Commerce Commission v Bunnings Ltd [2021] NZDC 8918 (3 September 2021)

Last Updated: 19 December 2021

SUPPRESSION OF PARTS OF PARAGRAPHS [84], [105] AND [107] PURSUANT TO S 205(2)(e) of the Criminal Procedure Act 2011

IN THE DISTRICT COURT AT AUCKLAND I TE KŌTI-Ā-ROHE KI TĀMAKI MAKAURAU

CRI-2017-004-000242 [2021] NZDC 8918

COMMERCE COMMISSION Prosecutor

v

BUNNINGS LTD Defendant

Hearing:	12, 13, 14, 15, 16, 19, 20, 21, 22, 26, 27, 28, 29 April 2021 4, 5, 6, 7, 11, 19 May 2021
Appearances:	J Dixon QC, I Brookie and C Farquhar for the Prosecutor A Ross QC, T Lindsay and T Westaway for the Defendant
Decision:	3 September 2021

RESERVED DECISION OF JUDGE B A GIBSON

[1] The Commerce Commission has preferred charges against the defendant, Bunnings Ltd (‘Bunnings’) for breaches of the Fair Trading Act 1986 (‘the Act’) arising from a marketing campaign conducted in various media across New Zealand with the advertisements, it is alleged, conveying the impression that Bunnings offered the lowest prices in the market on all of its products. The Commission claims a number of price comparisons of its competitors show this was, for a significant number of products, not so.

COMMERCE COMMISSION v BUNNINGS LTD [2021] NZDC 8918 [3 September 2021]

[2] The charges span the period 17 June 2014 and 29 February 2016. They are laid as representative charges. Seven charges allege that Bunnings, being in trade, breached s 10 of the Act with 38 charges laid under s 13(g) of the Act, the s 13 complaints alleging false or misleading representations as to the price of its goods. The offences are strict liability offences. The only defence available is that the representations were not misleading and the statutory defence in s 44 of the Act.

[3] The s 10 allegations allege that Bunnings engaged in conduct liable to mislead the public as to the nature or characteristics of goods, namely their price. The Commission claims the conduct was conducted at Bunnings stores and was liable to mislead members of the public into thinking its prices were the lowest prices on offer in the market, when they were not.

[4] Claims by Bunnings, took a number of forms, but for the s 10 charge the particulars of the misleading conduct alleged were posted on interior and exterior signs and posters, as well as on staff kit, stated:

• (a) lowest prices are just the beginning ...
  
• (b) lowest prices everyday
  
• (c) lowest prices
  
• (d) lowest price guarantee
  

[5] The s 13(g) charges relate to advertising on Bunnings’ website, television and radio, and in newspapers and catalogues. The advertising was frequent and repetitive. The representations, alleging that Bunnings offered the lowest priced goods, which the Commission alleges were false or misleading as to their price were, by way of sample:

Website representations:

(a) lowest price guaranteed

(b) nobody beats our prices

get set for Christmas with lowest prices

(c) nobody beats our prices

tackle all your summer jobs with the lowest prices

(d) nobody beats our prices

lowest prices for big Easter projects

(e) nobody beats our prices

lowest prices for your winter D.I.Y

(f) nobody beats our prices lowest prices guaranteed

Print representations – taglines:

Newspapers for a number of charges included the tagline “lowest prices are just the beginning ...” without other misrepresentations.

Print representations – tagline, plus other representations:

Other newspapers included the tagline “lowest prices are just the beginning

...”, together with the following representations (sample only):

unbeatable prices

lowest prices to help you pick off those summer jobs nobody beats our prices here’s proof

nobody beats our prices lowest prices for the big jobs lowest prices on trade quality tools

lowest prices to help you tick off those autumn jobs something for every Mum at lowest prices

tidy up unfinished projects before winter with lowest prices we’ll guarantee the lowest prices in New Lynn.

Catalogue representations – again the tagline “lowest prices are just the beginning ...” were published in taglines and the Commission claims other representations also breach s 13(g) of the Act, some of which are as follows:

what every Dad wants at unbeatable prices nail those spring jobs with unbeatable prices unbeatable prices store wide

tackle all your summer jobs with lowest prices

lowest prices on all your D.I.Y. jobs

lowest prices for the big Easter projects lowest prices means more for Dad

[6] The scheme of the charges are that as well as being representative, they also represent quarterly intervals, covering the charge period from 17 June 2014 to 29 February 2016, effectively one charge per three month period with the exception of the first and last periods which are slightly different. The charges are broken down into six different types of representation by medium, namely website, print, print with other statements, catalogue, TV and radio.

[7] As well as the advertising tagline “lowest prices are just the beginning ...” most advertisements also contained a lowest price guarantee which the Commission claims that in conjunction with the other representations conveyed or reinforced that Bunnings had the lowest prices in the market.

[8] The wording of the lowest price guarantee (‘LPG’) for the charge period was:

If you happen to find a lower price on the same stocked item, we’ll beat it by 15%*.

* Excludes trade quotes, stock liquidations and commercial quantities.

Legal analysis

[9] It was not disputed that Bunnings, for the purposes of the section, was both a person and in trade in terms of the definition of the same in s 2 of the Act.

Charges under s 10 of the Fair Trading Act 1986

[10] Section 10 of the Act states:

Misleading conduct in relation to goods

No person shall, in trade, engage in conduct that is liable to mislead the public as to the nature, manufacturing process, characteristics, suitability for a purpose, or quantity of goods.

[11] The Commission alleges the public were liable to have been misled by the characteristics of the goods, namely their price, with that conduct said to have occurred at Bunnings stores, by misleading members of the public into thinking Bunnings prices were the lowest prices on offer in the market when the Commission maintains on a significant number of occasions they were not. Characteristics, as with the other matters the subject of the section are to be interpreted “in their natural and ordinary meaning”; Taylor Bros Ltd v Taylors Group Ltd1 and the price of goods has been accepted as a characteristic; Sound Plus Ltd v Commerce Commission2. In Ben Rumble Communications Ltd v Commerce Commission3 Chisholm J said, with reference to the definition of price that the ordinary dictionary meaning applied, “namely ‘money or other consideration for which a thing is bought or sold’ (see Shorter Oxford Dictionary)”.

[12] The phrase ‘liable to mislead the public’ is not defined in the Act. It is unnecessary that the conduct in fact misleads or deceives, just that it is liable to do so. In Red Eagle Corp Ltd v Ellis4, a decision concerning s 9 of the Act, the Supreme Court noted that whether conduct is liable to mislead the public is a question of fact for the Court to determine objectively having regard to the particular circumstances of the case.

[13] For the purposes of s 10, to prove that the representations as to price were liable to mislead the public, the Commission must establish, to the criminal standard:

• (a) Bunnings was a person within the meaning of the FTA;
  
• (b) Bunnings was in trade within the meaning of the FTA;
  
• (c) A significant number of consumers would have considered the representations amounted to a claim that Bunnings had the lowest prices in the market with some tolerance allowed because of the numbers of SKUs and the competitive nature of general
  

1 [1988] 2 NZLR 1 (HC).

2 [1991] 3 NZLR 329.

3 (1999) 9 TCLR 204 (HC), 213.

4 [2010] NZSC 20; [2010] 2 NZLR 492 (SC).

merchandise/home improvement retailing leading to Bunnings not being aware within a reasonable time that competitors had lowered their prices on specific items (described in the amended particulars as ‘rare exceptions’);

(d) Bunnings’ conduct was liable to mislead the public, namely a significant section of the relevant group of consumers;

(e) The conduct was in relation to a characteristic of its trade, namely price in that Bunnings did not have, other than in rare exceptions, the lowest prices in the market.

[14] By contrast s 13 of the Act, which prohibits persons in trade making a false or misleading representation with respect to the price of any goods or services, makes no reference to misleading the public. Conduct cannot be categorised as misleading for the purposes of ss 10 and 13 unless it contains or conveys a misrepresentation; Taco Co of Australia Inc v Taco Bell Pty Ltd.5 Neither party contended the claims were simply advertising puffery. Bunnings maintained they were not misleading with respect to the s 10 charges and not false or misleading for the purposes of the s 13 charges and that the Commission, in any event, had failed to establish beyond reasonable doubt, that the charges were proved.

[15] Representations claimed to be misleading need to be to the particular segment of the public targeted by those representations which the Commission claims are misleading. In Godfrey Hirst NZ Ltd v Cavalier Bremworth Ltd6 Wild J, delivering the judgment of the Court said, at para [50]:

We reiterate our response to the question “who is the consumer?”. “The consumer” encompasses all the consumers in the class targeted by the allegedly misleading representations, except the outliers as we have explained them in [20] above. Or, if the representations are made to the public at large, it is all the public except the outliers.

5 [1982] FCA 136; (1982) 42 ALR 177.

6 [2014] NZCA 418; [2014] 3 NZLR 611 (CA).

[16] “Outliers” was defined by the Court at para [20] as being “... consumers who are unusually stupid or ill equipped, or those whose reactions are extreme or fanciful.” And at para [47] “the gullible”. At para [48] “if you take all the “average” or “ordinary” or “reasonable” (or, as we prefer, all the typical) members of the public generally, or of any targeted class of the public, you will end up with all the consumers in the class targeted except the outliers ...”

[17] The test as to whether a representation is false or misleading is an objective rather than a subjective one; Marcol Manufacturers Ltd v Commerce Commission7 where at 507-508 Tipping J said:

Once the Court is satisfied, this being a point of law, that the material in question is capable of being seen as a representation and capable of being regarded as misleading, then it is entirely a question of fact as to whether a misleading representation has in all the circumstances been made.

And further, at p 508:

As I have already said, the question whether a representation is misleading is to be judged objectively. It is not a question whether someone has actually been misled, although proof of that may well be helpful for an informant. The question for the Court is whether the mind of the representee, in the present case the average New Zealand shopper, would be misled. In my judgment the test can be put this way: the informant must establish beyond reasonable doubt that the mind of the representee would be misled in the sense that the representation is apt to mislead.

[18] The Commission submits that for the s 13 allegations the representations were both false and misleading. For a representation to be false the ordinary meaning of the word applies, namely erroneous or incorrect, not in accordance with the facts. The Commission assert Bunnings was making an explicit representation in its tagline, the LPG and in the various advertisements referred to as particulars of the s 13 charges that it offered the lowest prices. It is unnecessary to prove a deliberate falsehood. It is sufficient if the representation does not accord with the facts and is incorrect; Rural Management Ltd & Rutherford v Commerce Commission8.

7 [1991] 2 NZLR 502.

8 (1997) 6 NZBLC 99, 401.

[19] Further, the proscription on conduct that is misleading or deceptive was recently referred to by Campbell J in Commerce Commission v Viagogo AG9 where at para [25] he said:

The leading case on s 9 is the Supreme Court’s decision in Red Eagle Corp v Ellis. The Court said that the section proscribes conduct that, examined objectively, is misleading or deceptive. The question is whether a reasonable person in the consumer’s position would likely have been misled or deceived. It is not necessary to show that the defendant had any intention to mislead or deceive.

Section 13 Fair Trading Act

[20] Section 13(g) of the Act states:

False or misleading representation

No person shall, in trade, in connection with the supply or possible supply of goods or services or with the promotion by any means of the supply or use of goods or services,–

...

(g) make a false or misleading representation with respect to the price of any goods or services;

...

[21] It is unnecessary to show any particular person has been led into error by the false or misleading statement; Commerce Commission v Amark Publishing (NZ) Ltd10. Falsity, as Mr Dixon QC submitted, focuses on the words of the statement that is made, citing Deane J in Thompson v Riley McKay Pty Ltd (No.2)11 where it was said:

There is not, however, implicit in the word ‘represent’ any requirement that the representation actually reach, or be understood by, the intended representee. The act of representing is complete once the subject matter is irrevocably set forth or disseminated upon the course which is intended to lead to the intended representee or representees. A statement of fact made in ordinary conversation may constitute a representation notwithstanding the circumstance that the attention of the other party to the conversation was distracted and he either did not hear the statement that was made or, hearing it, did not appreciate its content. A statement of fact made on a billboard on the side of a busy highway will constitute a representation to passing motorists from the time it is exhibited notwithstanding the circumstance that it so

9 [2021] NZHC 1398.

10 (1989) 3 TCLR 567 (DC).

11 42 FLR 279 (FCA) at 289.

happens that, in the event, all passing motorists steadfastly keep their eyes upon the road ahead. A statement of fact made in the course of an advertisement on television will constitute a representation notwithstanding the circumstances that it is impossible to say whether any particular viewer actually watched it. A statement of fact made in the pages of a newspaper or journal offered for sale and sold at retail outlets will constitute a representation notwithstanding the circumstances that it is impossible to say whether the statement was read by any reader or by any particular reader.

[22] A false or misleading representation subject to s 13 of the Act is a strict liability offence. There are no shades of falsity; Megavitamin Laboratories (NZ) Ltd and Stewart v Commerce Commission12. For the purposes of these charges the elements the Commission must prove are that the representation the subject of the particular charge was made and that it was untrue, erroneous or incorrect; Commerce Commission v A & W Hamilton Ltd13, therefore a false or misleading representation by Bunnings as to the price of its goods.

Background to the charges

[23] Bunnings is a wholly owned subsidiary of West Farmers Ltd, an Australian public company. It commenced operating in New Zealand as a branch of the Australian business approximately 20 years ago and now has 50 stores of which 25 are large warehouse type stores, called by Bunnings ‘big box stores’, 19 smaller formatted stores and six stores dealing solely with tradesmen. Overall in New Zealand it had, in 2016, a turnover of approximately one billion dollars and approximately 3,785 staff.

[24] Bunnings offers for sale approximately 62,000 items, known as SKUs, or stock-keeping units. Its larger stores have in stock for sale up to 42,000 SKUs. Bunnings has a number of competitors in New Zealand, among whom are Placemakers, ITM, Hammer Hardware, and Mitre 10 (New Zealand) Ltd (‘Mitre 10’) which in 2016 operated 82 or 83 stores. A number were large format stores similar in size to Bunnings mega stores. Mitre 10 is Bunnings’ main competitor in New Zealand. Bunnings operates on a corporate model, owning all of its stores. Mitre 10 is a member owned co-operative with the store owners holding shares in Mitre 10 and with a support centre / head office based in Auckland.

12 (1995) 6 TCLR 231,242.

13 (1989) 3 TCLR 398, at 402.

[25] Like Bunnings, Mitre 10 carried a number of propriety and exclusive brands amounting to approximately 35 to 45 per cent of its SKUs. It operated in the same market as Bunnings, the home improvement market and as a supplier of building supplies to tradesmen.

[26] Bunnings for many years prior to the laying of charges built its marketing strategy around its position as a low priced retailer known in the trade and among marketing experts as an EDLP retailer (everyday low prices). This is a term used to denote retailers who tend not to use price fluctuations on products through sales or special discounted offerings. Their prices tend to be consistent.

[27] From 2002 Bunnings’ marketing model involved offering what it described as “lowest prices” together with a lowest price guarantee which advised customers that if they “happen to find a lower price on the same stocked item” elsewhere then Bunnings would beat its competitor’s price by 15%.

[28] Bunnings was not the only retailer using as a superlative the word “lowest” in advertising prices. For several years Mitre 10 also advertised its products using the same or a similar tagline. At some point the Commerce Commission took an interest in the veracity of these claims and on 4 May 2011 it wrote to both Bunnings and Mitre 10 warning them that their claims to have the “lowest prices” could constitute a breach of the Act.

[29] Specifically the letter to Bunnings said that it considered the lowest price representations were liable to mislead the public in contravention of s 10 of the Fair Trading Act and, secondly, that the representations also appeared to the Commission to contradict s 13(g) of the Act in that they contained misleading representations. The letter stated:

Lowest priced representations.

The Commission considers that there are two ways in which the lowest price representations can be interpreted:

(a) As an absolute claim that each item offered for sale is offered at a price lower than the price offered by competitors; or

(b) As a claim that, across its product range, a stores prices are generally lower than its competitors.

Which of these meanings will be preferred by consumers will depend on the nature of the goods being sold and the actual words used in the claim.

In this case, the Commission considers that the first of these interpretations is most consistent with the claims made in Bunnings advertising. It is more likely to fit the nature of shoppers at a hardware/home improvement store who are likely to visit to pick up supplies for a specific project, especially when seen in the context of Bunnings advertisements.

[30] Mitre 10 chose to discontinue its advertising campaign based around a lowest price tagline but Bunnings did not. In late 2014 Mitre 10, concerned that Bunnings was continuing to claim that it had the lowest prices on its products when its own prices on identical products were often lower, complained to the Commerce Commission. Other than this complaint, from its main competitor, there was no evidence of any complaints about Bunnings by members of the public generally or by any other competitor, somewhat remarkable given the length and prominence of the advertising campaign. Mitre 10’s complaint led the Commission to open an investigation and in January 2015 it conducted two informal price checking surveys, using Commission staff, in the Wellington area. Staff visited and checked prices on specific items at Bunnings stores in Lyall Bay, Wellington and at Porirua and then compared them with competing stores. The investigators, headed by Mr WJH Lourie, a principal investigator in the competition branch of the Commission, also examined and photographed the signage on the outside of the two Bunnings stores including the tagline “lowest prices are just the beginning ...” prominently displayed over Bunnings corporate colours (red and dark green) on the outside walls of these stores.

[31] As a result of that it commissioned a survey company, Perceptive Insight Ltd (‘Perceptive’), to conduct a price comparison survey between Bunnings shelf prices and those of its competitors. The survey results were provided to the Commission on 9 February 2016. An interview was conducted by Mr Lourie with representatives of Bunnings on 19 February 2016 and on 23 March 2016 the Commerce Commission wrote to Bunnings about its representations, saying “essentially all of the representations are about Bunnings offering its goods for sale at the lowest prices”. It claimed the representations were at risk of being false or misleading as Bunnings prices were not always the lowest and drew Bunnings’ attention to a number of surveys

including Bunnings own data obtained from HOED Mystery Shopping NZ Ltd (‘HOED’), a survey company, which Bunnings gave to the Commission and which showed that on average Bunnings did not have the lowest prices on 18.6% of the products surveyed. The Commission stated in its letter 23 March 2016 as follows:

Price surveys found that:

10.3.1 Bunnings did not have the lowest prices on 32% and 41% of products surveyed in the Lyall Bay and Porirua areas.

10.3.2 Bunnings did not have the lowest prices on 33%, 32%, and 20% of products surveyed in the Auckland, Wellington and Christchurch areas.

10.3.3 A retailer provided the Commission with survey data which showed that approximately 32% of the surveyed products were cheaper or the same price at this other retailer, when compared to prices advertised by Bunnings.

[32] That other retailer was Bunnings’ main competitor, Mitre 10 which had undertaken its own comparison survey of prices for products offered by it and Bunnings for sale to the public and which had provided the data it collected to the Commission to support its complaint.

[33] The Commission also advised Bunnings that it considered the LPG to be misleading as it essentially communicated to the customer that it would be a “rare event for Bunnings to be beaten on price”. Consequently Bunnings was requested to stop advertising using these claims.

[34] Bunnings did not do so. It wrote in May 2015 to say its own internal reviews and a review conducted on its behalf by HOED indicated that Bunnings offered the lowest prices on at least 90 per cent of the products tested since December 2014 and that Bunnings own internal monitoring of competitor’s prices, and the steps it would take to adjust prices downwards when it found a competitor was lower as well as the LPG allowed customers to be confident that Bunnings advertising was not false or misleading.

[35] An interview of senior Bunnings employees including its New Zealand General Manager, and its general counsel and external lawyers which took place in February 2016 at the Commission’s Wellington office revealed that Bunnings had for a number of years built its marketing strategy around the tagline ‘lowest prices are just

the beginning’ meaning that consumers could expect to get lowest prices as a matter of course. The importance of the price guarantee (LPG) was also emphasised by Bunnings Legal Manager, Ms Mustow. The Chief Financial Officer addressed the centrality of the marketing campaign stating that Bunnings held itself to a standard that its prices are the lowest prices. Various participants at the meeting emphasised, on behalf of Bunnings, that the intention was the company would offer the lowest prices across its range of products. The New Zealand General Manager said there were over a million transactions in December 2018 alone. The LPG was said by her to be there to enable Bunnings’ customers to trust it and was seen as a ‘backstop’. In any event only 0.11% of Bunnings overall customer sales resulted in customers returning after identifying a lower price elsewhere to claim on the LPG. The overall percentage of claims under the LPG was said to have dropped both in Australia and New Zealand over the preceding 10 years and was attributed to Bunnings policies which had ensured the company was able to offer the lowest prices to consumers.

[36] Bunnings adopted a number of ways to implement its policy of having lowest prices available to the consumer. Firstly it reviewed its prices twice a year, in January and June, as well as testing pack sizes to ensure that matches with other competitors were ‘like for like’. Secondly, it monitored its competitors’ prices by using external surveyor agencies such as HOED. In addition Mr Samuels, the Merchandise Planning Manager of Bunnings responsible for merchandise analysis and reporting gave evidence that once Bunnings learned a competitor was offering a lower price for a product it would either initiate a temporary price reduction (‘TPR’) or a permanent price change after verifying that the product offered for sale by the competitor was identical or ‘like for like’ to the Bunnings product. The change would always be to ensure that Bunnings could offer the lowest price for the particular product.

Expert evidence

[37] There was no survey evidence of consumers’ understanding of the representations made by Bunnings through its advertising, or, for that matter evidence of any consumer complaints to the Commission about Bunnings prices in relation to their advertising or of being misled by it.

[38] Evidence of consumers understanding is admissible as proving a public state of mind; Auckland Regional Authority v Mutual Rental Cars (Auckland Airport) Ltd14. Survey evidence can be of assistance but its absence is not fatal to a prosecution; Geddes v New Zealand Dairy Board15. Dr P J Gendall, an expert called by the Commerce Commission accepted that consumer survey evidence was the most reliable way of establishing what consumers would think about the meaning of the advertising. He considered it would have been helpful in the case but did not consider in his role as an expert witness it was for him to suggest to the Commission that it obtain one. He said it would have been possible to design a survey helpful to the Court and an understanding of what was in consumers’ minds could have been obtained so that it would have been “a halfway house between the parade of witnesses in court and the survey”.

[39] Dr B U Lang, another expert who gave evidence for the Commission, was not as definite, saying that academic research was moving away from surveys but it might have been beneficial as a further angle to consider the issues of what consumers would take from the advertising. Nevertheless while not necessary, a consumer survey may have been of assistance in considering whether the advertising the public was liable to mislead or were misleading, an essential ingredient of the charges, and by giving the experts a platform on which to ground their views.

[40] Dr Gendall, a Senior Research Fellow in the Department of Marketing at the University of Otago and for 20 years Professor of Marketing at Massey University in Palmerston North, opined that the overall impression conveyed by Bunnings advertising and marketing including the lowest price guarantee was that it offered the lowest prices in the market, across all of its products and at all times, other than on rare occasions when it had not been able to react to a sudden drop in price on a particular product. His evidence was supported by Dr Lang, a Senior Lecturer and former Head of Department in the Marketing Department at the University of Auckland. Dr Gendall said the claims of having lowest prices across all of its products is reinforced or strengthened by the LPG which gives the impression, through its wording, that the possibility of a consumer finding lower prices elsewhere is unlikely.

14 [1987] NZHC 213; [1987] 2 NZLR 647.

15 CA 180/03, 20 June 2005 at [94].

[41] The defendant also called expert evidence as to the meaning conveyed to consumers by Bunnings advertising.

[42] Professor A Stephen, a Professor of Marketing at the Saïd Business School, University of Oxford, England and an editor of leading academic journals in the field of marketing gave evidence on behalf of the defendant. Much of his published research and academic life concerned how consumers react to advertising and marketing. He opined that consumers would not see the taglines that Bunnings used promoting it as a retailer offering lowest prices when coupled with the LPG on the same stocked item as indicating anything other than that it was an everyday low price retailer. Taglines are branding devices, simple statements saying something about the brand and its position relative to competition. They rarely have a precise literal meaning to every product sold by a retailer.

[43] Professor Stephen said if consumers found that on particular items Bunnings were not the lowest priced in the market, the prominently displayed LPG was there to ensure its prices actually were the lowest by discounting a competitor’s lower price by 15%. He said the presence of the LPG would lead a consumer to think that while Bunnings would endeavour to have the lowest prices then if in particular instances they did not the LPG would, in effect “make it right”.

The relevant class of consumers

[44] Bunnings has a large number of stores throughout New Zealand and advertised regularly, during the relevant period, on television and in other media outlets such as daily newspapers as well as delivering catalogues to households. It spent a substantial amount on advertising, some $22 million to $26 million each year during the charge period, an amount not dissimilar to many of its competitors and several million less than its main competitor, Mitre 10. All of the advertising was emblazoned with Bunnings taglines and with the LPG prominently displayed in most cases.

[45] Professor Stephen’s view was the class of consumers targeted by Bunnings advertising were normal adults throughout New Zealand. However Dr Lang had a reduced view of the consumer class targeted and affected by Bunnings advertising,

with which I agree, as those interested in ‘do it yourself’ and home improvement products as well as general ‘living’ products such as outdoor furniture, plants and the like. His view was that this represented a broad cross-section of the community. There would be, he said, few consumers who would not be interested in some of the products offered for sale.

[46] For the purposes of assessing whether consumers in the relevant class were misled by the advertising the law assumes the consumer has seen and understood the advertisement; Thompson v Riley McKay Pty Ltd (No.2)16

[47] Two issues require determination. Firstly whether the advertising amounted to a particular representation and secondly, whether an average member of the consumer class targeted would derive from that a message which is liable to mislead or to which is misleading; Marcol Manufacturers Ltd v Commerce Commission17 where Tipping J added “the mind of the representee is likely to work more by impression than analysis and to be prone to some looseness of thought”. He reasoned in a way that argues against a literal interpretation of the words in a representation of the type caught by the legislation by analogy to the way the courts interpret innuendo in the law of defamation referring to the speech of Lord Reid in Lewis v Daily Telegraph Ltd18 where it was said:

There is no doubt that in actions for libel the question is what the words would convey to the ordinary man: it is not one of construction in the legal sense. The ordinary man does not live in an ivory tower and he is not inhibited by a knowledge of the rules of construction. So he can and does read between the lines in the light of his general knowledge and experience of worldly affairs.

The representations – the experts’ views

[48] The Commission alleges that Bunnings, through its advertising, including its various taglines, claimed it offered the lowest prices in the market with rare exceptions. This it claimed was liable to mislead or was false or misleading. Bunnings own internal policies and processes repeatedly referred to a policy of ensuring lowest prices for its products every day. It sought to have its employees

16 supra at 289.

17 [1991] 2 NZLR 502, 507.

18 [1963] 2 ALL ER 151.

monitor its competitors’ advertising and lower prices whenever it seemed it was beaten on price. It stated in its various communications with the Commission that its policy was to deliver the lowest prices on its products to consumers and did that with rare exceptions, the genesis of the amendment of the particulars for each charge to include that phrase. It saw the LPG as providing the necessary ‘backstop’ protection for customers should a competitor lower its prices on goods before Bunnings could reasonably become aware of it. Use of the LPG was seen by consumers, it believed, as being the exception rather than the rule as Bunnings policy of lowest prices would largely mean it was unnecessary.

[49] At the interview with the Commission on 19 February 2016 one of Bunnings’ employees said they would expect it to be “a rare event, a relatively rare event”, that a lower price was available elsewhere. Dr Gendall adopted this view, namely a rare circumstance was one where Bunnings was unaware a competitor had lowered its prices. Although he accepted Bunnings could not be expected to monitor approximately 60,000 SKUs on a daily basis he said consumers would expect Bunnings was monitoring its competitors’ prices all the time and so only an occasional one would slip through the net, hence it would be rare. He said once Bunnings became aware of a price change in a competitor’s product then, if necessary to maintain its claim to lowest prices it needed to react. Dr Lang attempted to prescribe a threshold, namely 5% as being an acceptable variation. Dr Stephen considered ‘rare exceptions’ as being something less frequent rather than rare.

[50] The views and opinions of Bunnings’ employees as to the representations that it had the lowest prices in the market with rare exceptions are not determinative as the test is an objective one which the Court must itself determine in terms of the s 10 charges as to what constitutes the representations and whether they were liable to mislead the public, or is false or misleading for the s 13 charges; Taco Co v Taco Bell19. Mr Dixon QC submitted evidence from Bunnings employees as to its policy is relevant in that the Court may be able to draw inferences, if appropriate, from it. He cited in support of that proposition a decision of the High Court of Australia in Australian Competition and Consumer Commission v TPG Internet Pty Ltd20, the inference said

19 supra at 202.

20 [2013] HCA 54; (2013) 304 ALR 186.

to be “... where a representation is made in terms apt to create a particular mental impression in the representee, as it is intended to do so, it may properly be inferred that it has had that effect”.21 Gendall J in Financial Markets Authority v Carrington Securities LP et al22 after referring to the summary of principles to determine whether conduct is actually or potentially misleading or deceptive as set out in Taylor Bros v Taylors Group Ltd23 said at para [25]:

Conduct that is misleading or likely to mislead so as to amount to a false representation is a question of fact to be decided objectively, the intention of a defendant being irrelevant, although if existing may be persuasive. What is necessary is there be a likelihood or reasonable possibility that persons to whom the offer is directed will be misled.

[51] Whatever inferences are drawn from the views expressed by Bunnings’ employees’ interviews and the various documents and policies referred to in evidence they are not determinative but may, as Mr Dixon QC submitted, assist in the Court in forming an objective view as to whether the advertising was liable to mislead or was false or misleading. Both parties called expert evidence to address the issue as to what representations were created by the taglines or headline representations and the LPG. Dr Gendall’s view of the taglines ‘lowest prices are just the beginning’ and ‘lowest prices every day’ emblazoned during the charge period on both the outside and the inside of Bunnings stores and the headlines to similar effect in its advertising was that they were extremely powerful in terms of their impact on consumers. His view was that the tagline and the advertisements with their headlines coupled with the LPG suggested to consumers that Bunnings would be offering lowest prices “plus something else” across the range of its products other than in the rare exception Bunnings had not done so for a particular product as it was unaware a competitor’s price were such it meant it was now offering a price lower than Bunnings.

[52] Headline statements were described by Professor Stephen as being “a shorthand for consumers that gets them thinking as opposed to telling them exactly how something is”.

21 At p 198 para [55].

22 Christchurch High Court, CIV-2011-409-000435, 12 May 2011.

23 supra at [24] and [25].

[53] Dr Gendall said the LPG itself reinforced the representation to consumers as the way in which it was worded “if you happen to find a lower price ...” suggested to consumers that would be “pretty unlikely”, presumably therefore dissuading the customer from searching for lower prices elsewhere. His view was that the statements were ‘concrete’ or specific as they concerned the lowest prices tagline and headlines given they were empirically testable. Notwithstanding the representations were expressed in superlative terms, he said they created the impression in consumer’s minds that Bunnings offered the lowest prices across the board and all of the other messages to similar effect in stores and in the advertising together with the LPG supported that proposition. Support could also be derived from the substantial monies Bunnings spent on advertising. While accepting that consumers form overall or general impressions from advertising he said in this case the messages promoted were quite specific and if consumers failed to take a literal meaning from them it represented a catastrophic failure on the part of Bunnings’ marketing department and its advertising agency. That of course pre-supposes the representations were as the Commission and Dr Gendall say they were and were taken literally by consumers.

[54] Dr Lang, who is employed in the Department of Marketing at the University of Auckland and also had experience gained in running a large market research company, gave evidence as to the impression consumers would take from Bunnings advertisements, mainly from the headlines in the advertisements and the tagline “lowest prices are just the beginning ...” and other variations of the same. He opined the advertisements would be taken literally by consumers who would as a result consider that Bunnings offered the lowest prices on all of their products across its range. He said there was no qualification expressed in the statements, tagline or in the headlines in a number of advertisements, particularly those used in television advertising where selected employees would appear and assert lowest prices applied to everything in the store, as well as headlines in advertisements expressly mentioning that lowest prices applied across all products in all aisles which would lead consumers to the impression that all products in Bunnings stores would be the lowest in the market. He would allow for a small number of products, 5% or less, where a consumer would understand that Bunnings were not able to match prices quickly enough or be the lowest, although he did not attempt to articulate the exception as Dr Gendall did when he adopted Bunnings own view of the occasions when it would not have the

lowest prices, but accepted that other than on a small number of products within this percentage consumers would understand Bunnings advertising to mean that its prices were the lowest across its entire range.

[55] Dr Lang accepted that consumers would not know how many SKUs constituted the entire range of products but his view was that the repetitive nature of the advertising, the simplicity of the message with the tagline “lowest prices are just the beginning ...” which he described as the spine of the advertising campaign and the millions spent on marketing meant that consumers would believe that Bunnings had the lowest prices across the range of its products. While he accepted there was no ‘hard proof’ that consumers would conclude as a result of the frequent and repetitive advertising that Bunnings was the lowest price retailer, he concluded it was a reasonable assumption to make. His conclusion was that the evidence, specifically the survey evidence of Bunnings’ prices, demonstrated that the claims were untrue and he said the four different surveys, although not conducted scientifically, all pointed to a conclusion that the real proportion of products sold by Bunnings that could be bought cheaper elsewhere lay somewhere in the range of 20 - 30%. He did not attempt to consider the nature of the information gathering for the price surveys and any possible errors in each but accepted them uncritically as evidence supporting a range of higher prices where Bunnings was higher by comparison with its competitors. He described the process of comparing the four surveys and drawing a conclusion as one of triangulation saying that where there are multiple items of evidence pointing the same way, then scientifically, one can be more certain about what the actual result is in reality.

[56] As for the LPG he observed it was generally reasonably prominently displayed and he expected it would be seen by consumers. He said it was not just a price matching guarantee but rather a price beating one. It was at the upper range of discounts offered for these types of guarantee and it operated to support the message that Bunnings had the lowest prices because of the broad nature of the statement and the absence of qualifications other than “same stocked item”. Consequently he said it operated as a reinforcing mechanism leaving the consumer confident that Bunnings would have the lowest prices and signalled to the consumer that Bunnings was so confident of its prices that it was willing to “bet against themselves”, describing the

message in the LPG as “oozing confidence”. He saw it as a way of reinforcing Bunnings’ campaign claiming it had the lowest prices.

Professor Stephen’s evidence

[57] Despite Dr Gendall’s evidence that consumers would take a literal meaning of the tagline “lowest prices are just the beginning ...” and the LPG he accepted under cross-examination that on their own these types of taglines were not misleading, a view with which Dr Lang, who expressed his view of the tagline and LPG in literal and absolutionist terms, did not agree. Dr Gendall said they were only misleading when coupled with the headlines in the advertisements as he said those headlines make it clear what the promise in the tagline means. Professor Stephen’s evidence was that taglines are simply a branding device, a proposition I did not understand Dr Gendall disagreed with provided they were seen in isolation. Professor Stephen said they were not to be seen as precise literal meaning type statements. In his view the tagline was not saying anything concrete, such as “this is the price of something” and so the tagline would be seen by consumers as holistic in establishing Bunnings’ brand position rather than as a specific representation. Professor Stephen, as Mr Ross QC submitted, is a world authority on consumer psychology and behaviour, viewed the research of the weight consumers put on taglines and LPGs as suggesting they play a relatively small role in driving consumer behaviour and it was difficult to say how much price-related claims influence individual consumers because there are many other factors influencing consumer behaviour such as the layout of a shop, good service, parking and the like.

[58] Professor Stephen said that a typical consumer would see the tagline “lowest prices are just the beginning ...” as an abstract general statement of policy ex ante and nothing more. The same would presumably apply to the other claims the subject of the s 10 charges re lowest prices every day, lowest prices and lowest prices guarantee. He saw the tagline and the LPG as indicating to the consumer that Bunnings were a “value retailer”, namely a retailer that wanted to give consumers a “fair deal”. The advertisements with the headline representations backed this up by suggesting to consumers that there was more to Bunnings than merely lowest prices, namely that this was a retailer which cared about having low or lowest prices and that the price

beating nature of the LPG confirmed to consumers that Bunnings overall pricing policy was about ensuring consumers would get a “fair deal”. He saw the headline statements, when used in addition to the tagline statements such as “Everything you need for your Easter projects at lowest prices” and similar statements in television advertisements, as speaking to the broader area of value, which for a retailer such as Bunnings with its wide range of products alerts the consumer to what is available and is not something related to price. He said academic literature refers to this as brand advertising. He said the wording of the LPG reinforced this image to consumers and also alerted them to the fact that Bunnings would not necessarily always have the lowest price on the same items stocked by its competitors but that it would deliver the lowest price through its price beating guarantee.

[59] Professor Stephen argued against a literal interpretation of the tagline and headlines, saying that consumers were not naïve, they used their common sense and were generally aware of the large number of SKUs stocked by Bunnings and its competitors and that it was “sort of preposterous” to believe that a system could be designed that would be 100% perfect at all times by always ensuring the lowest price on every item was available. He said consumers are also aware of the competitive nature of the particular industry Bunnings, Mitre 10 and others operate in, again leading consumers to treat the taglines, in particular, as more of a holistic expression of Bunnings’ price policy rather than a specific representation. He said a typical consumer would treat the “lowest prices are just the beginning ...” tagline as a “more abstract general statement” which, if they were interested, might motivate a store visit or a search.

[60] None of the experts considered that consumers would spend much time processing the taglines and headlines and the claim by Bunnings for lowest prices. Dr Gendall considered, based on work by Professor A Durant of the University of Middlesex, that while consumers might be literal in their interpretation they also take a common sense approach and consider advertising and marketing messages mean what they say without giving the claims too much thought and would think that if the claims were not true a third party agency, such as the Commerce Commission, would stop them from being made.

[61] Dr Lang adopted a more prosaic approach saying the type of products offered for sale by a retailer such as Bunnings meant that consumers were less motivated in dealing with items that are relatively cheap and tend to have low involvement in the decision making process. Accordingly they are inherently less likely to process the advertising claims prior to purchase.

[62] Professor Stephen agreed, saying that most consumers are not deconstructing advertisements and analysing them succinctly. Television advertisements, he said, generally only receive passing attention and, as with Dr Lang, the more expensive the item being purchased, such as for instance a barbecue or set of outdoor furniture, the more likely consumers would examine alternative pricing for the product from other competitors becoming, as Dr Lang described them, high involvement consumers. Consumers, Professor Stephen said, take advertising claims with a degree of scepticism.

[63] While Professor Stephen accepted consumers would not expect Bunnings to have a system that was 100% accurate in ensuring the lowest price on every item vis-a-vis its competitors, he accepted that consumers would expect Bunnings to be taking account of its competitors’ prices and would have good systems to enable it to do so, one example reinforcing this being Bunnings practice of placing its competitors’ catalogues on a board with Bunnings’ prices shown to its advantage, marked alongside.

[64] Nevertheless he said it was impossible to set the lowest prices in a competitive environment as Bunnings, at any given point in time, would not know what its competitors intended or were doing. It could only be a process of applying its clear policy of setting lowest prices and catching up once it knew of a competitor’s price and then lowering its own prices to beat them. Where he saw the LPG as giving effect to Bunnings policy was that while Bunnings could not possibly expect to know its competitors’ prices on every single item at every instance in time the LPG was in effect a type of insurance policy as well as being a branding device underpinning its claim of lowest prices. It also signals to customers that at times Bunnings does not have the lowest prices as if they were absolutely certain they did there would be no need for an LPG.

[65] Further academic literature supports this point as being a statement of the obvious, as for instance, in an article by S Baglione, L Tucci and J Talaga entitled “Consumer Preferences Among Low-Price Guarantee Offers” where it was said:24

“With price matching and price beating guarantees a retailer is effectively adding a disclaimer that the store may not offer the lowest prices.”

[66] The LPG, as Professor Stephen said, was very prominently displayed and served “to very clearly signal in my view that there is meat on the bone to back up the tagline”.

[67] The experts, including Professor Stephen agreed that Bunnings, through its integrated marketing campaign, was attempting to convey to consumers a message that it had the lowest prices on everything other than on occasions when it had not become aware that competitors had changed pricing on the same items it sold itself so as to render them lower than Bunnings, being the Commission’s “rare exceptions”. Certainly that was the thrust of Bunnings’ advertising policy and the marketing advertising campaign reflected that. The difference between the experts was whether consumers would interpret Bunnings’ claims in that way and what were an acceptable percentage of the same items that were sold elsewhere for prices lower than Bunnings.

[68] Dr Gendall agreed the taglines in themselves were not representations and would be seen only as such when combined with headlines, which he described as being extremely powerful in terms of their impact on consumers. By ‘headlines’ he meant claims such as Bunnings guaranteeing it had the widest range of product or the top quality brands, all for the lowest price. He said marketing research indicated that consumers often just read the headline and not the rest of the advertisement and that the headlines, combined with the taglines and the LPG would lead consumers to take the marketing messages from Bunnings more or less at face value, as Bunnings plainly intended.

[69] Dr Lang also accepted, as did Professor Stephen, that the thrust of the marketing and advertising campaign mirrored Bunnings’ intention and policy of having the lowest prices on everything. But, unlike Doctors Gendall and Lang,

24 infra at p 21.

Professor Stephen in my view correctly, considered that consumers, objectively viewing the taglines and various headlines, would not necessarily interpret them in the way contended. All experts agreed that consumers would expect Bunnings’ prices to be substantially the lowest, Dr Gendall using the ‘rare exceptions’ phrase derived from the Bunnings interview, Dr Lang formulating a non-trivial amount of 5% or less as the threshold, with Professor Stephen suggesting, given the size of the assortment and number of SKUs, that consumers would expect more often than not Bunnings would be the lowest price saying also that Bunnings worked hard and used sophisticated systems and policies to achieve its stated goal of offering the lowest prices.

[70] Professor Stephen’s evidence was that even on the worst scenario in the four price comparisons used which suggested Bunnings had the lowest prices approximately 70% of the time, given the vast number of products, the size and layout of the store, the number of different categories, the network and the complexity and monitoring each and every product price on a daily basis, having the lowest price 70% of the time, if that was an accurate statistic, was still a very good performance. Given the nature of the industry in which Bunnings and others operate it was, he said, impossible to set lowest prices in a competitive environment, which seems self-evident considering the number of products the large stores of the main competitors in the industry sell, the highly competitive nature of the industry and the active monitoring of prices.

[71] He accepted, given the lowest prices claim, consumers might be concerned if Bunnings had the lowest price only 70% of the time if the LPG was not part of Bunnings’ branding and advertising but that consumers would consider given Bunnings were the lowest the majority of the time, on those occasions where it was not, the LPG remedied the position.

Does the Advertising amount to a representation?

[72] The thrust of the Commerce Commission’s case is that a significant number of the class of consumers targeted by Bunnings would have considered Bunnings offered the lowest price on each individual product in all of its stores all of its times within the limits of rare exceptions as already noted. Where there are representations conveyed

by the advertising these show that Bunnings is asserting that on the whole it has the lowest prices in the market across the range of its products but there will be occasions when that might not necessarily be so. In marketing language Bunnings was representing to the relevant class of consumers, namely those interested in ‘do it yourself’ and home improvement products as well as ‘general living’ products such as outdoor furniture and the like that it was an everyday low price retailer.

[73] Neither do I accept Dr Gendall’s evidence that the words used in the LPG “if you happen to find a lower price ...” suggests to consumers in the relevant class that it was unlikely they would find lower prices elsewhere therefore reinforcing a representation claimed to have been made that Bunnings offered the lowest prices across the board. The words are expressed in colloquial language which does not carry with it the meaning Dr Gendall asserts. The phrase presupposes that consumers may find lower priced items for the same range of products at other stores and does not convey any message that this will occur so rarely it will dissuade consumer search. It simply tells the consumer that if you find a lower priced item in the same range elsewhere Bunnings will honour its guarantee to have the lowest price for the item by matching the price and discounting it by 15%. It says no more than that.

[74] Further, the LPG militates against any absolute and literal interpretations of the advertising messaging. It tells the consumer that claims such as “lowest price guaranteed” and “lowest prices every day”, part of the advertising alleged to be liable to mislead or false or misleading, that Bunnings might not necessarily have the lowest prices but will not only match lower prices on the same stocked item found elsewhere but discount it by 15%. That is a clear signal to consumers that not every product in Bunnings stores will be the lowest price. It cannot mean, as Dr Lang asserted, that consumers would understand it to mean all products would be the lowest in the market at all times as otherwise there would be no need for the LPG. Neither can the fact that consumers did not generally avail themselves of the LPG, as Dr Gendall noted based on statements of Bunnings own employees when interviewed by the Commission, mean it is to be considered of no effect or as reinforcing the representation. It is a clear signal to consumers that not everything at Bunnings stores will be the lowest price. It provides a consumer with a remedy in the event the consumer finds that to be and wishes to take advantage of the LPG.

[75] Professor Stephen, whose evidence I accept on the effect of the LPG reinforced his expert opinion evidence with reference to academic literature, in particular an article by S Baglione, L Tucci and J Talaga “Consumer Preferences Among Low-Price Guarantee Offers”25 where it was said that LPG’s are disclaimers of lowest prices and in an article by S Dutta, A Biswas and D Grewal “Low price signal default: an empirical investigation of its consequences”,26 where it was said that LPG’s perform a remedial function. Secondly consumers can be trusted to use their common sense. Most of the class of consumers in the advertisements targeted by Bunnings would be well aware of the nature of the type of stores that Bunnings and Mitre 10 operates. They carry huge numbers of SKUs. They would, as all the experts agreed, consider it was not possible for Bunnings to know on a daily basis exactly where its competitors pricing was pitched in relation to individual items. They would expect Bunnings to have systems to identify the opposition’s products and where identical products were sold by opposition retailers at prices lower than Bunnings they would expect Bunnings to adjust its prices but they would also accept this would not be an overnight transformation. Time would be needed for Bunnings to be able to complete surveys of competitors pricing and react.

[76] Consequently a purely literal reading of the representations would seem unlikely looking at the matter objectively from the position of the consumer in the class affected by the advertising. As was said by Lord Reid in Lewis v Daily Telegraph Limted 27 the ordinary man can and does read between the lines in the light of his general knowledge and experience. Similarly a degree of “robust realism” is necessary, as the Court of Appeal said in Geddes v New Zealand Dairy Board28 when determining whether advertisable promotional material is false or misleading, or in the context of this case, involving as it does allegations both under s 13 of the Act as well as s 10 of conduct liable to mislead, the advertising would be unlikely to mislead a consumer unless the Commission is able to establish a pattern of prices higher than Bunnings competitors across the range of its goods that exceed what would be reasonable in all the circumstances.

25 (2016) Atlantic Marketing Journal, Vol. 5, No. 1 at p 21.

26 (2007) J of the ACAD.mar.sci.35:76-88 at 86.

27 supra.

28 CA180/03, 20 June 2005 at 79.

The price comparisons

[77] While there was unanimity among the experts as to what Bunnings intended as representations but disagreement as to whether it amounted to conduct which objectively meant it was liable to mislead or amounted to false or misleading representations, the Commission adduced evidence of a number of price comparisons conducted during the charge period. None of the price comparisons were statistical surveys but were attempts to compare the same stocked items offered for sale by Bunnings and its competitors at various times covered by the charges. Mr Goss, Bunnings Merchandise Manager for New Zealand who attended the interview with the Commission on 19 February 2016 estimated that approximately 19,000-odd SKUs of the 62,000 Bunnings stocked were exclusive to Bunnings but that 69.5% of Bunnings SKUs were widely available in New Zealand.

[78] As for the s 13 charges, being those relating to false or misleading representations, the representations alleged being a combination of tagline and headline claims as set out in a schedule produced by the Commission, no evidence was led that on any single one of the items identified for sale in the Bunnings advertisements or published on its website, in newspaper advertisements, television advertisements or sales catalogues that Bunnings did not offer the lowest price for any of the products so featured. As Professor Stephen said the “lowest prices everyday” tagline was an abstract statement, and not about a particularised product or item for sale. Secondly Mr Ross QC for the defendant submitted that there was no reliable evidence about Bunnings prices across its whole range, which would include items exclusive to Bunnings, or on those products on which it competes in the proportion of products across its range that the Commission says Bunnings was not the lowest price during the charge period.

[79] Mr Ross QC, relying on the Court of Appeal decision in Commerce Commission v Bunnings Ltd29 submitted that none of the price comparisons were, in fact, auditable, as the necessary records to enable that to occur no longer existed. Consequently the summaries of the outputs could only be conjecture as no foundation

29 [2020] NZCA 310.

was available. He submitted that the Whitford guidelines as set out by Whitford J in Imperial Group PLC v Philip Morris Ltd30 were applicable as a guide to ensuring the underlying data for the price audits had been accurately collected.

[80] Dr J D Henstridge, Managing Director and Chief Statistician at Data Analysis Australia, an experienced statistician used to preparing and conducting surveys opined that none of the price comparisons surveyed in evidence could be considered to provide reliable statistical evidence of Bunnings prices across its whole range of products as none were reliable in a statistical sense and had the potential to be misleading. He considered there was a “significant mismatch” between the sampling and the actual Bunnings population of products saying that a particular sample might miss or under-represent certain types of products or certain categories, or alternatively over-represent them. His view was particular care needed to be taken in extrapolating from the price comparisons inferences that could be drawn as against the whole product population of Bunnings.

[81] The logic of Dr Henstridge’s expert opinion that reliable inferences or conclusions could not be extrapolated from the price surveys if the surveyed items were not representative of Bunnings whole range or of its entire range for non- exclusory products is illustrated, to use the HOED survey as an example, and which produced the most favourable percentages for Bunnings but gave weightings in terms of the items surveyed to products disproportionate to their percentage of Bunnings products as a whole. For instance, 14.9% of the survey concerned garden lifestyles products whereas the percentage of this category’s products sold by Bunnings as non- exclusive products was 28.6%.

[82] Consequently the weighting of items surveyed, for all of the surveys were defective in similar respects, by not concentrating on the items surveyed as a percentage equating to Bunnings non-exclusory products in the particular range meant that the percentages deriving as showing Bunnings as not being the lowest on an overall basis are unreliable. If Bunnings prices were lowest for a category of products that were not surveyed or if the percentage of products surveyed was proportionately

30 [1984] RPC 293 at 302 – 303.

less than those represented in the same categories of product sold by Bunnings then the overall percentage by which Bunnings was said not to be the lowest would be artificially distorted against Bunnings. Conversely if Bunnings prices were higher in categories of products not surveyed or surveyed disproportionately to the actual percentage of the range of Bunnings products then potentially the outcome would be more adverse to Bunnings.

Comparisons of the same stocked items

[83] The comparison, Mr Dixon QC submitted, which ought to carry most weight was a comparison conducted by Mitre 10, specifically conducted by Mr M Snowden. He was employed by Mitre 10 NZ Ltd as its Group Manager for Business Efficiency from February 2014 to August 2020. Prior to 2015 he said Mitre 10 would nominate an individual within the company who would be responsible for checking approximately 300 to 400 products per month and checking the same products with comparable retailers by utilising the competitors’ publicly available websites. Bunnings was one of the retailers. Websites were used as a way of obtaining up to date pricing for those products and if the product was not available online then a personal visit to competing retailers would be made on one or two occasions each month to complete the data.

[84] Mr Snowden said he developed, with the assistance of PriceTech, a software company, a system which enabled, what he described as ‘web scraping’ which involves collecting data by way of price comparisons from competitors’ websites and which enabled a large amount of data to be collected within a relatively short period of time. By comparison with the former practice of having an individual collect 400 competitive prices on a monthly basis, web scraping enabled approximately [suppressed from publication pursuant to s 205(2)(e) of the Criminal Procedure Act 2011] prices to be collected and analysed each day and could be repeated across multiple retailers. There were reasons, which concerned the amount of time taken up on a competitor’s website, to limit the amount of data collected but the data he was responsible for collecting over the charge period showed that for the same items sold by both Bunnings and Mitre 10, Bunnings was not the lowest price on an average of

over 1,200 products each month from June 2015 to February 2016, amounting to approximately 27% of the products the subject of the survey.

[85] The other advantage with web scraping was, Mr Dixon QC submitted, its reliability as the 2020 comparison data sheet contained 2,137 products that Mr Snowden identified as matching with 2,135 (99.9%) subsequently found to be matching. Other systems such as price matching, which relied on humans to physically check prices at a competitor’s store had the disadvantage of being subject to human error.

[86] The second set of price comparisons were those undertaken by Bunnings itself using the traditional approach of price matching. Visits were made to competitors stores to check prices for a random array of products between 3 June 2015 and 29 February 2016. Both Mitre 10 and Bunnings interest in competitors’ prices illustrate how aware operators in this type of industry selling large numbers of SKUs in the same market need to be of opposition prices. The Bunnings comparisons, which were obtained and analysed by the Commission indicate that, in comparing 3,679 individual products for 1,196 of them, or 32.51%, Bunnings did not have the lowest prices.

[87] The data collected by Bunnings and provided to the Commission was analysed by Mr N Prasad, a senior economist at the Commerce Commission. His analysis was disputed by Bunnings and Mr Samuel, currently the Merchandise Planning Manager at Bunnings, reassessed the results and found that a number of items had been incorrectly matched by Bunnings staff who had undertaken the competitor visits which provided the underlying data analysed by Mr Prasad. There were, he said, a number of incorrect matches or items without barcodes or wrong barcodes on the data shown in the underlying Bunnings spreadsheet sent to the Commission. His view was that Bunnings were lowest in these comparisons 84.5% of the time or, alternatively, were not the lowest in 15.5% of the comparisons rather than 32.51% as assessed by Mr Prasad.

[88] The next set of comparisons were those undertaken by HOED, a market research company instructed by Bunnings to undertake price comparisons of its and competitors’ products between June 2014 and February 2016. HOED used field

workers who visited nine other large retailers, Mitre 10 Mega, Hammer Hardware, Placemakers, ITM, The Warehouse, Super Cheap Autos, Carters, Countdown and Briscoes, initially in Auckland but then from July 2015 in addition Wellington, Christchurch and Dunedin, surveying each of the comparison retailers in the four centres monthly. They conducted a total of 6,193 checks in relation to a total of 692 different products. The products were supplied by Bunnings to check. A master spreadsheet was forwarded to HOED and in the spreadsheet Bunnings would select the product item it wanted checked. Checks were undertaken using the SKUs or barcodes. Exact products were sought as the comparison but each SKU for each different chain was different. Barcodes were more accurate. Clearly there were potentials for inaccuracies using this method but of the total of 6,193 checks in relation to a number of 692 different products, Bunnings were not the lowest on 807 occasions, approximately 13% of that time. Mr Samuel also re-examined the HOED results and assessed the difference as closer to 10%.

[89] The last set of price comparisons were those produced by Perceptive, an independent market research company which was instructed by the Commission to undertake price comparisons of Bunnings products against its competitors in November and December 2015. One hundred products were to be compared in Auckland, Wellington and Christchurch, a total of 300 collectively although in fact only 276 products were located and compared. The results indicated Bunnings were not the lowest price about 25% of the time. Again Mr Samuel reworked the comparisons and concluded that the results provided by Perceptive were inaccurate and that Bunnings was not the cheapest only for about 14% of the time.

[90] Mr D P Shaw, the Managing Director of Perceptive, which is now part of the Clemenger group of marketing companies, gave evidence about Perceptive’s engagement by the Commerce Commission and the surveys undertaken. The actual survey work was undertaken by a group of field workers used from time to time by Perceptive for these type of services and who seemed to have been engaged on an hourly rate. The Commerce Commission limited the survey to 100 products surveyed. Mr Shaw said that experience had taught him that a survey with such a limited number of products would result in an error rate of approximately 9.8%.

[91] It became apparent from Mr Shaw’s evidence that there were a number of problems with the survey. Firstly, Perceptive never had access to the full range of Bunnings products or to their SKUs. Field workers were required to record the barcode for the items being compared at competitors’ stores but a number of them failed to do so or recorded the store’s SKUs instead. The manufacturer’s barcode is its record for the product, described by Mr Snowden as the manufacturer’s SKU number, and so are often common to all stores selling its merchandise whereas a SKU number is one placed on it by the particular store or retail chain. They vary between retailers. The most accurate confirmation of an identical product is the barcode which is unique and provided by the manufacturer. If found at two different retailers, Mr Snowden, in evidence that was unchallenged, said the product would be identical in pack size and description.

[92] There were also clear variations in the way the field workers approached their task in the Perceptive survey and so it could not be certain that the same products were being truly compared. Secondly, human error was clearly present in recording a number of entries. As an example an entry in relation to certain products was ascribed to a visit to a Guthrie Bowron store which did not stock those products but instead ought to have been ascribed to The Warehouse.

[93] The primary guidance given to the Perceptive field workers was the product name and description. Mr Shaw accepted that the 100 items, chosen by Perceptive as a result of a previous survey it had done for Bunnings in 2008, did not reflect the weighting of Bunnings product offerings. The Excel sheets given to the field workers did not provide for barcodes or SKUs for the competitor products to be entered so it was not possible for anyone subsequently viewing the data to find out whether the product surveyed was identical to that sold by Bunnings. There were difficulties retrieving the information used in the survey when it was sought by the Commission as a result of disclosure requests in the proceeding and, as Mr Samuels was able to demonstrate in his evidence, there were several incorrect comparisons, for instance, a Stanley screwdriver set sold at Mitre 10 was shown on the Perceptive data to be cheaper than that sold by Bunnings but Mr Samuels’ re-analysis of the product as it appeared on Bunnings’ website showed it was of a different type, the one sold by Bunnings being sold with the name “Jeweller” and the one at Mitre 10 sold under the

name “Precision” and so, as Mr Shaw accepted, there was no match by description. Other examples included CRC Liquid Armour said to have been cheaper at a competitor’s store but sold there in a 250 millilitre container whereas Bunnings sold the product in a 500 millilitre container which would clearly be more expensive. The difference in container sizes had not been noted or more likely, noticed by the field worker.

[94] Without labouring the point there were several other errors in the comparison data. Consequently I agree with both Dr Gendall and Dr Henstridge who correctly traduced the evidential value of the Perceptive survey saying the small sample size and methodologies used meant not a great deal of weight could be placed on it. Dr Henstridge made the point, which I accept, that the Perceptive survey was not representative of the various percentages of Bunnings products in categories of non- exclusive products sold by it. He said 65% of Bunnings products were sold for less than $50 but for the survey, in the very limited number of products it did compare, 90% of those products were in the under $50 category and were hardly a representative range of Bunnings non-exclusive products which showed a clear bias towards lower cost products which necessarily distorted the conclusory percentage as to Bunnings prices not being the lowest.

The HOED survey

[95] The HOED survey was conducted at Bunnings’ request. HOED is a market research company that has been operating in New Zealand for at least 15 years and continues to provide a price checking service for Bunnings. In 2013 it was asked to undertake a price comparison of items sold by Bunnings, with the items selected by Bunnings, by comparing them with seven or eight other major competitors including Mitre 10 Mega, Hammer Hardware, ITM, Carters, and other similar stores. Initially the price comparisons were only done in Auckland but subsequently undertaken in Wellington, Christchurch and Dunedin from July 2015 onwards with monthly checks in all four centres being carried out.

[96] Mr R J Potton, the Managing Director of HOED, described the process as Bunnings forwarding a master spreadsheet to HOED setting out the items it wanted

price comparisons for. Mr Samuels confirmed the list was compiled after he, who was responsible for liaising with HOED, consulted with the managers of the various product lines sold by Bunnings. Eight hundred products were checked in each centre. However the same product would not be checked at different centres so that, if for instance, the price comparison for a particular hammer had been undertaken in Auckland from a Mitre 10 Mega store the same check would not be carried out in a Mitre 10 store in Dunedin. Bunnings would provide HOED with its SKU numbers, the product barcode and website pictures for the products it wanted price comparisons for.

[97] The HOED price checkers compared the shelf prices at the various competitors with the Bunnings prices. Analysis of the HOED results by the Commerce Commission indicated that Bunnings was the lowest price on 83.32% of the price checks and equal lowest on 3.65%, meaning that Bunnings, in these results, could claim to have the lowest prices 87% of the time for the products compared.

[98] As with the Perceptive survey and the other surveys, the HOED survey was not a statistical survey. It was one designed or intended to be used by Bunnings as part of its day to day business. Detailed underlying primary records of the competitor products and prices compared were not kept and so verification of the competitor product and price details was not possible.

[99] Both Dr Gendall and Dr Henstridge thought the sample range was wide enough and the survey reasonably rigorous. Over the charge period an average of well over 2000 products a month were compared, however Dr Gendall had reservations as the Bunnings prices were not checked by HOED and he was suspicious of the lowest price comparison as it seemed out of kilter with the other surveys, meaning presumably the Mitre 10 comparison and Bunnings internal price checks.

[100] One of the problems with the survey, as identified by Dr Henstridge is the survey was not designed to give a qualitative measure of the proportion of products where Bunnings was not the lowest. It was, primarily, a survey undertaken to assist Bunnings with its pricing and so the items identified by Bunnings for survey would presumably be those items where Bunnings considered there was a risk they were not

the lowest. That seems a reasonable inference to draw. It also suffered, as did all of the surveys, that the items actually surveyed, as I have already noted, did not correlate with the overall percentage of products in particular categories of goods sold by Bunnings in its non-exclusory range. It therefore, as with the other surveys, cannot be seen as a representative survey across the whole range of non-exclusory products.

[101] Dr Gendall also offered an opinion as to the quality of the other surveys adduced in the evidence and was clearly impressed by the Mitre 10 web scraping exercise not only because of its width, but because it did not involve the “human element in terms of collecting and comparing the prices”, although there was some human analysis for the purpose of checking the accuracy of the identification of products for sale as being the same as both Bunnings and Mitre 10. His opinion was that the price disparity evidence the survey revealed was compelling evidence that on a significant proportion of products Bunnings was not the cheapest compared to their major competitor. As for the Bunnings internal price checks he also thought they were compelling evidence as Bunnings had undertaken the checks themselves, had selected the product range to be analysed and saw it as reinforcing, in effect, the Mitre 10 evidence.

Mitre 10 survey – Mr Snowden’s evidence

[102] The Mitre 10 survey was a direct comparison of items offered for sale on the websites of Mitre 10 and Bunnings and involved Mitre 10 using a price optimisation software company, PriceTech, collecting prices from Bunnings’ website, a process known as ‘web scraping’ or ‘web crawling’, thereby enabling Mitre 10 to compare them with their own. Mitre 10 would advise PriceTech of the parameters of the exercise. PriceTech maintained the process by which they obtained the information was confidential to it and Mitre 10 however Mr A L Pascoe, the CEO of PriceTech said web data collection or web scraping was now a common practice for pricing comparison.

[103] As with the other surveys Mitre 10’s ‘web scraping’ survey was not a statistical survey but began to be used by it in 2015 to set the optimised retail price for its

products, particularly those in the home improvement market where there were often high weekly sales volumes.

[104] Mr Snowden said the survey was developed for business comparison reasons to enable Mitre 10 to consider its prices against its competitors and is still used today. He was solely in charge of the process when it started in 2015 but later in 2015 a data analyst was also added to the project.

[105] He described the process as checking the data received from PriceTech against the data taken from the competitor’s website, checking it against the competitor’s SKU number and then checking that the price received from PriceTech matched the price shown on the website, which invariably it did. Generally the information received from PriceTech through web scraping would include the SKU number of Bunnings’ website for the product sold by Bunnings, a product description and the price. Less frequently, additional information was sought such as [suppressed from publication pursuant to s 205(2)(e) of the Criminal Procedure Act 2011] manufacturing codes as well as other data. The primary limitation on collecting the greater range of data was the need to avoid taking up too much server time on competitors’ websites.

[106] Armed with the data received from PriceTech in the form of an Excel spreadsheet, Mr Snowden would then conduct what he described as ‘spot checks’ on the data received by matching it against the Bunnings website. Large numbers of Bunnings data sets were received, Mr Snowden saying that in July 2015, approximately 35,000, with similar large numbers collected each month. These would be compared with a table of Mitre 10’s data with its SKU, product description and price shown. The large numbers of monthly data sets did not represent distinct products as many occurred across many months and so the actual number of products at issue would be substantially lower than suggested in the total monthly data sets collected by PriceTech’s web scraping. Mr Snowden said at least 50% were physically checked by him or by those working under his supervision.

[107] For the comparison, physical examination had its limitations, with Mr Snowden saying he could not match more than 200 – 300 products on any single day. The most accurate way of establishing a comparison was to find the manufacturer’s

code on the Bunnings items, which usually is the same for all retailers selling the product. Once the limitation of physical matching, including the huge volume of data and inaccuracies caused through human error became evident, Mr Snowden said he [suppressed from publication pursuant to s 205(2)(e) of the Criminal Procedure Act 2011]. A physical check by either going to a Bunnings store or looking online on Bunnings’ website would then be carried out by Mr Snowden and later by a data analyst working under his supervision.

[108] While the manufacturer’s code was the best basis for comparison, where the code was simple with a small number of identifiers it might not be unique. Mr Snowden would use other functions in Excel to try and determine whether there was a match such as searching under product category and sub-category, description and price, especially if the price was similar between the two products.

[109] Between June 2015 and May 2016 several thousand products were found by Mr Snowden to be matching products. In August 2015 the highest total was 4,974. The lowest total was in June 2015 where 3,095 were matching products. Overall the Mitre 10 data as analysed by Mr Snowden between 1 June 2015 and 1 May 2016 showed the percentage of time where Bunnings was not the lowest or lowest equal on matching products to be 27.2%, and the percentage of time where Bunnings was not the absolute lowest to be 32.74%. Mr Dixon QC submitted that the table setting out this data showed there was a variation of just 2% above or below the 27% average when Bunnings was beaten on price during that period.

[110] The limitations of that evidence was addressed in the recent Court of Appeal decision Commerce Commission v Bunnings Ltd31 where the product/price database comparisons for products offered for sale on each companies websites was seen as fundamentally an automated process but where the actual comparative database for the results from June 2015 to May 2016 were lost. Other comparisons were undertaken by Mitre 10, specifically in 2017 when Mr Snowden created a sample to enable Bunnings to check that products he matched for the earlier survey did in fact match. Overall in the 2017 comparison 472 unique products were produced and were

31 supra at para 74.

checked by Mr Lourie of the Commerce Commission. He was able to find 404 products from the websites of various retailers and check whether the products were still listed on their websites and found all of them.

[111] Further in 2020 Mr Snowden recreated the process used by checking the products in Mitre 10’s current database with the prices he had for those products from 2015-2016. There were approximately 4,500 matches which were then checked against the data set for the products on both companies’ websites, with 2,300 matches available. As with the 2017 data set Mr Lourie checked the 2020 figures to see if the product matches corresponded to the product descriptions. His conclusion, which was not challenged, was that 2,137 products were available on both websites and 2,135 matched. For those products that were checked by Mr Lourie the percentage of times where Bunnings was not the lowest or lowest equal was 25.73% of the time and where it was not the absolute lowest, 31.08%. The flaw, however, was that although Mr Snowden organised the new Dataset in 2020 it was different from the database used in 2015/2016 and has not, as Bunnings submits, been produced in evidence and so is not testable.

[112] The Mitre 10 surveys were not, however, statistical surveys. Neither Mr Snowden nor the Commerce Commission contended they were. Evidence of web scraping from Bunnings website came from Mr Snowden, then working for Bunnings main competitor, Mitre 10. It is not known whether all of Bunnings products in its non-exclusory range, which were also sold by Mitre 10, were listed on its website. Consequently if there were products for which the two companies competed on price, and those products had not been shown by Bunnings on its website the overall percentages at which Bunnings was said to not be lower than Mitre 10 may have been distorted. Secondly the data scraped from Bunnings website by Mr Snowden in 2015/2016 is not available for analysis. It cannot be determined whether the web scraping process extracted everything correctly from the website. As for the later data sets there were also similar issues as to whether the range analysed was completely representative of Bunnings overall range of non-exclusory products for which it competed with Mitre 10 for sales.

[113] Bunnings may have had particular reasons for wanting to advertise some non- exclusory products on its website and not others. They may not have been representative, in percentage terms, of its overall product range. For instance Dr Henstridge’s evidence was that 31% of the 2018 data sets analysed by Mr Snowden concerned tools whereas tools only form 8.8% of Bunnings national product range. Overall Dr Henstridge concluded, and I accept, that the samples scraped from Bunnings website were not representative of Bunnings full range of products or of Bunnings non-exclusory full range of products. While I have no doubt that at times Mitre 10 were cheaper on a number of individual products the figures supplied cannot be used to extrapolate a conclusion that Bunnings overall were beaten by a factor of 27% on price during that period. They may have been beaten on price on some occasions but taken as an overall percentage of Bunnings products there was no evidence that the different percentages asserted by the Commission’s witnesses could be relied on as a statistic.

[114] To quantify an acceptable percentage for the number of same products not the lowest price in the market having regard to the huge number of SKUs, the volatility of pricing in the industry and the potential for manipulation of prices by a competitor deliberately undercutting Bunnings to grab market share, a proposition considered by Professor Stephen and which I accept would be a factor taken into account by consumers in not expecting every item offered for sale by Bunnings would be the lowest price. Having regard to those factors I think an acceptable range where Bunnings was not the lowest price or the same priced item would be approximately 15% provided the LPG in the form it was advertised was available.

[115] The other criticism I accept is that there was an over-representation of products sold by Bunnings for between $20 and $50 and a strong under-representation as to higher price items. That, of course, arises as a result of what it was Bunnings chose to advertise on its website which enabled Mitre 10 to make comparisons but it does not assist in being able to draw any satisfactory conclusion as to the representative nature of the overall percentages Bunnings are alleged not to be the lowest on price. For the reconstructed comparisons Mr Snowden agreed they were not scientifically calculated nor intended to prove the correctness of the larger data set which could not be reconstructed because the information underlying it was no longer available.

[116] Mr Snowden accepted that Bunnings had regional pricing that might differ from its website and also, as Mr Samuel in his evidence confirmed, had store-level pricing that could differ. Mr Samuel said there was some regional variation for items advertised on its website so that it was possible a person checking a price in Gisborne for an item might find the price different to that advertised for the same item for a Bunnings store in Auckland. Price would often depend on the costs to Bunnings incurred as a result of delivering the products to its stores through its various supply chains.

[117] Whether those variations would make a difference in terms of Mr Snowden’s analysis is unknown but the point is that, together with the selection of cheaper priced products published on Bunnings website, it is difficult to conclude Bunnings were overall significantly beaten on price by Mitre 10 to the degree claimed. Mitre 10’s analysis was intended for use as a commercial tool, not as a statistical survey. It did not purport to be a study of Bunnings prices across its range or even across the whole of identical products sold by both organisations, only those that Bunnings chose to display on its website. Reliable conclusions are therefore difficult to draw insofar as they provide an overall assessment of the extent to which Bunnings were not the lowest price on the same stocked goods and in any event their position is effectively saved by the LPG which when activated would mean they always would be.

[118] The Commerce Commissions submission was that the price comparison evidence, in particular the Mitre 10 evidence, established that Bunnings advertising was false and misleading and also liable to mislead. Mr Dixon QC submitted that although each of the price comparisons had different strengths and weaknesses there was consistency between the percentages of time where Bunnings was not lowest with rare exceptions as shown by the various analyses undertaken by Mr Snowden and also the checks conducted by Mr Laurie which showed Bunnings was being frequently beaten on prices. Many of the items consisted of only a difference of a few cents on products offered for sale under $50 and what should not be lost sight of is that the charges are framed on the basis the conduct complained of was that Bunnings various advertising in the forms it used was conduct liable to lead the public into thinking its prices were the lowest prices on offer in the market when they were not. No specific segment of Bunnings products was targeted by the charges with Bunnings having in

stock some 62,000 items or SKUs. Consequently tables showing that on a monthly survey, for instance in April 2016, 4,906 products matched as being sold by both Bunnings and Mitre 10, a figure of 26.01% shown in the table referred to in Mr Dixon QC’s submissions as representing a percentage where Bunnings was not the lowest or lowest equal cannot necessarily be extrapolated across Bunnings entire range of products.

Bunnings internal price checks

[119] Mr Prasad, a Senior Economist at the Commerce Commission analysed Bunnings internal price checks data provided in the form of 1,302 Excel spreadsheets which were received from Bunnings in the course of disclosure and recorded internal price checks conducted by Bunnings staff, mostly by competitor visits. He used a Stata programme to compile and analyse the Excel spreadsheets and using those spreadsheets reached the conclusion that between 3 June 2015 and 29 February 2016 of a total of 3,679 checks by Bunnings, competitors were cheaper on 1,196 occasions, equating to a percentage of 32.51%. These included checks of Bunnings’ main competitors but as against Mitre 10 the analysis showed that there were 1,850 checks with Mitre 10 Mega cheaper on 586 occasions, equating to 31.68%. A number of products were price matched more than once. When the period was extended to 5 April 2016 the analysis showed the competitor was cheaper on 30.93% of occasions. He did not, however, undertake any observation or checking of prices himself and was unaware of which products had been checked at competitors’ stores or when that occurred.

[120] For recurring products which were analysed, of the 2,117 products that fell within that category between June 2015 and May 2016, the percentage where Bunnings was always more expensive was 16.2%, suggesting Bunnings was responding by reducing prices once it identified cheaper prices for sale at competitors’ stores.

[121] One of the difficulties with the comparison, as became evident in cross- examination of Mr Prasad by Mr Lindsay, was that as many of the products identified by Bunnings through internal price checks conducted by its employees involved actual

visits to competitors’ stores, inaccuracies crept in. Secondly the Bunnings records often did not provide the dates the Bunnings staff member visited the competitor’s store. Unlike the Mitre 10 data, the Bunnings data comparing internal competitor prices was not consecutive. Mitre 10 conducted a comparison of a large number of the same products consecutively each month and on the same day.

[122] The competitive nature of the industry is also illustrated by Bunnings’ figures. Some products, presumably those with high sales volumes, were regularly checked. For instance Selleys Sugar Soap was checked 21 times by Bunnings at its competitors’ premises, Bunnings being cheaper on 16 occasions, and the competitors’ four. CRC 556 400 mls was checked regularly at competitors. Bunnings was cheaper on 26 out of the 29 times the product was checked.

[123] Market volatility and the way competitors watch each other’s prices and reduce them as appropriate is illustrated by the Selleys Sugar Soap example. Bunnings conducted a competitor check on 29 January 2016 and found it was selling the product at the same price as its competitor, namely at $5.85. On 2 February 2016 the competitor was higher and Bunnings reduced its price on 5 February. On 7 February another price check was done and Bunnings found it was selling the product for $5.93 and was higher than the competitor. Within a few days Bunnings lowered its price to

$5.34. Consequently there were certain days when Bunnings did not have the lowest price for the product but once it established that it took steps to ensure the price was the lowest.

[124] Some of the errors made in personal checks were also detailed by Mr Samuels, a Bunnings employee who conducted a retrospective analysis of the spreadsheets which indicated that some of the products featured, and which showed Bunnings as being more expensive, were in fact Bunnings’ exclusive products so the employee who had checked at competing stores had plainly not made an identical match. A mistake had been made. Similarly SKU numbers which were recorded in the competitor price matching sheet were, Mr Samuels said, not found in the Bunnings system for the particular time period, yet another example of Bunnings products being incorrectly recorded in the underlying data. An illustration of the error made and which crept into the analysis was of a spreadsheet describing an item being matched as a ‘garden wall

keystone rock face’ but with a competitor price matching sheet showing the reference in the spreadsheet was for a sander.

[125] Mr Samuels’ evidence was that the retrospective reconstructed analysis of the Bunnings spreadsheet showed a percentage of 15.5% where Bunnings did not have the lowest prices in relation to its competitors, which Mr Dixon QC submits is “still worlds away from the sort of performance that Bunnings needed to achieve to deliver on its lowest prices promise”. Mr Samuels was able to achieve this retrospective analysis by stripping out data from the Bunnings internal spreadsheets provided to the Commerce Commission, specifically SKU numbers which he said were invalid, incorrect matches or no barcodes and exclusive products.

Summary – the individual price audits – Bunnings submissions

[126] Bunnings submitted the price audits, not being proper scientific surveys, did not have the necessary foundation by observation to enable them to be relied on in evidence. It submitted the audits did not have a tendency to prove anything, of consequence to the determination of the proceedings; s 7(3) Evidence Act 2006 because there was no reliable foundation for the asserted results.

[127] Some of the price audits had errors, a number of which have already been identified in the preceding paragraphs, but more fundamentally it submitted that none of the price audits were supported by documents which allowed the defendant or anyone else to assess whether they were correct or not.

[128] The Commission’s primary submission was that the four different price comparisons conducted during the price period both individually and collectively proved that Bunnings were not offering the lowest prices in the market with rare exceptions. Overall I am not willing to draw a conclusion that the evidence from the four surveys, in particular the Mitre 10 survey which Mr Dixon QC submits is the most important of the price comparison evidence, is sufficiently robust and statistically reliable for me to extrapolate a conclusion that Bunnings prices were not the lowest so as to conclude that the advertising was liable to mislead as to price in terms of the s 10 charges. Further the LPG guaranteed consumers that they could obtain the lowest

price on the same stocked item if they found it at lower price elsewhere. The flaws in the various surveys are such that I am not prepared to accept the Commerce Commission has met the evidential burden of proving the advertising, for the purpose of the charges, was liable to mislead the public as to a particular characteristic, namely price.

[129] Further for the reasons already discussed the percentage margin by which it is reasonable to expect that Bunnings prices are not the lowest prices is higher than 5% as Dr Lang postulated, somewhat arbitrarily, but is also higher than Professor Stephen’s formula of Bunnings being more often than not the lower price. An acceptable range would be higher namely approximately 15% of the times when Bunnings was not the lowest provided the LPG was there to cure the position by ensuring a consumer could obtain the lowest price for the product from Bunnings should they seek to do so.

[130] The price comparisons because they are not statistical comparisons, do not provide sufficient support for the prosecution, given the standard of proof required is beyond reasonable doubt, for me to be satisfied that the Commerce Commission has proved the s 10 charges to that standard. Accordingly the s 10 charges are dismissed.

Section 13 charges

[131] The s 10 charges concerned alleged infringing conduct at Bunnings stores by what, in effect, were taglines such as “lowest prices are just the beginning ...” and the other taglines specified in the particulars accompanying the charges. That claim appeared on the interior and exterior signs at Bunnings stores and also on staff uniforms. I agree with Dr Gendall and Professor Stephen’s views that the tagline claims in themselves are no more than holistic statements of Bunnings position in the market and are not, unless accompanied by what they described in the evidence as headline claims, representations. In any event the other matters already mentioned does mean those charges fail.

[132] The s 13 charges concern advertising by Bunnings in various media such as their website, print, catalogue, television and radio statements. They were laid as

representative charges, as with the s 10 charges, with each charge covering a three month period across the 21 months of the charge period. The claims that Bunnings made false or misleading representations with respect to its goods fail for the same reasons as the s 10 charges. Representations in the media advertising such as lowest price guaranteed and nobody beats our prices and the like as set out in paragraph [2] of this decision are, to the extent they constitute taglines, not representations but much of the advertising was also accompanied by displays of specific products for sale with prices specified and so could be seen as such. These are not deceptive misleading as representations as the remedial nature of the LPG means a consumer would accept there may be other retailers selling products sold by Bunnings for a lower price and the LPG provided the consumer with the opportunity of having Bunnings provide them with the lowest price.

[133] Further the amendment ... with rare exceptions also applied to those charges and the evidence was not sufficient to conclude that the representations for a number of products Bunnings had for sale at a higher price than competitors did not fall outside an acceptable percentage or range for Bunnings products that would take into account the vast number of SKUs, the competitive nature of the market and the lag for Bunnings to be able to adjust its prices once it became aware competitors were selling the same products for a lower price.

[134] As for false representations the issue as to whether products were falsely advertised as to price with rare exceptions, the position seems to me to be the same even ignoring the contradiction created by the amendment that a price can be false but with rare exceptions. Either it is false or it is not. The LPG effectively remedies the position. No evidence was adduced that any product the subject of the advertising by Bunnings in the media charges was available elsewhere at a lower price. No specific item was taken from the items advertised by Bunnings in a newspaper on a particular day with that item then shown to be available at a lower price at a competitor on the same day.

Summary

[135] The Commission asserted, through Mr Dixon QC that consumers would have understood the words used by Bunnings in its advertisements in the light of their ordinary and natural meaning across the whole range of their products by pointing to words such as “everything” and “all” appearing in individual advertisements. As for the LPG it was effectively dismissed as being of no account as, in Dr Lang’s words, it depended on how motivated a consumer would be to check prices at competitors and secondly the use of the LPG by Bunnings signalled, in his view, a message that the consumer did not need to bother checking. On the Commission’s argument the impression conveyed by the advertising was to be taken in the light of its ordinary and natural meaning but that seemingly did not apply to the LPG. I do not accept that. Secondly expert opinion on the meaning of the words used in the advertisements was divided in the way I have already outlined, even as between the Commission’s own expert witnesses.

[136] Expert evidence is not, of course, determinative as to whether a representation is misleading or deceitful or liable to mislead which must be judged objectively, although the evidence of experts can assist in that respect. The issue is, in essence, whether the consumer in the class identified would be misled or deceived by the advertisements. I agree with Dr Gendall and Professor Stephens that the tagline statements, on their own are not representations and have already determined that the advertisements would not be seen by the consumer in a literal way. Consumers would understand them in the way described by Lord Reid in Lewis v Daily Telegraph Ltd32 “... in the light of his general knowledge and experience in wordly affairs”, and would take into account the nature of the industry, the size of the stores, the number of SKU’s and the general impossibility of ensuring that on each day every SKU in Bunnings stores was the lowest price. Consumers would also consider the LPG alerted them to the possibility that not every item in Bunnings may be the lowest price but providing a remedy to achieve that.

32 supra.

[137] Much of the argument among the experts was where would the threshold be for an acceptable percentage of times when Bunnings was not the lowest price, “the rare exceptions” as postulated by the prosecution. Again there was no general agreement as to what might constitute an acceptable percentage and I have arbitrarily selected one as would appear reasonable in all the circumstances. Further in establishing beyond reasonable doubt that the mind of a consumer would be misled in a sense that the representations were apt to mislead or that the consumer was deceived or actually misled, the prosecution must establish beyond reasonable doubt, having regard to what I have objectively determined as the meaning of the representations, that there was offending conduct. As Dr Lang correctly said there was no ‘hard proof’ consumers would believe Bunnings had the lowest prices across the range of its products. One way which might have been helpful would be to lead evidence of a series of consumer complaints about Bunnings prices not being the lowest as claimed for the same products. There was no such evidence and Mr Lourie, the investigator for the Commission, was unaware whether any complaints had been received. Such evidence would not in itself establish the conduct as misleading or deceptive or likely to mislead or deceive but it would be of assistance as noted in the judgment of Chambers J for the Court in Geddes v New Zealand Dairy Board.33

[138] There can be cases where the misrepresentation is made to the public at large where the misleading nature of the conduct is sufficiently obvious so that the Court can make findings without actual evidence of confusion. However evidence of complaint of deception or being misled would have assisted the Commerce Commission had it been available. The trial Judge in Geddes drew inferences that the absence of evidence from a relevant class of consumers as to their misunderstanding of the core representations amounted to an inference adverse to the appellant. Mr Ross QC submits the Court ought to draw inferences from the absence of consumer complaint evidence but it is not necessary to do so although I agree having regard to the large amount of advertising over a long period of time inferences adverse to the Commissions case are available with the absence of any evidence of consumer complaint or confusion. I am satisfied in any event that the evidence has not reached

33 supra at para 94.

the point where I could be satisfied beyond reasonable doubt that the advertising was false or misleading or deceptive or liable to mislead or deceive.

[139] Further I agree with Dr Gendall that a consumer survey about their view of the meaning of the advertising would have been of assistance to the Court. He rightly said it would not be the defining factor in the case but a well designed survey as to what consumers thought of the advertising would have assisted. Unfortunately such a survey was not commissioned.

[140] The survey evidence that was led was evidence that had considerable shortcomings as the underlying data was not available and could not be tested, except only in a limited extent in the latter Mitre 10 analysis and the evidence was not across the broad range of Bunnings product when the charges emphasised the offending conduct concerned the whole of Bunnings range. There were limitations as to the products actually selected and what percentage of Bunnings overall product range they represented both as to price and content, there were clear human errors in some of the surveys in collating the material and it was done in a non-statistical way when clearly a proper statistical survey could have been commissioned. This necessarily affects the weight to be given to the evidence and means that conclusions, having regard to the standard of proof, as to the falsity of the representations and the extent which a consumer would consider it was reasonable for Bunnings not to be the lowest price on the same items sold also by its competitors across the range of its products or that consumers were liable to be misled or had been misled could not be drawn to support the criminal standard of proof for the charges.

[141] As for triangulation theory, namely that inferences should be drawn from a range of statistically unsatisfactory surveys each with errors and bias, to justify a conclusion those inferences are not available on the evidence Dr Gendall said that although non-statistical, the product surveys of Bunnings prices compared to its competitors may not allow statistical inferences to be drawn in the way the Commission points to the Mitre 10 survey as showing Bunnings prices as being 27% higher than its competitor, Mitre 10, between June 2015 and May 2016 but becomes a matter of the weight to be given to the evidence. In my view the weight to be given

to it is such that it is not sufficient for the Commission to have established beyond reasonable doubt that the charges had been proved against Bunnings.

[142] Accordingly all charges, both under ss 10 and 13 of the Fair Trading Act 1986 are dismissed because on the evidence it cannot be said the representations were false or misleading or a reasonable person in the consumer class at issue would have or was misled or deceived or liable to be misled. In other words the Commerce Commission has failed to prove the charges beyond reasonable doubt.

....................................

Gibson DCJ