Budapest Bank that conduct can constitute both a “by object” and “by effect” restriction of competition: at [39]–[42]. Importantly, it was not the holding in Budapest Bank that a “by object” restriction is necessarily a “by effect” restriction. To my mind, this indicates that they are different analytical frameworks, and not that (as I discuss later in this judgment) a “by object” analysis is an evidential shortcut in obvious cases of “by effect” restrictions.

132 European Superleague Company SL v Fédération internationale de football association (FIFA) and Union of European Football Associations (UEFA), Case C-333/21, ECLI:EU:C:2023:1011 (“European Superleague”) was cited by the appellant for the proposition that “object” goes to the objective purpose or objective aims of conduct. ¹⁰³

133 The backdrop of the case was an attempt by several European football clubs to form a breakaway interclub continental competition. FIFA and UEFA are the global and European football governing bodies respectively. Their statutes made clear that prior approval was necessary before clubs could participate in such tournaments, and clubs and players that participated in non-approved tournaments would be expelled from FIFA- or UEFA-organised competitions: European Superleague at [23]–[31].

134 On the topic of “by object” restrictions of competition, the Court of Justice referred to the “objective aims which that conduct seeks to achieve from a competition standpoint”: European Superleague at [167]. Against that backdrop, the Court of Justice considered that the “specific nature of international football competition”, and the market structure in organising and marketing football competitions, meant that such rules could be legitimate. However, the lack of “transparent, objective, non-discriminatory and proportionate” criteria for a grant of approval meant that the FIFA and UEFA rules constituted by a “by object” restriction of competition. This was because the lack of such criteria made it more difficult for undertakings to set up competing competitions, primarily through restricting access to players and clubs: European Superleague at [175]–[179]. Apart from the clear exposition on the nature of a “by object” restriction of competition, I understand European Superleague to stand for the proposition that arrangements which ostensibly prevent competition or create barriers to entry will not necessarily constitute “by object” restrictions. Flexibility and practicality will be required in the analysis. The structure and nature of the market may justify the conduct or rules in question.

135 The appellant relied heavily on Banco BPN/BIC Português SA and others v Autoridade da Concorrência, Case C-298/22, ECLI:EU:C:2024:638 (“Banco”) and the Opinion of Advocate General Rantos in Banco, ECLI:EU:C:2023:738 (“Banco Opinion”). The case concerned an exchange of confidential information between credit institutions on their financing rates and the volume of loans disbursed. The undertakings managed 83% of all banking assets in the Portuguese banking sector, and the reference to the Court of Justice was explicitly “in the context of a concentrated market with barriers to entry”: Banco at [11]–[21].

136 It was held in Banco that the analysis for a “by object” restriction of competition was “first, the content of that agreement, decision or practice, second, the economic and legal context of which it forms a part and, third, its objectives”. So long as there was a sufficient indication that the conduct in question had been or would be deemed anti-competitive, the absence of direct precedent would not mean that conduct could not constitute a “by object” restriction: Banco at [41] and [44].

137 Three other points are worth noting from Banco. The first is that the Court of Justice, citing Thyssen Stahl v European Commission, C-194/99 P, ECLI:EU:C:2003:527, stated that transparency between undertakings (in non-oligopolistic markets) would likely lead to intensification of competition. However, this did not mean that uncertainty could be removed even as to the “timing, extent and details of any future changes in the conduct of its competitors on the market”: Banco at [53]–[54]. In other words, knowing what competitors are going to do is not necessarily objectionable. Knowing exactly when and how they will do it (when this is not otherwise known) is.

138 The second is the following passage from [55] of Banco:

139 The extract is to the effect that for an exchange of information to constitute a “by object” restriction of competition, it must be shown that the exchange of information could only or inevitably result in a distortion of the market. Such a reading is consistent with the further statements that the consideration is of the “immediate and direct aims” of the coordination or conduct, and that conduct constitutes a “by object” restriction if it “cannot pursue any objective aim other than that of distorting competition on that market”: Banco at [66] and [80]. In this sense, the judgment in Banco indicates that in identifying a “by object” restriction, the competition authority should exclude other reasons for the conduct in question, or at least consider whether there are such other reasons.

140 The third point arises from [92]–[93] of Banco, that even if an exchange of information does not directly result in undertakings immediately changing their conduct, or the structure of the market is such that undertakings are unable to immediately change their conduct, that does not mean there cannot be a “by object” restriction of competition. Even if parties cannot change tack immediately as a result of information exchanged, it is sufficient if they could react faster than they otherwise would.

141 In addition to the Banco judgment itself, the Banco Opinion is worth considering. The appellant has cited [61] of the Banco Opinion in relation to the meaning of “strategic” or “confidential” information. This is because information should generally be “strategic” or “confidential” before its exchange would constitute a “by object” restriction. The same paragraph indicates that the strategic usefulness of data depends on, amongst other things, the degree of concentration of the market. It was also indicated in the Banco Opinion at [49] that the pro-competitive effects of conduct can be considered, to rebut a presumption or initial finding that the conduct has caused or will cause a sufficient degree of harm to competition.

142 I turn now to European Commission v Servier SAS, Case C-176/19 P, ECLI:EU:C:2024:549 (“Servier”). Servier concerned disputes over the patents for a particular drug. I take three points from Servier relevant to the present appeal.

(a) First, that complexity by no means precludes a finding of a “by object” restriction. The need for a detailed analysis (as seen in Cartes Bancaires) of the agreement and the surrounding context indicates that a finding of a “by object” restriction is not reserved for arguments unlawful on their face. In broad detail, the patent-holder entered into agreements with potential competitors, that de facto extended the exclusivity period for the patent. This was not merely a “pay for delay” arrangement to prevent the entry of generic medicines, but a scheme of agreements involving the settlement of certain disputes, the grant of licences, and the assignment of patent applications: Servier at [3]–[25]. The Court of Justice did not definitively rule on the matter, remitting the case back to the General Court to determine whether the agreements were indeed “by object” restrictions.

(b) Second, that even where conduct is not intended to restrict competition or pursues a legitimate objective (in Servier, the settlement of disputes), conduct may nonetheless be prohibited if it also has an anti-competitive object, which is determined in the light of the surrounding context: Servier at [182] and [226].

(c) Third, because a “by object” analysis is separate from a “by effect” analysis, potential pro-competitive effects are not considered in determining if conduct constitutes a “by object” restriction: Servier at [288]. I shall return to this issue below, of whether conduct may be prohibited for being a “by object” restriction of competition, even if it has ambivalent or positive effects on competition.

143 I turn finally to the UBS Group AG v European Commission, Joined Cases T-441/21 etc, ECLI:EU:T:2025:337 (“European Government Bonds”). European Government Bonds is not groundbreaking in terms of the legal principles espoused, or the way in which the General Court considered the economic and legal context. In fact, much of the relevant sections appear to have been redacted for confidentiality reasons. Rather, European Government Bonds is relevant because it is factually analogous to the present appeal, involving employees sharing price-related information in a generally price-sensitive market. It is also relevant because the European Commission and the Court of First Instance were able to evaluate conduct in a complex market, including the way in which the primary and secondary bond markets worked.

144 The case concerned the market for sovereign bonds in the EU. The banks concerned were involved in both the primary and secondary market for bonds. It was found that traders from the banks were in communication, exchanging sensitive information on their trading strategies to increase their profits: European Government Bonds at [39]–[40]. No clear indication of the parties’ market shares appears to have been given, though the parties involved were “market makers”, particularly on the secondary bond market. On that basis, the Court of First Instance upheld the European Commission’s determination that the information exchanges constituted “by object” restrictions of competition.

145 I shall summarise the key principles to be drawn from the EU cases set out above, and the proper approach in Singapore law, later in this judgment. Suffice to say for now that the cases following Cartes Bancaires adopt a more flexible approach to identifying “by object” restrictions of competition, and that this may require evaluating complex conduct in complex markets.

146 With the relevant EU case law in mind, I turn now to the English case law.

147 The first English case I turn to is Bookmakers' Afternoon Greyhound Services Ltd v Amalgamated Racing Ltd [2009] EWCA Civ 750 (“Amalgamated Racing”). The case indicates that in determining whether conduct constitutes a “by object” restriction of competition, conduct that is ordinarily anti-competitive will not necessarily be sanctioned if it pursues or is part of an objective or purpose that is not anti-competitive.

148 The case involved the establishment of a second broadcaster for horseracing in the UK, to broadcast racing in licensed betting offices (“LBOs”): Amalgamated Racing at [4]–[6]. The English Court of Appeal accepted that to be viable, the new broadcaster would have to acquire exclusive rights to show races at LBOs, from a minimum number of racecourses, and these were not arrangements that the racecourses could make on their own. There was no prospect of a third broadcaster entering the market: Amalgamated Racing at [47]–[49] and [85]. Lloyd LJ referred to the Beef Industry Opinion, and considered that the case was one where the objective aim was to sponsor the entry of a new competitor into the market (which meant that the conduct should not be prohibited for being a “by object” restriction). Lloyd LJ also held that the fact that the conduct is designed to improve profitability of the undertakings concerned does not mean that conduct is anti-competitive. Profit is the motive of most commercial activity, and in that case, profit was achieved through introducing competition: Amalgamated Racing at [86].

149 The next case I consider is the decision of the Competition Appeal Tribunal in Sainsbury’s Supermarkets Ltd v MasterCard Incorporated [2016] CAT 11 (“Sainsbury’s”). Cartes Bancaires was referred to as setting out key principles for identifying a “by object” restriction of competition. In doing so, the Competition Appeal Tribunal added two points at [101] that are relevant in identifying “by object” restrictions:

(a) First, that because a finding of a “by object” restriction relieves a competition authority from undertaking an effects analysis, the former should not be used to avoid the latter, but instead should be deployed only where the harm to competition is “clear-cut and pronounced”.

(b) Second, that while identification of a “by object” restriction avoids the need for an effects analysis, the anti-competitive conduct needs to be considered in context, and that parties’ intentions can be relevant.

150 The complaint in Sainsbury’s was that companies in the Mastercard group implemented fees known as “Multilateral Interchange Fees”, which clients were required to pay on credit and debit card transactions: Sainsbury’s at [4]–[10]. The Competition Appeal Tribunal held that the conduct before it did not constitute a “by object” restriction. Although a price fixing agreement, it was a default provision that applied if banks did not apply their own interchange fees. This diluted any anti-competitive consequences. Furthermore, the fees were not an “ordinary” price fixing agreement, but instead reflected multiple factors (including encouraging card use balanced against encouraging merchants to accept them) and diverse interests of stakeholders, which prevented the companies in question from setting excessive fees: Sainsbury’s at [102].

151 The next case I consider is Agents’ Mutual Ltd v Gascoigne Halman Ltd (t/a Gascoigne Halman) [2017] CAT 15 (“Agents’ Mutual”), upheld on appeal in Gascoigne Halman Ltd v Agents' Mutual Ltd [2019] EWCA Civ 24 (“Agents’ Mutual (CA)”). The decisions of the Competition Appeal Tribunal and English Court of Appeal demonstrate that not all restrictions on an undertaking’s possible actions will be considered anti-competitive. One factor that may be relevant is the market power of the undertakings involved.

152 The Competition Appeal Tribunal cited Sainsbury’s (including the reference therein to Cartes Bancaires) for the principles relating to “by object” restrictions. However, the Competition Appeal Tribunal added a gloss to the principles from Sainsbury’s, that the need to show a sufficient degree of harm to competition meant that it would be insufficient that the conduct was “merely capable of resulting in the prevention, restriction or distortion of competition”. Evaluating the harm is distinct from considering the economic effects, though the agreement must be “capable of having some impact on the market or markets in question”: Agents’ Mutual at [149].

153 In that case, an association of estate agents operated an advertising portal. The association’s rules required, among other things, that members list properties only on one other competing portal. This was characterised as a “semi-exclusive purchasing obligation”. In relation to the economic context, the Competition Appeal Tribunal noted that in certain industries, undertakings (in that case estate agents) would have common commercial objectives, and the achievement of these through cooperation is not a breach of the UK Competition Act. The example raised in that case was common costs which estate agents may work together to reduce.

154 Furthermore, the parties involved were not alleged to hold any significant degree of market power (as the portal in question was a new one), and agents could choose whether or not to join the association (and thereby submit to its accompanying rules), and subsequently whether or not to advertise on another portal at all. The nature of the provision was that of a semi-exclusive purchasing obligation, and the economic and legal context “strongly [suggested] that its nature and purpose are not to harm competition”: Agents’ Mutual at [174]–[182].

155 The English Court of Appeal upheld the decision of the Competition Appeal Tribunal, finding that the nature and context of the arrangement did not reveal a sufficient degree of harm to competition. Like the Competition Appeal Tribunal, the English Court of Appeal identified Cartes Bancaires as setting out the relevant test for determining if conduct constituted a “by object” restriction of competition. In particular, the English Court of Appeal cited the Beef Industry Opinion in viewing the case as an example of an agreement being ambivalent in terms of its effect on competition (in the sense that any restriction on parties’ independence had to be assessed in light of its aim of promoting competition), and therefore not a “by object” restriction.

156 I turn finally to Lexon. The Competition Appeal Tribunal referred to Dole, Cartes Bancaires, and Budapest Bank (and the respective Opinions, where applicable) as setting out the applicable law on “by object” restrictions of competition: Lexon at [178]–[184]. It referred to the existence of a sufficient degree of harm to competition as the “essential legal criterion” for determining the existence of a “by object” restriction of competition. The Competition Appeal Tribunal stated that the consideration of the surrounding context was distinct from demonstrating the anti-competitive effects of the conduct. However, this did not mean that the effects should be ignored completely. The Competition Appeal Tribunal noted that conduct that would ordinarily be viewed as anti-competitive should not be sanctioned if the circumstances were such that it was incapable of producing harmful effects, or the conduct was pro-competitive. An example given, citing the Cartes Bancaires Opinion at [42], was that even cases of price fixing might not constitute a “by object” restriction if the undertakings “hold only a tiny share of the market concerned”.

157 The Competition Appeal Tribunal also made clear that “by object” restrictions should be construed narrowly. This was phrased variously in terms of requiring the conduct to have been “constantly prohibited”, or that the “harmful nature is proven and easily identifiable in the light of experience and economics” (both from the Cartes Bancaires Opinion), or “sufficiently reliable and robust experience” (from Budapest Bank).

158 Lexon involved the sale of a particular pharmaceutical drug. The market was concentrated, at first involving only two authorised suppliers of the pharmaceutical: Lexon at [5]. The undertakings involved exchanged information over a ten-month period, with the overall objective of slowing the decline in the price of the drug: Lexon at [3]–[8]. With the above principles in mind, the Competition Appeal Tribunal found that the information exchanged related to the undertakings’ commercial strategy, including volume of supply and prices quoted. This was against the backdrop of a new competitor entering the market, and the market was for a commodity product whereby undertakings could benefit by being “marginally more competitive on price”. There was also no indication of a legitimate competitive purpose. That being the case, the Competition Appeal Tribunal held that the information exchanges had an anti-competitive object: Lexon at [191]. The Competition Appeal Tribunal then considered that information exchanges was a recognised category of conduct which would cause serious harm to competition, and could be sanctioned as a “by object” restriction. Finally, the Competition Appeal Tribunal held that the Competition and Markets Authority’s consideration of the economic context at paragraphs 5.24 to 5.39 of its decision was sufficient. I set out the Competition and Markets Authority’s conclusion below:

159 I note also that the Competition Appeal Tribunal had itself undertaken a detailed consideration of the economic context before determining there was a “by object” restriction: Lexon at [64]–[84].

160 I now consider previous decisions by the appellant and CAB. Although these are not binding on me, it is useful to examine the analysis undertaken in this jurisdiction.

161 The first Singaporean case I consider is Fixing of monthly salaries of new Indonesian Foreign Domestic Workers in Singapore (2011) CCS 500/001/11 (“Foreign Domestic Workers”). The case involved a group of employment agencies that agreed to raise the salaries offered to foreign domestic workers whom they were engaging to work in Singapore. The consideration of a “by object” restriction was said at [61] to go to the “objective meaning and purpose of the agreement”.

162 There was only a brief definition of the market at [79]–[80] of Foreign Domestic Workers. This was on the basis that a distinct market definition is unnecessary in finding liability for a “by object” restriction and/or an appreciable effect on competition. This may be traced to the statement at [63], wherein the CCS took the position that there can be a “by object” restriction “even if an agreement does not have an effect on the market”.

163 Because the case involved price fixing, the CCS approached the matter as ipso facto a “by object” restriction. The CCS took the position that by collectively raising prices and forgoing competition in terms of price in that regard, employers had less incentive to switch between agencies: Foreign Domestic Workers at [71]–[75].

164 I turn now to Infringement of the section 34 Prohibition in relation to the price of ferry tickets between Singapore and Batam (2012) CCS 500/006/09 (“Batam Ferries”). The case concerned two ferry operators exchanging information on their pricing. The parties were the only two operators operating on the routes involved. Unlike Foreign Domestic Workers, the CCS clearly defined the relevant market. In that case, the relevant market was the ferry routes between the specific ports: Batam Ferries at [23]– [35].

165 The CCS then considered the economic context, and the operation of the relevant market. The relevant market was deemed a “highly concentrated market akin to that of a duopoly”. Citing T-Mobile, the CCS stated that there was no need to find an actual effect on the market, though it was acknowledged that undertakings might not face sanction if there was an insignificant effect on the market, such as by having a very small market share: Batam Ferries at [69]– [85]. The authority for the latter proposition was Völk v Vervaecke, Case C-5/69, ECLI:EU:C:1969:35 (“Völk”). Though phrased in terms of whether undertakings would be “subject to enforcement”, I note that the judgment in Völk referred to such a scenario falling outside of the prohibition altogether, and not merely meaning that enforcement would be unlikely.

166 Based on the principles above, the CCS found that there had been an agreement or concerted practice to exchange confidential price-related information. This removed each undertaking’s uncertainty as to their competitor’s future conduct. This constituted a “by object” restriction.

167 I turn next to Infringement of the section 34 prohibition in relation to the sale and distribution of fresh chicken products in Singapore (2018) CCCS 500/7002/14 (“Fresh Chickens”). The case concerned an agreement by several sellers of fresh chicken not to compete for each other’s customers, and to coordinate their price movements. The CCS cited Cartes Bancaires for the proposition that the prohibition on “by object” restrictions arises because certain types of conduct are, by their very nature, so injurious to normal competition that they may be sanctioned without regard to their effects: Fresh Chickens at [128] and [132].

168 In relation to the relevant market, the CCS returned to its position in Foreign Domestic Workers, that certain practices do not require a distinct market definition. I set out [173] in full:

169 However, the CCS also noted at [520] that the parties were estimated to control over 90% of the relevant market.

170 The short point that can be drawn from the cases above is that the CCS itself has not been entirely consistent in the way that it approaches the question of whether conduct constitutes a “by object” restriction of competition.

171 Before I set out the applicable test and principles in finding a “by object” restriction of competition, I address the potential argument that a “by object” restriction sanctions the same conduct as a “by effect” restriction. If correct, this means that the prohibition on “by object” restrictions essentially targets conduct that, if a full effects analysis was conducted, would necessarily be deemed a “by effect” restriction. If such a position is adopted, any sign that conduct might not harm competition should compel the competition authority to abandon its “by object” analysis. It would instead have to undertake a full effects analysis.

172 There are three related points or categories of arguments that appear to support the proposition above. The first of these is encapsulated in the Opinion of Advocate General Bobek in Budapest Bank, ECLI:EU:C:2019:678 (“Budapest Bank Opinion”) at [24]:

173 The statement that a “by object” restriction and a “by effect” restriction are “not ontologically different” is ambiguous. If it means that they both concern conduct that distorts or restricts competition, this is fully in line with, and supported by, the cases I have considered above. If the statement means that they are the same type of conduct, I am unsure if this meaning can be supported. The relevant legislation (whether under EU law, English law, or the Competition Act) does not go so far as to state that the distinction is simply a matter of procedure. I note also that in the Budapest Bank Opinion at [29], Advocate General Bobek accepted the possibility that certain conduct may be classified as both a “by object” restriction and a “by effect” restriction of competition. The Court of Justice held similarly at [43] of Budapest Bank. That being the case, I am unsure if a “by object” restriction can be deemed as only covering conduct that would, if a full effects analysis was conducted, inevitably result in an adverse effect on competition.

174 The second of these is demonstrated by two Opinions that may indicate that a “by object” restriction of competition is essentially a procedural or evidential shortcut. If correct, the object analysis excuses a competition authority from spending the time and resources investigating the precise anti-competitive effects of conduct in clear cases.

(a) The first example is the Cartes Bancaires Opinion at [35]. Therein, the Advocate General referred to “by object” restrictions as furthering “procedural economy”. Allowing competition authorities to sanction certain categories of conduct without needing to establish their specific effects reduces the need to undertake a complex and time-consuming analysis.

(b) The second example is the T-Mobile Opinion. The Advocate General therein noted at [43] that a prohibition per se (on “by object” restrictions) “conserves resources of competition authorities and the justice system”.

175 I do not think that the Opinions above are justificatory in nature. Rather, they are descriptive, of the side effects of a “by object” analysis. Just because the finding of a “by object” restriction helps conserve resources, does not mean that procedural or evidential economy is the raison d’etre of a “by object” analysis.

176 The third category is certain elements of European case law, an example being the decision of the Court of First Instance in GlaxoSmithKline Services Unlimited v European Commission, Case T‑168/01, ECLI:EU:T:2006:265 (“GlaxoSmithKline (CFI)”), upheld on appeal ECLI:EU:C:2009:610 (“GlaxoSmithKline”). The case concerned, in broad terms, an agreement to limit the parallel trade and resale of certain pharmaceuticals: GlaxoSmithKline (CFI) at [8]–[21]. The T-Mobile Opinion cited GlaxoSmithKline (CFI) at [147] as a possible case where the determination that conduct constituted a “by object” restriction required some consideration of its effects. If this characterisation is correct, this means that the difference between an “object” and an “effects” analysis is again one of degree rather than one of kind.

177 I disagree with such a characterisation. Read in context, [147] of GlaxoSmithKline (CFI) is best understood as a rejection of the European Commission’s determination that there was a “by object” restriction. A consideration of the effects of the agreements was therefore necessary. It was in any event held on appeal that a finding of a “by object” restriction does not require proof of disadvantages to consumers: GlaxoSmithKline at [62]–[64].

178 I have already set out above the relevant case law, particularly in the EU and the UK. Suffice to say, it points away from the need to consider the effect of conduct before deeming it a “by object” restriction. Nevertheless, it remains for me to decide which approach the competition authority in this country should follow. I turn to that task now.

179 The reason why I set out the numerous cases above (in particular the EU and English cases) is because the appellant, respondents, and CAB all claimed to be applying the law as espoused in those jurisdictions. However, the way in which the relevant court or tribunal applies a stated test is more important than the fact that a specific test is said to apply. Even if a court or tribunal purports to follow the same test as the cases that preceded it, it may nonetheless apply it in a different way. The court or tribunal may also put a gloss on the pronouncements from preceding cases, as was done in Sainsbury’s, for example.

180 What then is the “object” in s 34 of the Competition Act? Based on the cases above, I consider that this refers to conduct where the objectively assessed aim, purpose, or rationale is to prevent, restrict, or distort competition within the relevant market. I find useful the language of the Opinion of Advocate General Emiliou in CD Tondela – Futebol, SAD v Autoridade da Concorrência, Case C-133/24, ECLI:EU:C:2025:364 (“CD Tondela Opinion”, adopted by the Court of Justice in ECLI:EU:C:2026:361 at [41]), that in determining that conduct causes a “sufficient degree of harm” to competition, that is a determination that the conduct has a manifest anti-competitive economic rationale.

181 I do not think that this can be reduced to simply looking at the “objective purpose” or “objective aims” of conduct, as the appellant has proposed. ¹⁰⁴ This is because the EU cases that the appellant had cited require a competition authority to consider the “content of [an agreement’s] provisions, its objectives and the economic and legal context of which it forms a part”: see for instance Budapest Bank at [51]. I agree that these are all factors that should be considered. The placing of “objective” as a distinct but not entire part of the “object” of conduct risks confusion, and may be somewhat circular in nature. I therefore prefer the language of a “manifest anti-competitive economic rationale”.

182 I do, however, agree with the appellant’s characterisation of “by object” restrictions as cases “where the facts point objectively to the purpose of the conduct as being anti-competitive/restrictive of competition”. ¹⁰⁵ This is why conduct which only affects competition in a minor or ancillary way (one example being certain non-compete agreements following corporate buyouts) is not generally deemed a “by object” restriction. In such cases, there is no sufficient harm, as the facts may not point objectively to the purpose of conduct being anti-competitive (ie, there is no anti-competitive economic rationale).

183 As stated in the CD Tondela Opinion, an anti-competitive rationale does not go to undertakings’ subjective intention. This explains why the conduct in Beef Industry was sanctioned as a “by object” restriction of competition. Even if the intention was to reduce oversupply, what was being pursued was the removal of competitors from the relevant market. In this regard, I find helpful the Opinion of Advocate General Emiliou in ROGON GmbH & Co. KG v Deutscher Fußballbund e. V. (DFB), Case C‑428/23, ECLI:EU:C:2025:363 at [52]:

184 I also consider that Banco at [56], which follows the sections referred to at [137]–[139] above, is relevant:

185 In essence, the question that may be asked is “what else could this be for?” If the answer is that the conduct in question can only be explained as being for the sole or primary purpose of preventing, restricting, or distorting competition, then the conduct would be prohibited as a “by object” restriction. If, on the other hand, the question yields the answer that the conduct has some other possible purpose that is not anti-competitive, such as the reduction of common costs in Agents’ Mutual or to allow the entry of a new competitor as in Amalgamated Racing, then the conduct should not constitute a “by object” restriction. Whether there is some other possible purpose will usually require consideration of the economic context (though not an analysis of the precise effects of the conduct in question), which I will discuss in the next section of this judgment.

186 Based on my decision above, and the cases surveyed, I consider that there are several key principles inherent in any “by object” analysis. Not all of these are specifically contested in the present appeal. However, they are relevant in determining the present matter, and may be useful for future cases.

187 The first key principle apparent from the case law surveyed above is that the prohibition on “by object” restrictions relates to their harm to competition. This is reflected in s 34 of the Competition Act itself, which refers to the “prevention, restriction or distortion of competition within Singapore” (emphasis added), as well as Cartes Bancaires at [50] and ING Pensii at [32], which refer to “by object” restrictions causing harm to the “proper functioning of normal competition”. The focus on harm to competition is distinct from any harm that may arise to consumers, or any economic harm generally. After all, competition law does not only relate to consumers’ and companies’ interests, but the structure of the market and competition itself: Dole at [125] and European Government Bonds at [1549].

188 This is also the case in Singapore. The Competition Act is intended to “promote the efficient functioning of our markets and hence enhance the competitiveness of our economy”: Singapore Parliamentary Debates, Official Report (19 October 2004) vol 78; at col 864 (Dr Vivian Balakrishnan, Senior Minister of State for Trade and Industry). That being so, I think it is important to make clear that the appropriate focus is on the functioning of the relevant market and harm to competition, rather than a certain economic outcome.

189 The second, and related, key principle is that the prohibition on “by object” restrictions of competition and “by effect” restrictions of competition are intellectually distinct exercises. They consider the same conduct from different perspectives. This is supported by the decision in Budapest Bank, that conduct can constitute both a “by object” and a “by effect” restriction of competition. Such an interpretation coheres with the wording of s 34 of the Competition Act. The provision prohibits conduct that has the “object or effect” of preventing, restricting, or distorting competition. I agree with the interpretation in STM, that the disjunctive “or” indicates that these are distinct analyses. To collapse them into a single set of conduct, or to hold that one is a subset of the other, would be to place on the provision an interpretation that its words cannot bear. Such an interpretation would also, by rendering otiose the word “or” in the phrase “object or effect”, fail to give full meaning to each of the words used by Parliament. This is to be avoided: Blackstone Asia Real Estate Partners Ltd v Standard Chartered Bank (Singapore) Ltd [2026] 1 SLR 251 at [24(a)].

190 That is not to say that the constituent elements of the tests do not overlap, nor that conduct constituting “by object” restrictions cannot also be “by effect” restrictions. I acknowledge the argument, canvassed above at [172]–[175], that “by object” restrictions are a subset of “by effect” restrictions on competition. I agree that “by object” restrictions can also be found to be “by effect” restrictions on competition (assuming that an effects analysis is conducted). Moreover, as a matter of practice and practicality, conduct will usually first be assessed on whether it is a “by object” restriction, before assessing if it is a “by effect” restriction of competition. However, that does not mean that one is a subset of the other, nor that a failure to show or prove an anti-competitive effect means that conduct cannot constitute a “by object” restriction.

191 The third key principle is that in considering if conduct constitutes a “by object” restriction of competition, parties’ subjective intentions are not determinative. As can be seen from the cases above, the analysis is an objective one, but this is not to say that subjective intention is never relevant. As set out in Cartes Bancaires at [54], a competition authority, tribunal, or court may consider subjective intention. English law is even clearer in this regard, as seen from the statement in Sainsbury’s that subjective intention can be a relevant factor (at [149] above).

192 That said, a lack of subjective intention does not obviate or prevent a finding that there has been a “by object” restriction. As seen from Beef Industry, the fact that undertakings may not believe that conduct prevents, restricts, or distorts competition will not necessarily mean that their conduct does not constitute a “by object” restriction.

193 However, the fact that the undertakings involved do not subjectively intend to restrict or distort competition may be a relevant factor. The lack of anti-competitive intention, or the existence of a pro-competitive subjective intention, may well indicate that there is an alternative intention or economic purpose that makes it inappropriate to find there is a “by object” restriction. This follows from my holding that the “by object” limb of s 34 of the Competition Act asks whether the conduct in question is known to or can only be to distort the market. The fact that undertakings hold a genuine and credible belief that conduct is not anti-competitive might indicate that the conduct is not known to or can only be to distort the market.

194 The fourth key principle is that although “by object” restrictions go to conduct that causes a “sufficient degree of harm” to competition, this does not imply a particular level of anti-competitive impact. This would place “by object” restrictions as a type of “by effect” restriction. I have stated this not to be so. To my mind, the phrase “sufficient degree of harm” is a qualitative characteristic rather than a quantitative one: CD Tondela Opinion at [26]–[28]. In other words, the phrase is descriptive: Pablo Ibáñez Colomo, “Restrictions by object under Article 101(1) TFEU: From dark art to administrable framework” (2024) 43 Yearbook of European Law 224 at p 227–228. It identifies the kind of harm that the conduct is inherently liable to cause, not the degree to which that harm has materialised or can be measured.

195 In this regard, the approach is not formalistic in nature. It is not a matter of finding that the moment conduct fits a particular description, such as price fixing, then it would constitute a “by object” restriction. The cases surveyed above do not say that there is anything wrong with having a starting point that particular classes of conduct constitute “by object” restrictions. What the cases do not support, is making that starting point also the end point of the analysis. There is a need for a holistic consideration, including of the economic context and the factors listed in Cartes Bancaires at [53]. It is only after these factors are considered in sufficient detail that a determination can be made as to whether the conduct in question causes a sufficient degree of harm, and thereby constitutes a “by object” restriction.

196 It thus cannot be said that the moment conduct falls within a particular class of anti-competitive conduct, it would automatically and necessarily constitute a “by object” restriction. I note the CAB’s reference to the possibility that price fixing will always constitute a “by object” restriction. ¹⁰⁶ In Sainsbury’s, the conduct investigated was a form of price fixing. However, the particular context and purpose of the Multilateral Interchange Fees (as a default rate, and one intended to balance encouraging the use of credit and debit cards and their acceptance by merchants) meant that it was not deemed a “by object” restriction. As this particular question does not arise for my determination in this appeal, I say no more.

197 The fifth key principle is that “by object” restrictions are determined narrowly and restrictively: Cartes Bancaires at [58] and Budapest Bank at [54]. The appellant does not dispute this. ¹⁰⁷ This is consistent with the consequences of a determination that conduct constitutes a “by object” restriction. In such a scenario, the competition authority is excused from proving the anti-competitive effects of the conduct. Instead, the conduct can be prohibited and sanctioned without further review. This peremptory effect on economic conduct in turn justifies, and even demands, that the power to proscribe be limited to cases where the anti-competitive purpose or aim is clear.

198 This principle does not mean that a competition authority must show that the exact type of behaviour has previously been classified as a “by object” restriction. If that was the case, it would never be possible to find a “by object” restriction save where it has already been found. The parties have not argued that there is a numerus clausus of “by object” restrictions of competition. Neither is there a specific number of examples which the competition authority needs to show, where the conduct in question has been deemed to be harmful to competition. So long as it can be said, in advance and with a degree of certainty, that the conduct is of the sort that would have anti-competitive effects or has previously been deemed a “by object” restriction, that will do.

199 The sixth principle is that even though “by object” restrictions are determined narrowly and restrictively, this does not limit them to conduct that appears clearly anti-competitive on its face. This can be seen from Servier. A “by object” restriction concerns conduct whose sole or primary purpose is preventing, restricting, or distorting competition (ie, the manifest anti-competitive economic rationale). Determining conduct's purpose is a necessary precondition essential for finding a “by object” restriction. Finding a “by object” restriction therefore requires identifying this purpose, even where identification may be difficult.

200 Perhaps the fifth and sixth principles explain the origin of the CAB’s determination that only the “most well-established, egregious and obviously harmful forms of anti-competitive conduct” would constitute a “by object” restriction. ¹⁰⁸ To the extent that this is simply a statement that “by object” restrictions should be determined restrictively, based on there being a manifest anti-competitive economic rationale, the statement is correct. If, however, the intended meaning is that only where the anti-competitive impact is clear on its face can a “by object” restriction be found, I consider that this is inconsistent with cases such as Servier (the principles of which I do not take to be controversial or in dispute), and such an approach should not be followed.

201 I note at this juncture that there is some indication in earlier case law that conduct may constitute a “by object” restriction if it has the “potential to have a negative impact on competition”. This sits uneasily with the “sufficient degree of harm” approach seen from Cartes Bancaires and the cases applying it. A test of “potential negative impact” should not be adopted. I explain why.

202 First, the supposed test or requirement that conduct has the “potential to have a negative impact on competition” is derived from T-Mobile at [31]. That paragraph referred to and agreed with [46] of the T-Mobile Opinion. The latter has to be read in context. The T-Mobile Opinion rejected at [45] the argument that conduct will not be deemed a “by object” restriction if it can be shown that no negative consequences will result. The next paragraph of the T-Mobile Opinion went on to reject the view that a finding of a “by object” restriction is dependent on showing an anti-competitive effect. The T-Mobile Opinion then went on to compare “by object” restrictions to drink-driving, where the wrong is in the act done, regardless of whether any accident was caused. Read in context, Advocate General Kokott was not saying that mere potential to impact competition is enough. Rather, she meant that the anti-competitive object is identified independently of any finding concerning the consequences (or effects) of that particular conduct.

203 Second, a test of “potential negative impact” would conflict with the search for a manifest anti-competitive economic rationale and/or a sufficient degree of harm. If all that was needed was “potential” impact, it should follow that there is no need to show a “sufficient degree of harm”. That is not the case. In Cartes Bancaires, the General Court had established that the conduct in question was “capable of restricting competition”. I do not see this as any different from the conduct having the “potential to have a negative impact on competition”. Yet the Court of Justice found that there was no “by object” restriction of competition, because it had not been shown that there was a sufficient degree of harm: Cartes Bancaires at [69].

204 Third, a test of “potential negative impact” conflicts with the requirement that “by object” restrictions be interpreted restrictively, which the appellant appears to agree with. ¹⁰⁹ If the requirement is merely “potential negative impact”, then this risks opening up a wide range of conduct to sanction.

205 On a related point, rejecting the “potential negative impact” and requiring a narrow interpretation of “by object” restrictions would not unduly impede a competition authority’s ability to investigate and sanction undertakings. As Advocate General Wahl noted in the Cartes Bancaires Opinion at [62], the competition authority is still free to engage in a “by effect” analysis.

206 Fourth, I consider that a test of “potential negative impact” would collapse the distinction between “by object” and “by effect” restrictions. For conduct to constitute a “by effect” restriction, it must have an actual adverse effect on competition. For conduct to have an actual adverse effect on competition, it must necessarily have also had a potential negative impact (or effect) on competition, which the competition authority would naturally have had in mind when investigating the conduct. If that is so, any “by effect” restriction of competition would also have been a “by object” restriction. If a test of “potential negative impact” is adopted, there would be no need to ever resort to the “by effect” restriction. This would render otiose the reference to “by effect” restrictions in s 34 of the Competition Act. For these reasons, such a test should not be adopted.

207 This exposition of key principles can be summarised in the following manner: Conduct that has a manifest anti-competitive economic rationale may be sanctioned as a “by object” restriction. This is where the conduct can only or must necessarily be explained as having an anti-competitive aim.

208 The final issue concerns the proper role of economic context in determining if there is a “by object” restriction of competition.

209 The CAB held that the economic context must be “adequately evaluated (without necessarily engaging in a full-blown effects-based analysis of the impugned conduct)”. ¹¹⁰ Such an evaluation is necessary to a conclusion that the conduct “in itself poses a sufficient degree of harm to competition”. ¹¹¹

210 The appellant’s submission is that the CAB set out the correct test or standard of inquiry. ¹¹² However, the appellant disagrees with the Decision insofar as the Decision indicates that some consideration of the economic effects is necessary. ¹¹³ Although the economic factors considered may be the same, consideration of the effects, on the appellant’s view, is distinct from considering whether the objective purpose or aim of the agreement is anti-competitive. ¹¹⁴

211 The appellant’s position is that an analysis of the economic context is to “ensure that there are no legal or factual circumstances surrounding the agreement, decision or concerted practice that could preclude or prevent the conduct involved from having a sufficient degree of harm to competition”. ¹¹⁵

212 The appellant’s view is that the economic context can or may shed light on the objective purpose or aim of the conduct. There are two ways which the appellant suggests that this analysis would militate against the finding of a “by object” restriction. The first is that the practice may not be, when considered in context, be inherently anti-competitive. The second is that the conduct may be “incapable of having a sufficient degree of harm to competition”, for example where there is no competition to restrict. ¹¹⁶

213 In contrast, the respondents submit that the appellant has mischaracterised the CAB’s views on the matter. ¹¹⁷ The respondents are, unsurprisingly, in agreement with the CAB’s holding on this point. The respondents have also submitted that the consideration of the potential or likely effects of the conduct on undertakings’ market behaviour would be necessary in determining if the conduct constituted a “by object” restriction. ¹¹⁸

214 Unlike the preceding section of this judgment, the issues in this section are relatively specific. There is no real ongoing debate over the nature of the economic context, though the appellant and respondents in this case appear to differ. I shall therefore briefly consider the EU and English case law, and the general role of the examination of the economic context. I shall then consider the specific issues that the parties have raised in this regard, ie, the relevance of market share or market power in considering the economic context, and the nature or extent of evaluation necessary.

215 I begin first with Cartes Bancaires, and the Cartes Bancaires Opinion. The Court of Justice held in Cartes Bancaires that in determining if there is a sufficient degree of harm to competition, “regard must be had to… the economic and legal context of which it forms a part”: Cartes Bancaires at [53]. It was similarly stated in the Cartes Bancaires Opinion that the object of the agreement would have to be considered in the light of the economic context: Cartes Bancaires Opinion at [38]. However, the economic context should only reinforce or weaken the finding that there is an anti-competitive object. Evaluation of the economic context should not, on the Advocate General’s view, be used to establish an anti-competitive object where the competition authority cannot determine that there is such an object: Cartes Bancaires Opinion at [44], [139], and [147]. In summary, both Cartes Bancaires and the Cartes Bancaires Opinion indicate that the economic context is useful insofar as it sheds light on the existence or non-existence of a manifest anti-competitive economic rationale (or sufficient degree of harm to competition).

216 The Beef Industry Opinion set out the view that consideration of the legal and economic context plays an exclusionary role. In other words, the legal and economic context would be considered if it “cast doubt on the existence of a restriction of competition”: Beef Industry Opinion at [50]. In this regard, the Advocate General set out three categories at [51]–[54], whereby the surrounding context would militate against a finding of a “by object” restriction:

(a) First, where limitations on freedom may not have an effect on competition, such as where the undertakings involved may not be competitors.

(b) Second, where conduct is “ambivalent in terms of its effects on competition”, in the sense that the conduct may be pro-competitive.

(c) Third, ancillary arrangements to primary agreements that are not anti-competitive, such as non-compete agreements following buyouts or restrictions on certain activities due to professional ethics.

217 Although this part of the Beef Industry Opinion was not cited in the Beef Industry judgment itself, the categories set out in the Beef Industry Opinion have been referenced and adopted in several cases:

(a) The first of these cases is Amalgamated Racing. Lloyd LJ identified the case as falling within the second category from the Beef Industry Opinion. The overall aim of the conduct was pro-competitive (to enable the entry of a second broadcaster into the market), and the conduct in question (exclusive broadcasting rights with several racecourses) was necessary for the second broadcaster to be economically viable. These factors meant Amalgamated Racing was a case where the economic context militated against a finding of a “by object” restriction.

(b) A second case falling within the second category from the Beef Industry Opinion is Agents’ Mutual (CA). The English Court of Appeal found that economic context was not, and did not need to be, determined merely by examining if there was a pro-competitive purpose. Instead, particular weight was given to the undertaking's lack of market power. This justified finding that the economic context was such that restrictions on the number of advertising portals that estate agents could join did not reveal a sufficient degree of harm, and did constitute a “by object” restriction of competition: Agents’ Mutual (CA) at [59]. I also note that the Competition Appeal Tribunal had stated that while the conduct must not “merely” be capable of preventing, restricting, or distorting competition, there must be some capability of impacting the market in question: Agents’ Mutual at [149]. I take this to mean that while harm to competition must be shown to be sufficient, it suffices to show that there will be some economic effect (without any criteria of sufficiency).

218 The appellant has referred me to E.ON Ruhrgas AG and E.ON AG v European Commission, Case T-360/09, ECLI:EU:T:2012:332 (“E.ON”). E.ON was a case where the economic context meant that there was no “by object” restriction of competition. Because competition rules generally relate only to consequences on competition, there could not be a “by object” restriction of competition in sectors that were not open to competition. The conditions of competition would be examined in the light of actual and potential competition. The latter would be based on whether entry into the market was economically viable, with intention to do so being relevant but not essential: E.ON at [84]–[87]. This would not require a legal monopoly, with the analysis instead focusing on whether there would be “real concrete possibilities for the undertakings concerned to compete among themselves or for a new competitor to enter that market and compete with established undertakings”: E.ON at [105].

219 E.ON concerned the joint construction and operation of a gas pipeline by the leading French and German natural gas suppliers. The agreement included a limitation on each undertaking selling in the other’s market, with regular meetings on the undertakings’ arrangement: E.ON at [14]–[27] and [164].

220 The French market was the subject of a legal monopoly until 2000, so there would be no potential competition (and therefore no “by object” restriction of competition) until the monopoly ended. Similarly, there was a de facto monopoly in Germany until 1998, due to the existence of exclusive supply agreements that were permitted under German law. Although there were two other market entrants that were different from each other around the relevant time, this did not mean there was potential competition. The two were somewhat unique, one being a joint venture between a very large German gas consumer and a large Russian producer, and the other being a large gas producer: E.ON at [89]–[113].

221 It is also worth noting that the European court in E.ON held that the fact that the parties entered into an agreement was not evidence of there being potential competition. This was because the economic context indicated that there was no competition to restrict. At most, it showed that the undertakings involved feared competition: E.ON at [115]. I take from this case the point that even if parties enter into a supposedly anti-competitive agreement, there still must be an examination of the economic context. The examination may find that the agreement is not anti-competitive at all, because the economic context is such that there is no competition to restrict.

222 The appellant also referred me to Toshiba, for the proposition that the consideration of the economic context is limited to what is “strictly necessary” (see [128] above). ¹¹⁹

223 Similar to E.ON, Toshiba concerned a market sharing agreement under which Japanese undertakings would refrain from selling products in the European market: Toshiba at [6]–[11]. In that case, the Court of Justice held that market sharing agreements constituted “by object” restrictions, and that “such an object cannot be justified by an analysis of the economic context”. In the circumstances, the consideration of the economic context could be limited to the extent “strictly necessary” to find there has indeed been a “by object” restriction: Toshiba at [28]–[29]. I shall return to this point below. Suffice to say for now that there is an apparent conflict between the existence of a category of practices which cannot be “justified” by the economic context, and the need to consider the economic context (if only to the extent of what is “strictly necessary”) in determining if there is a “by object” restriction of competition.

224 However, the Court of Justice in Toshiba did also uphold the General Court’s finding that there was potential competition in the relevant market. This was seen by Japanese undertakings accepting projects from European customers. The fact that a gentlemen’s agreement was entered into also indicated that there was potential competition. This latter point may conflict with E.ON, which does not appear to have been referenced by the European court in Toshiba. However, the issue of the existence of actual or potential competition does not arise in the present case, as the ID and Decision both proceeded on the basis that the respondents in this case were operating in the same market as Mac-Nels and Penanshin.

225 I finally consider Lexon. I have set out the general points listed by the Competition Appeal Tribunal above at [156]–[159]. The first point worth repeating from that case is that there is a distinction between considering the economic context, and demonstrating anti-competitive effects. The second point is that even if conduct is generally regarded as anti-competitive, it should not be prohibited if it is incapable of producing harmful effects or may be pro-competitive. One example is that cases of price fixing may not constitute a “by object” restriction if parties “hold only a tiny share of the market concerned”.

226 I agree with the appellant that the economic context can or will be relevant to the determination of whether there is a “by object” restriction of competition, insofar as the economic context sheds light on the aim of the conduct viewed objectively. This is clear from Cartes Bancaires, and the cases that apply the different categories from the Beef Industry Opinion, among others. This is why there is usually no “by object” restriction of competition if the undertakings are not competitors. In such cases, it will be relatively rare for there to be a manifest anti-competitive economic rationale.

227 The relevant factors that will usually be considered include the “nature of the goods or services affected, as well as the real conditions of the functioning and structure of the market or markets in question”: Cartes Bancaires at [53]. As may be seen from this extract, consideration of the relevant economic context will likely require identifying the relevant market or markets, as identifying and considering the relevant market will (usually) shed light on whether the objective aim or intention is to restrict or distort the market. If the relevant market is not defined, there may be a risk that the economic context will not be considered sufficiently. For example, if a competition authority does not properly define the relevant market, the fact that undertakings are not competitors may not be properly considered. Of course, it is open to the competition authority to consider other factors, and for undertakings to raise such factors if relevant.

228 As for the difference between an examination of the economic context and an examination of the effects of the conduct in question, I also agree with the appellant’s submission that consideration of the economic context is, and should be, distinct from analysing of the effects of the conduct in question. If the former entailed the latter, this again risks collapsing the distinction set out in s 34 of the Competition Act between the “object” and the “effect”. However, this delineation is not one that brooks no exception.

229 First, the different analyses do not necessarily consider completely distinct issues. In Agents’ Mutual, the semi-exclusive purchasing obligation allowed estate agents to use other listing portals, provided they limited themselves to just one additional portal. This formed part of the economic context that determined no “by object” restriction existed. Yet this same fact, that estate agents could and likely did advertise on another portal, would have been relevant to an effects analysis. Indeed, the Competition Appeal Tribunal noted at [184] that considering this factor could shade into examining the conduct's effects.

230 Second, the fact that evaluating the “object” of conduct is distinct from considering its “effect” does not mean that the nature or existence of effects should be ignored. Cases like Gøttrup-Klim demonstrate that conduct cannot be considered in abstract. I think this accords with, and explains, the comment by the Competition Appeal Tribunal that while the distinction between economic context and economic effect must remain, a competition authority is “allowed to consider whether [conduct] is likely to have any impact on the market”: Agents’ Mutual at [184]. Such an approach also fits with the pronouncement in Lexon that a competition authority only needs to consider if there could be a “material effect” on the market: Lexon at [231]. These considerations are for the purpose of finding a manifest anti-competitive economic rationale (or sufficient degree of harm). The fact that there is unlikely to be any impact, or if the impact would not be material, could point away from there being such a rationale or harm to begin with.

231 Similarly, the categories in Beef Industry Opinion (at [216] above) are examples of conduct that may appear to be “by object” restrictions, but the economic context is such that they should not be deemed as such. I note that the first two categories in the Beef Industry Opinion are phrased in terms of “effect”. I do not think that collapses the distinction between “by object” and “by effect” restrictions. Rather, it demonstrates that in cases where a competition authority or undertaking can show that there is unlikely to be an anti-competitive effect (or such an effect is minimal), or that the likely effects are not anti-competitive (but may instead be pro-competitive) in nature, it would be less likely for there to be a manifest anti-competitive economic rationale.

232 I should reiterate that the analysis of the economic context should not cross into an evaluation of the effects of the conduct. The search is for a manifest anti-competitive economic rationale, and insofar as economic effects may be relevant to this search, it would generally suffice to show that there is some impact, without weighing what that impact is precisely. To the extent that the consideration is whether the conduct is “capable in an individual case of resulting in the prevention, restriction or distortion of competition” (to borrow the phrase from T-Mobile at [43]), I do not think that conduct analogous to that in Gøttrup-Klim or falling within the categories in the Beef Industry Opinion can properly be described as being capable of preventing, restricting, or distorting competition.

233 With this general survey of the relevant law, I turn to the two more specific issues raised in the present dispute.

234 To recapitulate, the respondents have a combined market share of about 5.75% of the relevant market. The respondents submit that undertakings’ market share should be considered as part of the relevant market structure and economic context. ¹²⁰ The respondents’ counsel also took the point that the appellant had not shown any consideration of the respondents’ market share. ¹²¹

235 The appellant has referred to the Commission Decision of 23 April 1986 relating to a proceeding under Article 85 of the EEC Treaty (1986) Commission Decision IV/31.149 (“Polypropylene Decision”) for the point that collusion involving undertakings with a small market share can constitute a “by object” restriction. Specifically, the appellant has raised the fact that the smallest producer sanctioned had a market share of less than 2%. ¹²² The respondents do not contest the possibility that undertakings with a small market share may be sanctioned. Their submission is instead that market share is relevant, and inadequately considered. All I say at this juncture is that while the smallest undertaking sanctioned in the Polypropylene Decision had a market share of less than 2%, the four major producers which participated in the infringements accounted for about half the market: Polypropylene Decision at [5] and [8].

236 The appellant has also referred to the Expedia Opinion at [50], that “it must be presumed that undertakings which enter into an agreement with an anti-competitive object always intend an appreciable effect on competition, irrespective of the size of their market shares and turnover”. ¹²³ As I have noted above, the judgment itself stated that conduct would not be prohibited by Article 101(1) of the TFEU if it had an “insignificant effect on the market”: Expedia at [16].

237 The appellant further referred to European Government Bonds and the underlying contested decision of the European Commission, for the similar proposition that “an undertaking cannot use its small size to claim that its actions are incapable of harming competition in the market”. ¹²⁴ I think it is correct to say, as the appellant has, ¹²⁵ that the General Court in European Government Bonds rejected the claim that the European Commission did not account for the supposedly small size of the banks concerned. However, the General Court’s decision was that the banks involved had not provided evidence that it had a small market share, nor had they shown how their size would have been relevant to the finding in question: European Government Bonds at [1492]–[1495]. This was not a pronouncement that market share will never be a relevant part of the economic context.

238 On the respondents’ part, I was referred to Batam Ferries at [71], that CCS was of the view that:

239 I have noted above at [165] that the authority cited in the quoted extract is Völk, and Völk indicates that a small market share may mean conduct is not a “by object” restriction. However, the respondents’ reference to Batam Ferries faces the issue that the quoted extract appears on its face to go to the CCS’s decision, discretion, or willingness to enforce s 34 of the Competition Act, rather than the CCS’s jurisdiction to do so. That being said, the relevant quote from Völk is that the “weak position which the persons concerned have on the market of the product in question” may mean the conduct is not prohibited at all under (what is now) Article 101(1) of the TFEU.

240 I consider next the FSL Holdings Opinion. The FSL Holdings Opinion was referred to favourably by the Court of Justice at various points in the judgment. The Advocate General took the position that the undertakings’ small size and market share in that case did not detract from the existence of an anti-competitive object: FSL Holdings Opinion at [107]. However, the Advocate General did not exclude the possibility that size and market share could be relevant in determining if there is a “by object” restriction of competition.

241 As for the role and relevance of market power, Allianz Hungária Biztosító Zrt v Gazdasági Versenyhivatal, Case C‑32/11, ECLI:EU:C:2013:160 (“Allianz Hungária”) concerned agreements between insurers and vehicle repair shops in Hungary on the conditions and rates for certain repair services: Allianz Hungária at [6]–[11]. The Court of Justice explicitly stated at [48] that companies’ market power can be a relevant factor in determining if there was a “by object” restriction of competition. Allianz Hungária was referred to in Agents’ Mutual (CA), and the English Court of Appeal in that case referred repeatedly to the undertaking’s lack of market power in concluding that there was no “by object” restriction of competition: Agents’ Mutual (CA) at [49], [59] and [61]–[63].

242 Amalgamated Racing also referred to market power as relevant to determining whether there was a “by object” restriction. This could be for the purpose of confirming that there would be a “by object” restriction, or indicating that there was no “by object” restriction of competition.

243 From the above cases, I can draw a simple, though perhaps not entirely satisfactory, conclusion. Undertakings’ market power, and relatedly market share, can be, but is not necessarily, a relevant part of the economic context.

244 A hypothetical example may be useful. The CAB posited at [110] of the Decision that smaller undertakings such as the respondents would often act as price-takers. Let us assume that a larger undertaking raising its prices means that other undertakings will inevitably raise their prices as well. In the circumstances, I accept that in considering whether two smaller undertakings discussing and agreeing that they would raise their prices would constitute a “by object” restriction of competition, their respective or collective lack of market power could be a relevant part of the economic context. This is because a lack of market power may be indicative of a lack of manifest anti-competitive economic rationale. In other words, if two small undertakings discuss and disclose what they were already going to do, in circumstances where the discussion and disclosure would not and could not affect their actions, there may not be a sufficient degree of harm to competition. Of course, this is subject to the indication in T-Mobile (which I accept) that there may be distortion of competition if the undertakings can reduce uncertainty as to the timing, extent, or details of their strategy. All I say is that as a general point, market power (or lack thereof) may be relevant to the analysis.

245 In relation to market share specifically, I consider that it may similarly be relevant to the economic context, and the issue of whether there is a “by object” restriction of competition. I accept that the EU and English authorities have often rejected the role of market share as being relevant to the economic context. I also agree that it is important that there be no safe harbour for anti-competitive conduct. In that regard, the appellant has correctly referred me to the Polypropylene Decision and its related cases, ¹²⁶ to demonstrate that even small undertakings may be sanctioned for participating in a “by object” restriction of competition.

246 Again, I do not think that market share is always irrelevant. There may be cases where the undertakings’ small market share is indicative of a lack of market power, which might in turn indicate that there is no manifest anti-competitive economic rationale (or sufficient degree of harm to competition). Indeed, the Competition Appeal Tribunal in Lexon and Advocate General Wahl in the Cartes Bancaires Opinion left open the possibility that even in cases of price fixing, there may not be a “by object” restriction if the undertakings only have a tiny market share. I do not close the door on this possibility, though I admit that it might be a rarity. If undertakings wish to raise a similar point, they should be allowed to do so in representations, and before the relevant appellate fora if necessary.

247 The other specific issue raised in the present appeal is the extent of the consideration of the economic context.

248 The appellant’s submission is that this is in the nature of a “basic reality check”. ¹²⁷ This is in apparent contradiction to the finding in the Decision that the economic context must be “adequately evaluated”. ¹²⁸ I note that these are qualitative benchmarks, rather than quantitative requirements.

249 The first case I consider in determining the appropriate extent of the evaluation of the economic context is Budapest Bank. The appellant did not raise Budapest Bank itself, but the Budapest Bank Opinion. ¹²⁹ Specifically, the appellant cited the Advocate General’s view that the analysis of the economic context is a “basic reality check”. I note the potential merits of this view. Although this point does not appear to have been directly cited or adopted in the judgment itself, and the Advocate General’s statement was made in the context of agreements that have already been found to be generally anti-competitive, it has been cited in cases such as Lexon.

250 In any event, the Court of Justice did undertake a fairly extensive and detailed analysis of the surrounding context: Budapest Bank at [59] onwards, and in particular [80]–[85]. That being the case, I am unsure if the consideration of the economic context is necessarily, only, or always, a high-level one. As can be seen from Budapest Bank itself, and cases like Servier, the analysis has often been fairly comprehensive.

251 There have, however, been cases where the consideration of the economic context was relatively brief. These have generally been in cases where the anti-competitive object is clear.

(a) The most obvious example would be Toshiba. In that case, the Court of Justice permitted an analysis of the economic context that was limited to what was “strictly necessary”: Toshiba at [29]. However, it is worth noting that the Court of Justice had preceded this statement by commenting that market sharing agreements “cannot be justified by an analysis of the economic context”: Toshiba at [28]. Similarly, it was held in FSL Holdings that where an anti-competitive object was found based on price fixing, the consideration of economic context can be a limited exercise: FSL Holdings at [107].

(b) Another example is Balmoral Tanks. That case involved a cartel in the English market for cylindrical galvanised steel tanks. The cartel members had involved themselves in price fixing and dividing the market. The parties involved in that case were the only manufacturers in the English market: Balmoral Tanks at [1] and [11]. The consideration of the law on “by object” restrictions and the role of economic context was brief. This can be easily explained on the basis that the parties involved were the only manufacturers in the relevant market, and since the cartel had been in operation for some time, it would have been understood that any information provided by one undertaking would be used by the others. In any event, the undertakings involved were actively sharing information with each other: Balmoral Tanks at [77]–[88].

(c) Lexon purported to follow the “basic reality check” approach from the Budapest Bank Opinion. However, this meant in practice something a little more involved than in Balmoral Tanks (or in the ID, for that matter). It involved, among other things, the level of market concentration and competition, the nature of the product, and the downward pressure on prices (set out in full at [158] above).

252 Both Balmoral Tanks and Lexon adopted and applied the “basic reality check” approach to considering the economic context. However, the extent of the analysis involved very much differed between the two cases.

253 The cases listed above demonstrate that there is no particular standard of analysis involved when considering the economic context. Cases where the parties involved make up a large section of the market may only require a brief consideration before there is a finding of a “by object” restriction. Cases with a highly complex market or where the market share of the parties involved is not precisely ascertainable may call for a more detailed analysis.

254 The appellant has submitted that the consideration of the economic context should only go as far as what is strictly necessary, citing the Banco Opinion at [47]. I do not disagree. However, that statement was made (as the appellant correctly states) in the context of cases where the anti-competitive object is easy to perceive. As I have stated above, a finding of a “by object” restriction of competition is not based on obviousness per se. That means there will be cases where determining the existence of an anti-competitive object will necessitate a more involved consideration of the economic context. In such cases, a more involved evaluation will be “strictly necessary” to confirm or negate an initial finding that there has been a “by object” restriction.

255 This is consistent with the Budapest Bank Opinion at [49]. The appellant has cited the Budapest Bank Opinion for the proposition that consideration of the economic context was a “basic reality check”. ¹³⁰ That is incomplete. To set out the paragraph in full:

256 In summary, evaluating the economic context does not require a particular or universal quantitative standard. The language of “basic reality check” or “strictly necessary” might indicate that this is necessarily a brief consideration. However, this is not necessarily the case. To my mind, a more helpful standard is set out in Lexon at [231], that the real requirement is for “sufficient examination”. This makes clear that the requirement is not necessarily the same in every case, while also making clear that the competition authority is entitled in certain cases to undertake only a brief evaluation of the economic context.

257 I finally note the possibility from Toshiba that market sharing agreements cannot be justified by an analysis of the economic context. I have not been asked to decide when and whether it may ever be possible to dispense with an analysis of the economic context. I am also unsure whether the authority cited in Toshiba for this point really stands for the particular proposition. In any event, the parties in the present appeal have proceeded on the basis that some consideration of the economic context was necessary. In the circumstances, I say no more on this matter.

258 I turn now to the CAB’s decision that the conduct in the present appeal did not constitute a “by object” restriction of competition. I consider first if there is sufficient precedent or experience such that the conduct can be labelled a “by object” restriction. I then consider whether the conduct has a manifest anti-competitive economic rationale, such that it constitutes a “by object” restriction.

259 The first issue can be dealt with shortly. Exchanges of price-related information have been, and can be, classified as “by object” restrictions of competition. As I have stated above, the requirement is for there to be a sufficient body of experience pointing to the conduct being harmful to competition. That is the case here. I raise just three examples. The first is T-Mobile. That concerned a single meeting between mobile telecommunications network operators on standard dealer remuneration. The second is Lexon. That case involved several email exchanges over a period of 10 months: Lexon at [89]–[153]. The communications involved the exchange of a variety of sensitive information that would affect pricing: Lexon at [191]. The third case is Batam Ferries. That case concerned the exchange of price-related information between ferry operators on several occasions: Batam Ferries at [107]–[140]. All three cases resulted in a finding that there was a “by object” restriction of competition.

260 These cases demonstrate that exchanges of information, particularly that relating to prices, have been and can be classified as “by object” restrictions. These range from single meetings or exchanges, to drawn-out or repeated instances. Batam Ferries shows that such conduct has been deemed a “by object” restriction in Singapore.

261 I turn now to the second issue, of whether the information exchange in the present case constituted a “by object” restriction. I consider that the CAB was wrong to hold that the exchange of information in the present case did not constitute a “by object” restriction.

262 The conduct at the heart of the present appeal is the Communications. The Undertakings exchanged information on whether they would follow HSC and CLS in imposing the FTZ Surcharge.

263 As I have held at [207], the question of whether conduct constitutes a “by object” restriction of competition asks whether there is a manifest anti-competitive economic rationale. In my view, the Communications had a manifest anti-competitive economic rationale. The underlying and fundamental economic justification for the respondents’ conduct was anti-competitive. The respondents were party to a series of communications designed to determine if they, among other warehouse operators in Keppel Distripark, would raise their prices. In doing so, the respondents received information on whether other undertakings would be raising their prices.

264 This was not a purely hypothetical conversation, nor was it something natural and proper in the context of individuals and undertakings in the same space. The Communications came on the heels of the two largest warehouse operators raising their own prices. In ordinary circumstances (and subject to what I say below on the economic context), warehouse operators would have to determine whether they would similarly raise their prices. Alternatively, they could decide whether to maintain existing prices, and compete based on price.

265 I disagree with the submission by the respondents’ counsel, that this was merely “casual talk”. ¹³¹ By taking part in the Communications, the respondents obtained information on their competitors’ pricing strategies and intentions. They knew, or had indications, about what their competitors would do. In turn, they used that information to convince their clients to accept the imposition of the FTZ Surcharge. That does not comport with the conditions of normal competition.

266 An anti-competitive object can be found if the conduct will “directly or indirectly fix purchase or selling prices or any other trading conditions”: Dole at [124]. The conduct at the heart of the present appeal influenced the respondents to impose similar increases in price to their competitors. With this in mind, it would not matter if the FTZ Surcharge was economically necessary or justified. As seen from Gerhard Züchner, a concerted practice can constitute a “by object” restriction even if the undertakings have another justificatory reason, such as the offsetting of costs. The focus is on the fact of concertation and the harm to competition.

267 In addition, even if it is likely that many undertakings would have gone ahead with the price increases anyway, the fact is that the respondents obtained this information earlier than they otherwise would have. As can be seen from European Government Bonds, undertakings exchanging information on their strategies in advance of when they implement them can constitute a “by object” restriction. As stated in Banco, conduct can constitute a “by object” restriction where uncertainty is removed or reduced as to the “timing, extent and details of any future changes in the conduct of its competitors on the market”. The Communications went to the fact that other undertakings in the market would match the price increases that had been imposed by HSC and CLS. Even if there was no competition in relation to the fact of price increases (and this has not been proved to me), there might still have been competition in relation to the extent of the price increases.

268 Furthermore, there is no requirement that the respondents must have communicated with each other directly, or that all members of a concerted practice must have communicated with each other. As seen in JJB Sports, multiple undertakings may be sanctioned as parties to a concerted practice even if they do not communicate directly. In any event, I have not received any submission from the respondents to the contrary.

269 I now consider whether the appellant had had sufficient regard to the economic context. I consider that the appellant had done so, and that the CAB was incorrect to hold that there was insufficient consideration in the ID.

270 As I have held above at [256], a competition authority is required to undertake “sufficient examination” of the economic context. This does not entail any particular depth or breadth of analysis. As stated in the Budapest Bank Opinion at [49], the complexity of the analysis required will depend on all the relevant factors. In relatively straightforward cases, there would be nothing wrong with a competition authority considering the economic context briefly.

271 I turn now to the economic context as set out in the ID. The appellant directed me to [113]–[126] and [150]–[172] of the ID. I note the following points in particular:

(a) First, the respondents would seek their clients’ agreement before imposing new charges. ¹³² CNL explicitly stated that it had low bargaining power. The Notes of Information from Gilmon’s personnel tell a similar story. Mr Chua stated that Gilmon would follow the standard pricing in Keppel Distripark as it was not the largest operator. ¹³³ Similarly, Mr Teo stated that “Gilmon cannot start implementing a new surcharge as it does not have enough market power”. ¹³⁴

(b) Second, that the undertakings in Keppel Distripark would find it easier to convince their clients to accept price increases if their competitors were doing so. It was risky for undertakings to unilaterally impose price increases (unless they were large operators, which the respondents were not) as their clients would compare the rates offered by the different warehouse operators.

(c) Third, and relatedly, that the respondents knew that raising their prices could cause customers to switch to other warehouse operators. Knowing that their competitors would be imposing similar price increases gave the respondents certainty, and strengthened their bargaining power. ¹³⁵

272 The appellant also drew my attention to [188]–[189] of the ID. ¹³⁶ The appellant had considered and explained therein how the exchange of information on a shared or mutual intention to increase prices would reduce uncertainty, in the context of the particular market. In brief, the market was price sensitive, so the Communications would reduce undertakings’ uncertainty as to the risk of increasing their prices. ¹³⁷ The appellant also referred to its assessment at [248]–[255] of the ID that there was still uncertainty among warehouse operators about whether to impose the FTZ Surcharge. ¹³⁸

273 The Decision did not contain any pronouncement that the above analysis was incorrect. Rather, it took aim at the “disparate observations” and lack of a “coherent analysis” in the ID. ¹³⁹ I therefore approach the elements set out above on the basis that they are an accurate description of the circumstances, and consider only if the extent of the analysis was sufficient. Before turning to this task, I find it relevant that the present appeal concerned a homogenous product, in a fixed market (in the sense that there was no apparent way or prospect of expanding Keppel Distripark).

274 Having considered the analysis undertaken by the appellant in the ID, I am satisfied that the ID contained sufficient consideration of the economic context. It can fairly be said that the consideration of the economic context in the ID is more restrained than that in Lexon or Banco. Yet in the circumstances, the ID does enough. As I have said above, there is no particular quantitative requirement that a competition authority must undertake. The requirement is qualitative in nature.

275 The ID identified that the respondents did not have sufficient market power to unilaterally impose the FTZ Surcharge on their clients. The ID identified that there was uncertainty on whether to impose the FTZ Surcharge even after HSC had done so. The ID identified that the imposition of the FTZ Surcharge entailed risks for whichever undertaking chose to do so. Crucially, the ID identified that the Communications reduced the respondents’ uncertainty in this regard, and made it easier or more likely for them to impose the FTZ Surcharge on their clients. I therefore hold that the CCS had sufficient regard to the economic context when coming to its conclusions in the ID. The CAB erred in holding (in the Decision at [109]) that the consideration of the economic context in the ID was insufficient.

276 There is, however, some merit to the criticism in the Decision that the consideration of the economic context in the ID was “disparate”. Without the benefit of the appellant’s counsel’s skeletal submissions, it would have been more difficult to identify and assess the appellant’s evaluation of the economic context. At the same time, there is no need for a specific section on the economic context (though this would help with analytical clarity). All that is needed is that the consideration of the economic context can be identified and assessed by an appellate tribunal or court. In the present case, the ID contains sufficient analysis of the economic context, even if not presented optimally.

277 The CAB in the Decision and the respondents have both taken the position that something more must have been done by the appellant before finding that there was a “by object” restriction in this case. The trouble with such an argument is that what more might have been done has not been identified or explained. I asked the respondents’ counsel this during the hearing, and his answer was that for the second respondent at least, the appellant should have noted that the second respondent was operating in a different market to Mac-Nels and Penanshin. ¹⁴⁰ The appellant’s counsel argued that this had not been raised before, ¹⁴¹ but I find it tolerably clear from the respondents’ submissions to the CAB that such a point was raised. ¹⁴² However, this point was in the respondents’ submissions to the CAB, and the CAB did not then find for them on this point. In fact, the CAB’s analysis proceeded on the basis that the conduct in question between competitors did not constitute a “by object” restriction of competition. The CAB must be taken, even implicitly, to have accepted that Gilmon is or was operating in the same market as Mac-Nels and Penanshin. I decline to depart from this conclusion. To the extent that this is a question of fact, this court cannot disturb the CAB’s finding. To the extent that this is a question of law, Gilmon has not shown that the majority of its revenue being tied to a single client (which was at least part of the basis for the submission) means they were operating in a different market.

278 I have not been directed to any fact or factor that would indicate that in a market such as Keppel Distripark, an exchange of price-related information in a price-sensitive market would not have the objective aim or intention of preventing, restricting or distorting competition. On its face, if the economic context is such that operators and clients are conscious of price and sensitive to changes in price, undertakings exchanging information on their pricing strategy would very likely be anti-competitive. In the circumstances, the appellant was entitled to undertake a relatively brief consideration of the economic context.

279 Regarding the respondents’ lack of market power, I do not agree that this militated against a finding that the respondents’ conduct constituted a “by object” restriction. Rather than being a casual conversation which would have no impact on competition or the market generally, the exchange of information reduced the risk in imposing the FTZ Surcharge. The respondents’ contact, communication, and concertation were of the sort that could help them overcome their individual weaknesses, and get their clients to agree to the increases in price.

280 For the reasons above, I am satisfied that the respondents’ conduct constituted a “by object” restriction of competition. The CAB was incorrect to hold otherwise. I therefore hold that the CAB erred in finding that the conduct in the present appeal did not constitute a “by object” restriction. I also hold that the CCS had sufficient regard to the economic context in coming to this decision, and the CAB had erred in holding otherwise. It is therefore appropriate to set aside the Decision, and restore the finding in the ID that the respondents had infringed s 34 of the Competition Act.

281 The respondents have submitted that something more should have been done in the present case, but I have not been shown anything that was overlooked. The consideration of the economic context in the ID is sufficient for the analysis of the facts in this matter, and the nature of the anti-competitive conduct. That does not mean it will always be enough. As I have held above, there will likely be cases where a more complex analysis is called for. This is not one of them.

282 A “by object” restriction of competition is where conduct has a manifest anti-competitive economic rationale. This is objectively assessed, though a lack of subjective intention to prevent, restrict, or distort competition may be indicative of a lack of such rationale. In determining whether conduct constitutes a “by object” restriction, a competition authority must pay sufficient regard to the economic context. This is a qualitative, not quantitative, requirement, going only to the question of whether there is a manifest anti-competitive economic rationale.

283 In the premises, the CAB was incorrect to hold that the respondents’ conduct did not constitute a “by object” restriction of competition, and that the appellant had paid insufficient regard to the economic context. I allow the appeal against the Decision, and restore the finding in the ID that the respondents had infringed s 34 of the Competition Act.

284 What remains is the question of costs as well as whether any consequential or ancillary directions are required from the court. Parties have 14 days from the date of this judgment to raise any consequential or ancillary directions. Further, if parties are not able to agree on costs, I direct them to file written submissions on costs within 14 days of the date of this judgment, limited to 10 pages each. Unless requested to hear parties orally, I will proceed thereafter to decide on the incidence and quantum of costs on the basis of their submissions.