 OFFICE OF THE INFORMATION & PRIVACY COMMISSIONER  British Columbia Canada
INQUIRY REGARDING LIQUOR DISTRIBUTION BRANCH DATA ON ANNUAL BEER SALES David Loukidelis, Information and Privacy Commissioner April 19, 2000 ISSN 1198-6182 Summary: Applicant brewer sought 1994 through 1999 aggregate sales data for each of two competitor brewers. Those sales data were generated by the public body but are deemed by the Liquor Distribution Act to be information supplied in confidence by the third party brewers. Information is financial or commercial information of those third parties. Reasonable expectation of significant harm to competitive positions was shown; public body was required by s. 21(1)(c)(i) to refuse to disclose information under s. 21(1). Disclosure also could reasonably be expected to result in undue financial gain or loss, so public body also was required by s. 21(1)(c)(iii) to withhold the information. Information was not gathered for the purpose of collecting a tax, so public body not required to withhold on that basis. Key Words: Commercial or financial information – supplied in confidence – competitive position – significant harm – negotiating position – interfere significantly with – undue financial loss or gain. Statutes Considered: Freedom of Information and Protection of Privacy Act, s. 21(1); Liquor Distribution Act, s. 36. Authorities Considered: B.C.: Order No. 57-1996; Order No. 238-1998; Order No. 323-1999; Order 00-02; Order 00-03. Ontario: Order M-920; Order 125; Order P-340; Order P-561; Order P-1105; Order P-1175. Alberta: Order 99-018 Cases Considered: Workers’ Compensation Board v. Ontario (Information and Privacy Commissioner) (1998), 164 D.L.R. (4th) 129; Howard Smith Paper Mills Ltd. v. The Queen (1957), 29 C.P.R. 6 (S.C.C.); Big Canoe v. Ontario (Minister of Labour) 181 D.L.R. (4th) 603 1.0 INTRODUCTIONThis case involves an attempt by one brewer to use the right of access to information created by the Freedom of Information and Protection of Privacy Act ("Act") to obtain, at no cost, valuable confidential information that is likely to give it a competitive advantage as against its two largest competitors. The competitors resist this, of course, and argue that the Act’s right of access is intended to further the public interest and not private interests. The battle in this case was fought in a flurry of submissions, replies and further replies. Once the dust settled, it was clear the two sides agreed on almost nothing of relevance in this inquiry. Having reviewed the material, I have concluded that the public body – assisted by the third parties, Molson and Labbatt – has established a reasonable expectation of significant harm to competitive position or negotiating position for the purposes of s. 21(1)(c)(i) of the Act. I have also decided that disclosure of the information in question could reasonably be expected to result in undue financial loss or gain within the meaning of s. 21(1)(c)(iii). The LDB must therefore withhold the information. In April of last year, the applicant, Pacific Western Brewing Company ("Pacific Western"), submitted a multi-part request to the public body – the Liquor Distribution Branch ("LDB") of the Ministry of Small Business, Tourism and Culture – for records related to sales figures for Molson Breweries ("Molson") and Labatt Breweries ("Labatt"), the third parties in this inquiry. After some discussion, Pacific Western and the LDB agreed that the request would be for records related to the annual provincial sales figures in British Columbia for Labatt and Molson domestic products for each of the previous five fiscal years (April 1994 to March 1999). The LDB created a record, under s. 6(2)(a) of the Act, to respond to the request. The LDB later denied the request, under s. 21 of the Act. Under s. 52 of the Act, Pacific Western requested a review of that decision. There was no change in the issues during mediation and the matter came to me in this inquiry. In addition to argument and affidavit evidence submitted by the LDB and Pacific Western, both Molson and Labatt filed argument and affidavits. Labatt’s submissions supplemented Molson’s, since Labatt expressly adopted Molson’s argument and evidence. The LDB relied on, and adopted as its own, the evidence and argument produced by Labatt and Molson. Since s. 21(1) creates a harms-based test, a public body’s reliance on third party evidence on the harm issue is both sensible and acceptable. Pacific Western submitted written argument, but not any evidence. The Craft Brewers Association of BC was given intervenor status, but made no submissions. 2.0 ISSUESIn this inquiry, I have the task of determining whether or not the LDB was correct in deciding that it was required by s. 21 of the Act to refuse to disclose the requested records. Under s. 57(1) of the Act, the LDB must establish that the applicant has no right of access to the records in dispute. 3.0 DISCUSSION3.1 Description of the Information in Dispute – The first thing I should address is the nature of the information in dispute. As was noted above, two records are in issue, each of which was created by the LDB in response to Pacific Western’s request. One record shows aggregate dollar amounts for Labatt’s annual domestic product sales in British Columbia for each of the five fiscal years of 1994-1995 through 1998-1999, while the other shows the equivalent information for Molson’s. The amounts are shown in what are called "display dollars". According to the affidavit of Theo Kellner, filed by the LDB, display dollar amounts are the total shelf prices for a given product, including container deposits, provincial sales tax and federal goods and services tax. Theo Kellner’s affidavit makes it clear that the information in issue here is generated from data compiled by the LDB from its operations. It is also clear that this information is not publicly available. The LDB will sell information through its Supplier Data Sales System ("SDSS"). Subscribers to this service pay an annual fee for access. The data sold through the SDSS includes domestic beer product sales data only for the subscriber. A subscriber cannot get that information about another producer. The LDB discloses some data under the Financial Information Act ("FIA"), which requires the public agencies it covers to disclose certain financial information. It appears the LDB has, since the 1970s, complied with the FIA. In this inquiry, however, the LDB indicated that its legal counsel had advised it that the FIA in fact does not cover the LDB. At all events, the LDB said that it has to date published, in accordance with the FIA’s requirements, "total annual payments, which may include payments for products and services and container deposit charges" made to "suppliers to whom it has paid $10,000 or more in the relevant year". The data at present disclosed by the LDB under the FIA, it might be thought, resemble the information in dispute here. If the provincial and federal taxes included in the display dollar figures in the records here were subtracted, leaving total shelf prices and container deposits, would this not be the same as the FIA data? This supposition may explain why Pacific Western repeatedly asserted that the sales figures for brewers other than Labatt and Molson are published by the LDB. According to the LDB’s evidence, however, the FIA numbers "may include payments for … services". This means, the LDB argued, that the two data sets are "quite different". There is no evidence as to what "services" a brewer might provide to the LDB. There is no evidence before me as to whether the sales figures in dispute here reflect payments to Molson or Labatt for "services" of some kind. Nor is there any evidence that the data disclosed under the FIA include payments for services. Inclusion of "services" payments of some kind may well distinguish the two sets of data. Even if the FIA information about Pacific Western and other brewers were the same as the information considered in this order, however, it would not follow that the playing field could, or should, be levelled by disclosure of the information to Pacific Western. The question whether disclosure could reasonably be expected to harm significantly the competitive position of either brewer cannot be avoided. This issue is dealt with below. 3.2 Scope of This Inquiry – The notice of written inquiry, and the fact report, issued by this office make it clear this inquiry is limited to disclosure of the record created by the LDB, as noted above. Pacific Western’s initial submission also said that Pacific Western seeks "the annual provincial sales figures of domestic products" for Labatt and Molson. That submission also noted that Pacific Western had narrowed its request from its initially broader scope. As Pacific Western’s request for review stated: … we settled upon a request for the disclosure of the total provincial sales for Labatt and Molson domestic products for the fiscal years commencing April 1994 to March 1999 (see our letter dated May 18th, 1999). Yet in Pacific Western’s reply submission, counsel for Pacific Western said Pacific Western’s request was not limited to the two records created by the LDB in response to the request: The Applicant does not limits it’s [sic] request to these two Documents, and maintains that it’s [sic] application for disclosure is broader and encompasses all documents or records kept by the Public Body which show the annual volume of sales in British Columbia of the Third Parties’ products. In its reply submission, Pacific Western also contended, in the alternative, that if its argument for disclosure did not succeed, 5. ... the Applicant submits that portions of the Documents that are brand-specific should be severed from the records, and the remainder of the records including the total overall sales figures of the Third Parties in British Columbia should be ordered disclosed to the Applicant. 6. In the further alternative, if the Commissioner declines to order production of the Documents on the grounds that they contain confidential information which originated with the Third Parties and could reasonably cause harm upon disclosure, the Applicant reiterates that its request for information was not limited to the two Documents created by the Public Body in answer to this request. In this alternative, the Applicant submits that an Order should be made that the Public Body disclose all of it’s [sic] self-compiled ‘product movement reports’ for the last five years as they relate to Molson and Labatt sales through Public Body stores and distributors. These are documents that meet the description of the kind of information that the Applicant requested. The first request – that brand-specific information should be severed – does not avoid the s. 21(1) issue in this case. The question would remain whether s. 21 prohibits disclosure of "overall sales figures" for Labatt and Molson. In any case, the records in issue contain only the aggregate domestic product provincial sales for Molson and for Labatt, in "display dollars". Brand-specific information is not found in the records. The second argument, in the "further alternative", amounts to a new access to information request. Up to the point of its reply submission, Pacific Western was content to proceed on the basis that the two records created by the LDB are the records in issue. It cannot, at this late stage, purport to abandon that position and say that a broader request is in issue or demand access to the LDB’s "self-compiled ‘product movement reports’ for the last five years". The material before me establishes that the LDB made its decision in relation to the two created records before me. This inquiry is limited to those two records alone. 3.3 Does Order No. 238-1998 Govern the Outcome Here? – A preliminary point arises. The LDB argued that the outcome here should be the same as in Order No. 238-1998. The LDB based this on its contention that Order No. 238-1998: dealt with the same applicant and with one of the same third parties (i.e., Labatt); addressed similar information (i.e., information that would disclose market shares of Labatt and of Molson); and involved some of the same arguments as were made in this case. In Order No. 238-1998, my predecessor upheld the application of s. 21(1) of the Act to a letter from Labatt to the LDB which addressed the positioning of beer products in liquor stores. Although I am not bound to follow previous orders, consistency of approach is valuable (especially where third party interests are involved, as in this case). Order No. 238-1998 is clearly a relevant consideration for me, to the extent it dealt with information and issues similar to those involved here. But that order did not involve the same information as that in issue here, and it may or may not have involved the same issues. I therefore decline to analyze Order No. 238-1998 as if it presented truly the same cause or issues as are involved in this case. 3.4 Section 21 of the Act – Although the LDB’s response to the applicant referred only to s. 21, it is clear that all parties consider s. 21(1) to be the relevant provision. That section reads as follows:
This mandatory section creates a three-part test. A public body is required to withhold third party information only if all three parts of the test have been satisfied. 3.5 Trade Secrets, Commercial or Financial Information – The first part of the s. 21(1) test requires evidence that the disputed information would, if disclosed, reveal a third party’s trade secrets, commercial information, financial information, or other of the specified kinds of information. Pacific Western sensibly conceded at the outset that the "records could be said to reveal commercial or financial information of Labatt and Molson" for the purposes of s. 21(1)(a)(ii). Pacific Western, however, for some reason chose, in its reply, to reverse its position on this point. This is, to say the very least, regrettable. Since the other parties were given a further opportunity to be heard on this point, however, I do not need to decide, in this case, whether such a practice should be curbed by preventing a party from reversing itself in this way (absent innocent error in doing so). Pacific Western’s argument was that the use in s. 21(1)(a) of the word "of" limits the application of s. 21(1) to information that belongs to a third party. The LDB disputed this argument and said that, especially in light of the purposes of s. 21(1), the word "of" must be read as meaning "about" or "concerning" a third party. The LDB rightly accepted, however, that the word "of" in s. 21(1)(a)(i), in relation to "trade secrets", includes information that is owned by a third party. I do not entirely rule out the possibility that use of the word "of" in s. 21(1)(a)(ii) has both these meanings. At the very least, the word "of" in s. 21(1)(a)(ii) means commercial or financial information about a third party. There is no doubt that information revealing the annual sales of a company is financial or commercial information of that company and I so find in this case. Molson also argued that the information in dispute qualifies as a "trade secret" for the purposes of s. 21(1)(a)(i). Both in its affidavit evidence and argument, Molson sought to persuade me that the information meets all four elements of the definition of trade secret found in Schedule 1 to the Act. That definition reads as follows:
Molson’s submissions on this point focus on the elements found in paragraphs (a) through (d) of the definition, as opposed to the words following "including" in the first line of the definition. Since I find that the information in issue so clearly falls within the classes of "commercial" and "financial" information, I need not decide whether the information also belongs to the class of information defined as a "trade secret". Nor do I make any finding about Molson’s argument that disclosure of the information would "implicitly" reveal trade secrets. 3.6 Supply of Information in Confidence – The next element of the s. 21(1) test is the requirement, found in s. 21(1)(b), that the information must be information "that is supplied, implicitly or explicitly, in confidence" to the public body. Counsel for Pacific Western did not concede that "records showing historical sales figures were supplied" to the LDB in confidence for the purposes of s. 21(1)(b), since "all other competitors in the market are required to make public their comparative sales figures". Counsel did not say how or when such sales data of others are required to be made public. The LDB’s affidavit evidence contradicts this assertion, a position in which Molson and Labatt joined.
The LDB, Labatt and Molson each pointed out that s. 36 of the Liquor Distribution Act addresses the confidential supply issue. Here is s. 36: For the purposes of Section 21(1)(b) of the Freedom of Information and Protection of Privacy Act, information in the custody or under the control of the branch [the LDB], whether or not supplied to the branch, is deemed to be supplied to the branch implicitly or explicitly in confidence, if the information concerns the branch’s a) acquisition of liquor from a manufacturer, manufacturer’s agent, distributor, authorized importer of liquor or other person who supplies liquor to the branch, or b) sale of liquor acquired by the branch from a person referred to in paragraph (a) This provision clearly establishes that information described in s. 36 is, whether or not supplied to the LDB, deemed to have been supplied to the LDB in confidence for the purposes of s. 21(1). This amendment to the Liquor Distribution Act may have been designed to overcome the fact that a public body may generate sensitive commercial or financial data about a third party that are therefore not "supplied" by the third party to the public body. Molson pointed out that Brewers Distributor Ltd., a company owned by Molson and Labatt, supplies beer to the LDB for Labatt and Molson. Molson said that data supplied by Brewer’s Distributor Ltd. to the LDB are used to generate the annual sales revenue figures for Labatt and Molson that are in issue here. The LDB, Labatt and Molson all argued that s. 36 of the Liquor Distribution Act deems this information to have been supplied to the LDB in confidence. Molson also supplied affidavit evidence to establish that the information is in fact supplied and that it is supplied in confidence. In my view, s. 36 is, in this case, a sufficient answer and on that basis I find that the information in issue here satisfies the requirements of s. 21(1)(b). 3.7 Harm, Loss or Gain – The nub of this case is whether disclosure of the information could reasonably be expected to cause harm of a kind described in s. 21(1)(c). The LDB, Labatt and Molson all argued that disclosure could reasonably be expected to "harm significantly the competitive position" of Labatt and of Molson (s. 21(1)(c)(i)). They also argued that disclosure could reasonably be expected to result in undue financial loss or gain to any person or organization (s. 21(1)(c)(iii)). Apart from arguing that my predecessor’s decision in Order No. 238-1998 essentially dictates the result in this case, the LDB adopted the arguments made by Molson and Labatt on the harm issue. Private Advantage and Public Interest Under the Act It is necessary, first, to deal with a point raised by Molson in setting the context for its submissions. Molson says, "it is crucial to have a proper understanding of this dispute’s context". It says Pacific Western’s access request 17. … is an attempt to exploit public access legislation for purely private commercial advantage. The Act’s avowed objectives, as expressed in s. 2, are to make public bodies more accountable to the public and to protect personal privacy. The current request for access is not concerned with transparency, accountability, or personal privacy. It aims solely to acquire trade secrets and other highly sensitive commercial data that can be used for financial gain. The commercial value of that information is tremendous, but the applicant seeks to obtain it for almost nothing. 18. Pacific Western has established an ongoing pattern of invoking the Act to extract commercial data for competitive advantage. This is plainly not the purpose for which the Act exists. To the contrary, s. 21 of the Act provides a legislative acknowledgement of the need to protect the legitimate interests of private businesses. There is no evidence before me to support Molson’s contention that Pacific Western "has established an ongoing pattern of invoking the Act" for its private advantage. True, an access request by Pacific Western led to Order No. 238-1998, but this alone does not establish a pattern. Nor would such a ‘pattern’, on its own, be a bar to this request. Absent a successful application by the LDB under s. 43 of the Act, for an order permitting it to ignore requests from Pacific Western, past use of the Act by an applicant is not a factor in deciding the merits of a current request. Then there is Molson’s concern that Pacific Western is using the Act for private advantage. Molson did not argue that the Act should be interpreted or applied differently where a requester is seeking a private commercial advantage. I would not, in any case, agree: the Act must be interpreted and applied despite such considerations. No doubt the rights of access created by the Act are intended to serve the public interest in openness and accountability, as s. 2 of the Act demonstrates. Of course, s. 21 itself makes it clear that the Act’s right of access may be used by a business for its private advantage. A requester may, in a given case, be able to acquire information of a competitor, and therefore gain a private advantage, because disclosure would not cause "significant harm" to the competitor. The Legislature contemplated this result when it chose the standard of significant harm for s. 21(1). If recent trends in the forming of public-private partnerships, or in outright privatization of public body services, continue, businesses involved in such arrangements must assume some level of possible exposure through the Act. Public scrutiny of the kinds of public-private business arrangements just described accords with the Act’s legislative goals. The Legislature has tempered the Act’s objectives, of course, by protecting private interests through s. 21(1). Where s. 21 permits disclosure of information, it will accord with the Act’s objectives of openness and accountability. Where s. 21(1) prohibits disclosure, the Act’s public objectives bow to private needs. Standard of Proof Section 21(1)(c) requires a public body to establish that disclosure of the requested information could reasonably be expected to cause "significant harm" to the "competitive position" of a third party or that disclosure could reasonably be expected to cause one of the other harms identified in that section. There is no need to prove that harm of some kind will, with certainty, flow from disclosure; nor is it enough to rely upon speculation. Returning always to the standard set by the Act, the expectation of harm as a result of disclosure must be based on reason. I have, at times, described this standard of proof in terms of whether the release of information is more likely than not to result in the harm described in the relevant section of the Act. See Order No. 323-1999 and Order No. 00-02. Focussing on contextual considerations, however, in Big Canoe v. Ontario (Minister of Labour) 181 D.L.R. (4th) 603, the Ontario Court of Appeal recently rejected the idea that there must be a probability of harm in assessing the risk of harm where an exception to disclosure aimed at personal safety and bodily integrity is involved. The exception involved in that case is the Ontario version of our s. 19(1). On that point, I indicated, in Order No. 00-03, that the application of the "could reasonably be expected" standard should reflect the nature and inherent qualities of the harm which the particular exception is intended to prevent: Although s. 19(1) involves the same standard of proof as other sections of the Act, the importance of protecting third parties from threats to their health or safety means public bodies in the Ministry’s position should act with care and deliberation in assessing the application of this section. A public body must provide sufficient evidence to support the conclusion that disclosure of the information can reasonably be expected to cause a threat to one of the interests identified in the section. There must be a rational connection between the disclosure and the threat. Some of my predecessor’s decisions took the view that ‘detailed and convincing evidence’ of harm was needed for a harm test to be met. This is not inconsistent with what I have just said about this standard of proof. The Ontario Court of Appeal, I note, also upheld a stipulation for ‘detailed and convincing evidence’ in connection with the harm test in access to information legislation. In Workers’ Compensation Board v. Ontario (Information and Privacy Commissioner) (1998), 164 D.L.R. (4th) 129, the Court dealt with a judicial review application regarding a decision under a section of Ontario’s access law that is similar to s. 21(1). Ontario’s Assistant Commissioner had used the words ‘detailed and convincing’ to describe the evidence necessary to prove a reasonable expectation of harm. At pp. 142-143, Labrosse J.A. said the following: [T]he use of the words "detailed and convincing" do not modify the interpretation of the exemption or change the standard of proof. These words simply describe the quality and cogency of the evidence required to satisfy the onus of establishing reasonable expectation of harm. Similar expressions have been used by the Supreme Court of Canada to describe the quality of evidence required to satisfy the burden of proof in civil cases. If the evidence lacks detail and is unconvincing, it fails to satisfy the onus and the information would have to be disclosed. It was the Commissioner’s function to weigh the material. Again, it cannot be said that the Commissioner acted unreasonably. Nor was it unreasonable for him to conclude that the submissions amounted, at most, to speculation of possible harm. To sum up for present purposes, the standard of proof for harms-based exceptions is to be found in the wording of the Act – here, whether the disclosure of information could reasonably be expected to cause the specific harm to be protected against. Evidence of speculative harm will not meet the test, but it is not necessary to establish certainty of harm. The quality and cogency of the evidence must be commensurate with a reasonable person’s expectation that the disclosure of the requested information could cause the harm specified in the exception. The probability of the harm occurring is relevant to assessing the risk of harm, but mathematical likelihood will not necessarily be decisive where other contextual factors are at work. Significant Harm to Competitive Position The next point is that s. 21(1)(c)(i) requires proof that disclosure could reasonably be expected to harm "significantly" the "competitive position" of a third party. Use by the Legislature of the word "significantly" is to be contrasted with the language used in s. 17(1), which protects certain public body interests where "harm" to the public body – including to its financial interests – could reasonably be expected to flow from disclosure of information. By adding the word "significantly" in s. 21(1)(c)(i), the Legislature clearly indicated that something more than "harm" is needed. As is discussed below, by choosing a standard of significant harm, the Legislature clearly contemplated situations where disclosure could simply harm the interests of a private business, but still be permitted. It is neither possible nor wise to attempt an exhaustive definition of what is meant by "harm significantly". It is something more than mere harm, but it is difficult to go further than that in defining it. At the very least, the party bearing the burden of proof must prove that the anticipated harm is, when looked at in light of the circumstances affecting the third party’s competitive position or negotiating position, a material harm to that party’s competitive position. Among other things, in determining whether a feared harm is "significant", the extent of the harm in relation to the assets or revenues of the third party may be relevant. Take an example where disclosure could reasonably be expected to cause a $100,000 loss to a third party. The impact of that loss on the competitive position of a struggling startup business will obviously be greater than the impact of the same loss on a large multi-national corporation. This difference was recognized in Order No. 57-1995, where my predecessor found it "highly unlikely" that disclosure of environmental test results about one gas station site could reasonably be expected to harm significantly the competitive position of Chevron Canada Limited. The LDB and the third parties did not specify, in dollar terms, the competitive harm the third parties could be expected to suffer from disclosure. On Molson’s behalf, it was said in the affidavit of Scott Ellis, Molson’s Vice President of Corporate Affairs for its Western Division, that disclosure of the information would allow competitors to refine existing data about Molson’s business, the result being that "given the magnitude of Molson’s British Columbia sales the economic significance of even a one per cent refinement represents millions of dollars" (para. 72). This was echoed in parts of the affidavit of Paul Smith, sworn in support of Labatt’s case. In this case, as in many others, it is not possible to analyze these issues in quantifiable, dollar terms. Apart from the affidavit evidence just referred to, the evidence in this case is, necessarily, more qualitative than quantitative. Other cases will similarly not permit precise quantification, in monetary amounts, of gain, loss or harm. This does not mean a reasonable expectation of significant harm from disclosure has not been shown. To the contrary, for the following reasons I find that the LDB has established a reasonable expectation of significant harm to the competitive positions of Molson and of Labatt. Both Labatt and Molson readily admit that information similar to that in dispute is often obtained by beer industry players through independent market research. This research – which is costly to conduct - may yield statistics on sales, including by brand, that can be used to estimate competitors’ market share, brand share, pricing strategies or profitability. They say, however, that the information in issue here is much more precise than data gained through such private research. It is therefore jealously guarded by Molson and Labatt and is not available to anyone else by any other means (including through the LDB’s program of selling data, known as the Supplier Data Sales System ("SDSS"). As for the uses to which the information could be put, the following passage from Molson’s initial submission bears quotation at length: If the Annual Sales Revenue Figures were disclosed to Molson’s other competitors, it would allow them to fine-tune their estimates and for the first time identify with some precision the success or failure of Molson’s specific marketing programs, pricing initiative, and SKU [stock-keeping unit] introductions. Molson’s competitors could do so without the heavy costs of market analysis that Molson bore by itself. Molson’s competitors could also, for example:
These submissions were supported by affidavit of Scott Ellis. According to his evidence, the information would permit Molson’s competition to "bring a new level of precision to what have previously remained only estimates". Again, although he candidly conceded that the increased precision "may not exceed a percentage point or even a fraction of a point", Mr. Ellis deposed that even a 1% improvement in clarity would have an impact in the "millions of dollars". At first sight, it is not clear how aggregate annual sales figures for Labatt or Molson could be used to derive the kinds of information just described. How could a competitor glean such information from a lump sum? This point was addressed as follows in Labatt’s reply submission: The risk to Labatt derives from a process where analysts project information through a combination of known facts and assumptions. The more accurate the factual information, the more accurate the conclusions drawn. Analysis of competitive information is by its nature largely dependent on historical data. In fact, one of the biggest competitive advantages enjoyed by any member of the beer industry is its knowledge of both consumer trends and competitive behavior based on historical observation and experience. It would seem obvious that a later entrant to any marketplace is disadvantaged relative to any long-established business. Confirming assumptions made about past behavior with newly acquired factual information allows a competitor to "fine tune" its current reading of the marketplace and plan future activities more precisely. Historical sales data continues to have a great deal of relevance for any competitor who is attuned to using such data in determining business strategies. According to Labbatt and Molson, this analytical exercise would lead to the sharpening of data described in Molson’s initial submission. Certainty as to the actual amount of annual sales would more precisely define the envelope within which, applying other known facts and certain assumptions, a clearer picture could be gained of competitively sensitive information. As I noted above, this process was also addressed in some detail in Scott Ellis’ affidavit. As he put it, the annual sales figures are not harmless; they are, in his view, "the door to a wealth of other highly sensitive information". His affidavit outlined, in detail, how the figures could be used, in conjunction with other data or assumptions, to determine with some accuracy the following information: annual sales volumes, average revenue per hectolitre of production, performance over time, and the relationship over time between Molson’s sales volume and revenue and how specific marketing initiatives affected performance. Since Labatt adopted Molson’s evidence, this evidence relates to Labatt also. I have some trouble with Mr. Ellis’ evidence about how the information could be used to establish the relationship, over time, between Molson’s sales volume and revenue and how specific marketing initiatives have affected performance. My concern stems from the mention in his affidavit of how this information could be derived by blending a competitor’s store audit information with "a proportional breakdown of the Annual Sales Revenue Figures month by month". The information in question is stated on an annual, and not monthly, basis. Perhaps it would be possible to take the annual amounts and – using knowledge of average or store-audited seasonal sales variations – derive the month by month breakdown. Even if Mr. Ellis is mistaken as to the nature of the requested information, however, his evidence remains that other competitive information could be derived from the annual figures. Again, he deposes that the data could be used to fine-tune, to a greater level of precision, estimates already in the hands of Pacific Western and others. Labatt’s evidence is to the same effect. Pacific Western, which does not bear the burden of proof here, did not provide any evidence to rebut that of Scott Ellis and Paul Smith regarding use of the requested information to derive the kinds of competitive information just described. Counsel for Pacific Western argued, however, that it is … unbelievable that they mean to argue that, by disclosing their historical sales figures in British Columbia to the public, anyone could extrapolate without time, cost or effort and with a degree of preciseness all of the items set out in paragraph 73 of the Molson submission [quoted above]. First, it is reasonable to conclude that the industry experience of both Mr. Smith and Mr. Ellis lends credence to their evidence. Second, no one has suggested here that underlying information could be ‘extrapolated’ without effort or cost. To the contrary, it is clear the underlying data could be derived only using data gathered by competitors using costly store audits and other means. In addition to the above-described arguments about harm, Labatt’s and Molson’s evidence speaks to the harm feared from disclosure of this information to foreign competitors. According to Labatt, disclosure to Pacific Western would ultimately amount to disclosure of the data to foreign competition. In his affidavit on behalf of Labatt, Paul Smith deposed that the British Columbia beer market is the third most important in Canada. Mr. Smith deposed that the beer industry in North America is "mature", which has resulted in a "highly competitive marketplace". Access to Labatt and Molson sales data would, he said, "provide significant competitive advantage to any foreign brewer contemplating an entry, or re-entry, into the British Columbia market". Both Labatt and Molson also contended that disclosure of this information could expose them to liability for breach of confidentiality clauses in licensing agreements each has entered into with other brewers. In Molson’s case, its licensing agreements for brands such as Miller, Coors, Heineken and Fosters require Molson not to disclose or share information on sales volume or sales revenue for those brands. Labatt made the same argument regarding its licensing agreements for foreign brands, including Budweiser, Carlsberg and Tuborg. I am not in a position to find that these concerns are well-founded, since none of the licensing agreements has been provided to me. It is not clear whether disclosure of this information under compulsion of law would constitute an actionable breach of any of those agreements. I also note that the LDB says the information in question discloses Labatt’s and Molson’s annual sales figures for "domestic beer products". It is not clear that this class includes sales of foreign brands under licence to Labatt or Molson. For these reasons, I place little weight on this aspect of Labatt’s and Molson’s evidence. Pacific Western made a number of points in support of its case. A consistent theme was the fact that annual sales figures for Pacific Western and others are disclosed by the LDB under the FIA. (Whether this practice will change with LDB’s recent discovery that it is not covered by the FIA remains to be seen.) Labatt and Molson have, for cost reasons, incorporated Brewers’ Distribution Ltd. to distribute their products jointly. Since data published under the FIA are for that corporation, Labatt and Molson are effectively shielded from scrutiny, whereas Pacific Western and others are not. In Pacific Western’s view, the playing field should be levelled by disclosure of the information it has sought under the Act. Here is how Pacific Western’s reply submission made this point: The Applicant submits that no "harm", significant or otherwise, to the Third Parties’ competitive position could reasonably be expected to occur if they were required to publicly disclose the same information [again, the LDB and third parties say the FIA information is not the same] that every other competitor in the market is required to disclose. In fact, the failure to publicly make available the Third Parties’ annual sales figures causes and continues to cause harm to the competitive positions of all other beer manufacturers in British Columbia. The Third Parties are currently able to do the kind of market analysis regarding their competitors that they set out in their submissions (Molson, paragraph 73) as being harmful and likely to cause them loss if their sales figures were disclosed. If publicly making available annual sales figures of the Third Parties does affect the Third Parties adversely or does result in some gain to other parties, these results could hardly be said to be significant or undue. Competing in an open market with equal access to information from all participants cannot be considered a large detriment or an excessive burden. Even if I assume, for the moment, that the requested information and the FIA data are the same, my function is not to help level the domestic beer industry playing field. Labatt and Molson have entered into a business arrangement that incidentally protects certain commercial or financial information. This business arrangement is perfectly legitimate. The same course of action is, in theory at least, open to Pacific Western and others. The fact that Pacific Western or others have not done the same thing cannot be turned on its head and be used to favour disclosure. Another point advanced by Pacific Western was that disclosure by Labatt and Molson to each other, through Brewers’ Distribution Ltd., of their annual sales figures sets a "threshold as to what constitutes harm, loss or gain". As it was put in Pacific Western’s reply submission, If allowing their largest competitor to have a Third Party’s annual sales information is not too harmful to engage in, how can similar disclosure to the public, including to small competitors, be any more harmful? Not only do the Third Parties bear the burden of demonstrating that allowing public access to sales information is more harmful than giving it to your largest competitor, they must demonstrate how it would be "significantly" more harmful, or how it would "unduly" cause financial loss to themselves or gain to others. It is one thing for Labatt and Molson to gauge the situation and decide it makes business sense to disclose information to each other through Brewers’ Distribution Ltd. and quite another to compel them to give the same information to Pacific Western and a host of others in a global brewing industry. Moreover, the fact that Labatt and Molson presumably have made a deal that works for each of them does not, in my view, govern the issue of significant harm that may result from disclosure to Pacific Western or to the world at large. Pacific Western made a similar point about disclosure of the information in issue, by Labatt and Molson, to their various foreign brewer licensing partners. How, Pacific Western argued, could Labatt and Molson claim that their sales figures for certain foreign brands are confidential, because of licensing agreement confidentiality provisions, yet claim that public disclosure of the same data would raise a threat from "those partners". It is reasonable to assume that any data disclosed to a foreign brewer under a licensing agreement relate to British Columbia or Canada sales by Labatt or Molson of that brewer’s brands only. This is not the same as disclosure of the aggregate sales figures for Labatt and Molson as a whole, across all brands. Even if the information so disclosed relates to Molson and Labbatt products, Pacific Western’s point assumes that all foreign competitors are parties to licensing agreements with Labatt or Molson. This is unlikely to be the case, which means the threat of harm at the hands of foreign competition cannot be ignored. There has been no real suggestion here that, because disclosure of FIA data for Pacific Western and others has not harmed them, any claim of significant harm to Labatt or Molson cannot be made or is weakened. In other words, no one is saying that because Pacific Western and others can survive despite disclosure, Labatt and Molson can also. Assuming again that the FIA data and the requested information are the same, the fact that Pacific Western and others can still carry on business despite scrutiny of their sales figures by the competition (including Molson and Labatt) does not mean disclosure of the disputed information would be harmless. As was noted above, Labatt and Molson enjoy – through their legitimate business arrangements – an advantage that Pacific Western and others do not. If that advantage is ignored, and the information is disclosed, it is still possible that significant harm to their competitive positions could result even though others have managed to get by despite disclosure of their data under the FIA. Earlier in this order I said that use of the word "significantly" in s. 21(1) requires proof of harm that, in all the circumstances affecting a third party, is a material harm and not just mere harm. I also noted that the size of a corporation’s revenues or assets could be a factor in deciding whether an anticipated harm is significant. Last, I observed that it might not be possible to quantify, in dollar terms, the amount of anticipated harm. There is evidence in this case that disclosure of the information could be expected to result in losses in the millions of dollars. Even if one assumes that Labatt and Molson each have annual sales in the hundreds of millions – in fact, billions – of dollars, I am persuaded in this case that losses in the "millions of dollars" could reasonably be expected to harm significantly the competitive position of Molson and of Labbatt. In reaching this conclusion, I have also kept in mind the further evidence, adduced by Labbatt and by Molson, that the data could be used to derive competitive information about their operations. The data could lead to a longer-term strategic impact in a way that could reasonably be expected to significantly harm the competitive positions of the third parties. The possibility of even greater losses in this way is relevant to my findings on the harm issue. Undue Financial Gain or Loss Molson argued that disclosure of this information could, within the meaning of s. 21(1)(c)(iii), reasonably be expected to "result in undue financial loss or gain to any person or organization". Labatt agreed with this. Molson submitted that disclosure of the information would cause loss to it because the information would hurt its competitive position, thus causing a loss of revenue. Molson also said it would allow Pacific Western or another competitor to make profits without having invested any capital to do that: "Molson’s competitors would reap unfair monetary benefits, and Molson itself would sustain undue monetary losses". As was noted earlier, the evidence establishes that this information has value. A buyer could be found for it because it would enable a competitor to fine-tune, at the very least, existing data about Molson and Labatt. The information would add value to that data and has value in its own right. Similarly, there is evidence that disclosure of the information could reasonably be expected to cause harm to Molson and Labatt. Whether or not the expected harm is significant, it might also be "undue". The central question is exactly that – would the gain or loss be "undue". When is a financial gain or loss "undue"? As is the case with the significant harm test under s. 21(1)(c)(i), this test obviously requires one to consider what loss or gain might be ‘due’ in trying to define what is ‘undue’. The ordinary meanings of the word "undue" include something that is unwarranted, inappropriate or improper. They can also include something that is excessive or disproportionate, or something that exceeds propriety or fitness. Such meanings have been approved regarding the similar provision in Alberta’s freedom of information legislation. See Order 99-018. The courts have also given ‘undue’ such meanings, albeit in relation to other kinds of legislation. See, for example, the judgement of Cartwright J. (as he then was) in Howard Smith Paper Mills Ltd. v. The Queen (1957), 29 C.P.R. 6 (S.C.C.), at p. 29. As Cartwright J. noted in Howard Smith, above, interpretation of the word ‘undue’ is not assisted by simply substituting different adjectives for that word. That which is undue can only be measured against that which is due. The Legislature did not, however, provide such a frame of reference for the purposes of s. 21(1)(c)(iii). It is necessary, therefore, to approach the issue of what is undue financial loss or gain in the circumstances of each case. This analysis can to some extent be guided by decisions in previous similar cases, which will give some sense of what may be undue in the present situation. In this case, Molson and Labbatt argue, disclosure of the information would give Pacific Western something for nothing. It would be given valuable competitive data without having had to pay for it through independent research or other means. As I understand it, they argue this information would present Pacific Western with a windfall. This issue is considered in a number of my predecessor’s decisions regarding s. 21. Although none of them explicitly offers any guidance on what is "undue" financial loss or gain, it is possible to conclude that David Flaherty considered the s. 21(1)(c)(iii) to be similar, in some ways, to the s. 21(1)(c)(i) test. They are, of course, different tests. The Legislature clearly created two tests under s. 21(1)(c) and each must be given different meaning. Ontario’s Freedom of Information and Protection of Privacy Act is similar to our Act. Section 17 of the Ontario legislation is similar to s. 21 of ours. Under s. 17(1)(c) of the Ontario Act, an institution must refuse to disclose information where the disclosure could reasonably be expected to "result in undue loss or gain to any person". The two provisions differ, clearly, in that the British Columbia version is expressly concerned only with undue "financial" loss or gain. The Ontario version is arguably wider, since it refers to loss or gain generally. Decisions respecting s. 17 of the Ontario Act certainly suggest it is wider, e.g., that damage to reputation can be a loss even if financial value cannot be placed on that damage). See, for example, Order P-1175. (That decision – and Order P-340 – indicate, however, that it is difficult to prove loss of a non-financial kind.) In any case, it is plain that the Ontario and British Columbia provisions both protect against financial gain or loss that is undue. Ontario decisions consistently show that if disclosure would give a competitor an advantage, usually by acquiring competitively valuable information, effectively for nothing, the gain to the competitor will be undue. See, for example, Ontario Orders 125, P-561, P-1105 and M-920. In the last case, the City of Toronto denied access to its contract with a third party computer service provider. The third party successfully argued that disclosure of the contract’s details – including the terms between the third party and its sub-contractors – would enable its competitors to "replicate the company’s technologies and services" and thus would cause it undue loss. The inquiry officer did not find that the result would also be an undue gain to the competition, although such a finding would appear to be supportable in that case. Again, a similar argument was made here (and in Order No. 238-1998, which is discussed above). Perhaps because he found that the test under s. 21(1)(c)(i) had been met, my predecessor made no finding on this issue in Order No. 238-1998. In the circumstances, I find that the evidence supports the argument that disclosure of the disputed would result in undue gain to Pacific Western and undue loss to Labatt and Molson. On the question of gain, it is clear from the evidence that Pacific Western will gain some competitive advantage from disclosure of the requested records. Pacific Western did not seriously dispute this. Nor did it refute the suggestion by Labatt and Molson that Pacific Western’s purpose in seeking the information is purely competitive. I accept the evidence of Labatt and Molson that the information would enable Pacific Western and other competitors to fine-tune existing data, derived from independent and costly research, regarding Labatt’s and Molson’s business in British Columbia. This would give the competition something for nothing. It would give them valuable competitive information for free and that information could then be used to make inroads into the market share of the third parties. In my view, this financial gain to Pacific Western, and others, would be undue. It would not be undue because the gain would be large. The evidence does not permit me to make any finding on the costs saved by Pacific Western if it were to obtain information that it would otherwise have to pay for. Nor does it allow me to decide what price Pacific Western would pay to buy such information if it were available. But the information doubtless has value to Pacific Western and to others. The gain to Pacific Western from having that information would be undue because it would be unfair, and inappropriate, for Pacific Western to obtain otherwise confidential commercial information about two of its competitors and thereby reap a competitive windfall. One final point is necessary in passing. Of course, Pacific Western does not bear the burden of proof here. But it is interesting that Pacific Western did not point to anything that would suggest any gain to it would not be undue, e.g., because of some overriding public interest in disclosure that would render any gain incidental to the public interest advantage and thus not an undue gain. Levelling the playing field in some vague way is not such a factor. Further, the resulting financial loss to Labatt and to Molson would be undue. As was discussed earlier, the evidence shows that the loss to Molson, at least, can reasonably be expected to be in the "millions of dollars". Although the tests of significant harm to competitive position and undue financial loss or gain may in some cases overlap, the tests differ. In this case, the expected loss would be undue both because of its size, or significance, and because it would be inappropriate and unfair. Although it will not necessarily always be true, in this case, at least, the same considerations establish the undue nature of the expected financial loss and the gain. I find that the LDB is required by s. 21(1)(c)(iii) to refuse to disclose the requested information to the applicant. Tax-Related Information and Section 21(2) In Molson’s view, the LDB was required to withhold the disputed information under s. 21(2) of the Act, which reads as follows: The head of a public body must refuse to disclose to an applicant information that was obtained on a tax return or gathered for the purpose of determining tax liability or collecting a tax. This is because, according to Molson, "[o]ne of the purposes" for which data are collected by the LDB is "to determine what taxes the LDB must remit to federal and provincial authorities in respect of the sale of liquor products in British Columbia". In advancing this argument, Molson referred to the ‘Policy and Procedures Manual’ issued by the Province’s Information, Science and Technology Agency to assist public bodies in their application of the Act. Labatt agreed with Molson’s argument regarding s. 21(2). According to the passage in the Manual that Molson and Labatt rely on – which is by no means binding on me – the s. 21(2) phrase "gathered for the purpose of determining tax liability" means that … the information is gathered for the purpose of determining if a person or organization owes past, present or future taxes to a federal, provincial or municipal government. The LDB did not support Molson and Labatt on this issue. In its reply submission, the LDB said s. 21(2) "must be read as protecting only information that was collected for the sole purpose, or perhaps not the sole purpose but at least the primary purpose, of determining tax liability". Here, the LDB said, the information could at best be said to have been collected "only incidentally for the purpose of determining tax liability". I agree with the LDB. In my view, the broad reading of s. 21(2) suggested by Molson and Labatt is not warranted by the language of the section, especially in light of s. 2(1) of the Act. That section says the Act’s goals of openness and accountability are, among other things, to be achieved by "specifying limited exceptions to the rights of access" (s. 2(1)(c)). To be sure, this does not permit a narrow reading of s. 21(2) or any of the other exceptions and I do not read them that way. I do think, however, that s. 2(1)(c) is an important part of the legislative context within which s. 21(2) appears and that it is relevant to the interpretation of the section. The material before me establishes that the data provided to the LDB by its suppliers – and the data generated by the LDB and in dispute here – are supplied for a whole host of reasons. Judged on that material, I cannot agree that the disputed information was gathered for the purpose of determining tax liability or collecting a tax. I note, also, that the information in dispute consists of annual sales figures for each of Molson and Labatt for each of several years. Molson says these figures "are an annual cumulation" of the "weekly ‘display dollar’ data streams" submitted to the LDB by Molson and Labatt. It follows, Molson says, that the figures fall under s. 21(2). The records sought by Pacific Western were created in response to its request; the lump sum figures found in the records were created by the LDB from underlying data (including, in part, data provided by Molson and Labatt). Strictly speaking, however, the information sought by Pacific Western was not, directly, "gathered" by the LDB. On that basis alone, the information is not covered by s. 21(2). 4.0 CONCLUSIONFor the reasons given above:
April 19, 2000 ______________________ David Loukidelis Information and Privacy Commissioner for British Columbia
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||