Canadian Copyright Licensing Agency (Access Copyright) v. British Columbia (Education)

Canadian Copyright Licensing Agency (Access Copyright) v. British Columbia (Education)
Court (s) Database
Federal Court of Appeal Decisions
Date
2017-01-27
Neutral citation
2017 FCA 16
File numbers
A-93-16
Decision Content
Date: 20170127
Docket: A-93-16
Citation: 2017 FCA 16
CORAM:
NADON J.A.
DAWSON J.A.
GAUTHIER J.A.
BETWEEN:
THE CANADIAN COPYRIGHT LICENSING AGENCY, OPERATING AS ACCESS COPYRIGHT
Applicant
and
BRITISH COLUMBIA MINISTRY OF EDUCATION AND ALL ENTITIES NAMED IN SCHEDULE “A” HERETO
Respondents
Heard at Ottawa, Ontario, on November 22, 2016.
Judgment delivered at Ottawa, Ontario, on January 27, 2017.
REASONS FOR JUDGMENT BY:
GAUTHIER J.A.
CONCURRED IN BY:
NADON J.A.
DAWSON J.A.
Date: 20170127
Docket: A-93-16
Citation: 2017 FCA 16
CORAM:
NADON J.A.
DAWSON J.A.
GAUTHIER J.A.
BETWEEN:
THE CANADIAN COPYRIGHT LICENSING AGENCY, OPERATING AS ACCESS COPYRIGHT
Applicant
And
BRITISH COLUMBIA MINISTRY OF EDUCATION AND ALL ENTITIES NAMED IN SCHEDULE “A” HERETO
Respondents
REASONS FOR JUDGMENT
GAUTHIER J.A.
[1] The Canadian Copyright Licensing Agency, operating as Access Copyright (Access) seeks judicial review of a decision of the Copyright Board of Canada (the Board), certifying the royalty rates to be collected by Access for the reproduction of works in its repertoire by elementary and secondary educational institutions (K-12 schools) represented by twelve provincial and territorial ministries of education (outside of Quebec) and all Ontario school boards (collectively the Consortium) during the 2010-2012 (First Tariff) and 2013-2015 (Second Tariff) tariff periods.
I. Background [2] In its decision, the Board approved an annual royalty rate of $2.46 per full-time equivalent student (FTE) for the First Tariff and $2.41 per FTE for the Second Tariff. It applied a “volume times value” methodology, pursuant to which the volume of compensable copying is multiplied by the estimated value of each page of the copied work for one of the four genres included in Access’ repertoire — books, periodicals, newspapers and “consumables”.
[3] This methodology had been proposed by the parties and adopted by the Board in its previous decision issued in June 2009 dealing with a previous tariff of Access for K-12 schools (Access Copyright (Educational Institutions) 2005-2009 (26 June 2009) [K-12 (2009)]). This decision was challenged before our Court and was ultimately quashed by the Supreme Court of Canada in Alberta (Education) v. Canadian Copyright Licencing Agency (Access Copyright), 2012 SCC 37, [2012] 2 S.C.R. 345 [Alberta]. As the Supreme Court had identified several issues that rendered the Board’s assessment of the fair dealing exception unreasonable, the current decision under review was meant to comply with the teachings of Alberta. The Board held a hearing over nine days in April, May and September 2014. It heard the oral arguments in September 2014. In June 2014 and after the September hearing, the Board put several technical questions to the parties and the evidentiary record was only perfected on December 19, 2014 once the parties answered the Board’s additional questions. The decision was issued in February 2016.
[4] Because of the high cost of the evidence required to apply the methodology referred to at paragraph 2 above, the parties agreed to rely on the results of the volume study, previously undertaken by them in 2005-2006, as a reasonable proxy for the copying taking place during the two tariff periods at issue here.
[5] As mentioned, the Board, in applying the agreed methodology, had to determine the volume of compensable copying or compensable exposures. To do so, it had to first identify which of the copied works came within Access’ repertoire. Then, the Board had to assess which of those copying events reproduced a “substantial part” of the work involved within the meaning of section 3 of the Copyright Act, R.S.C., 1985, c. C-42 (the Act).
[6] The next step required the Board to exclude from these compensable exposures the copying that should be excluded from compensability pursuant to any applicable provision of the Act. The only exclusion relevant to this proceeding is the copying that falls within the user’s right as set out at section 29 of the Act, which provides that fair dealing for the purposes of education and private study does not infringe copyright. The Board also deducted any copying required for a test or examination (subsection 29.4(2) of the Act) as well as those for display for the purposes of education or training (subsection 29.4(1) of the Act). These findings are not challenged.
[7] Access accepts most of the findings made by the Board in its detailed and lengthy decision. With respect to what works should be included in its repertoire, Access only contests the Board’s decision to disregard any errors in coding made in the volume study in respect of who owned the copyright, as well as its decision to exclude from the volume of compensable exposures any copying of a book that included less than one or two pages per copying event on the basis that these events did not involve the reproduction of “a substantial part” of the work within the meaning of the Act (see section 3 of the Act in Appendix).
[8] The other six issues raised by Access all relate to the deductions made to the total number of compensable exposures on the basis of fair dealing in respect of books, newspapers and periodicals, including the methodology used by the Board to quantify those deductions, which Access argues was procedurally unfair and fundamentally flawed.
[9] Rather than summarizing the lengthy decision of the Board under a distinct heading, I will refer to the most relevant findings when dealing with each issue. I will also refer to the applicable standard of review under each separate heading.
II. Analysis [10] I have regrouped the issues raised by Access under the following headings:
A. Repertoire
(1) Did the Board err in ignoring expert evidence provided by Access to correct and clarify the breadth of its repertoire as described in the volume study?
(2) Did the Board err in restricting the “substantiality” (term used by the parties) of compensable exposure under section 3 of the Act?
B. Fair Dealing
(1) Did the Board err in its application of the burden of proof?
(2) Did the Board breach its duty to act fairly?
(3) Was the Board’s methodology unreasonable and did it err in assessing the relevant factors?
A. Repertoire (1) Did the Board err in ignoring expert evidence provided by Access to correct and clarify the breadth of its repertoire as described in the volume study? [11] In K-12 (2009), the Board found that the works captured by the volume study that were published by non-affiliated rights holders (NARH) would be included in Access’ repertoire for the purpose of calculating the tariff on the basis of an implied agency relationship where the NARH ratified Access’ administration of their rights by accepting the payment of royalties (K-12 (2009) at para. 133).
[12] In the proceeding at issue, the Consortium objected to such an inclusion with respect to the First Tariff and Second Tariff stating that this category of rightholders should not be considered. In its reasons (Reasons), the Board rejected Access’ argument that issue estoppel applied because the decision was only quashed by the Supreme Court in respect of the fair dealing exclusions. In refusing to apply “issue estoppel”, the Board exercised its discretion to ensure the fairness of the First Tariff and the Second Tariff. Additionally, the Board noted that Access itself sought to revisit other findings made during the K-12 (2009) proceeding (Reasons at paras. 125-126). The Board’s conclusion on issue estoppel is not contested by Access in the present proceeding.
[13] Because of its position that issue estoppel applied, Access says that there was no need for it to correct the coding errors made in the volume study where some works were wrongly attributed to NARH or no coding at all was given (0). Indeed, the Board had included all these works in Access’ repertoire in its K-12 (2009) decision.
[14] However, this information became more relevant to answer specific technical questions posed by the Board that asked Access to confirm and clarify the meaning of certain fields in the data used, including identifying who signed agreements with Access and whether such entity owned the copyright (see Access’ letter to the Board dated October 14, 2014).
[15] In Access’ reply letter dated October 14, 2014, Access raised the issue of coding errors in response to the Board’s technical questions. It further explained that if the Board rejected its issue estoppel argument, Access would need to conduct a more detailed analysis of the data in order to correct the breadth of its repertoire since the coding errors had the effect of significantly underestimating the number of copied works comprised in Access’ repertoire.
[16] After the Consortium submitted its own expert report dated October 14, 2014 (also filed in reply to these questions of the Board), Access did provide such a detailed report and quantified the impact of the underestimation. The expert report prepared by Circum Network Inc. (Circum) dated November, 28, 2014 is attached to a letter dated December 5, 2014. This report makes it clear that the calculations provided by the Consortium’s experts should not be accepted because of the underestimation resulting from the Consortium’s assumption that only works expressly attributed (i.e. coded as such) to an affiliated righholder or to organizations in other jurisdictions (referred to as “Reproduction Rights Organizations” (RROs)) should be considered as part of Access’ repertoire.
[17] Access’ December 5, 2014 submissions and expert report dated November 28, 2014 were accepted by the Board and exhibit numbers (AC-114 and AC-114A) were attributed to this evidence.
[18] The Consortium did not object to the filing of this evidence, nor did it offer any comments as to its validity. In fact, the Consortium itself filed on December 5, 2014, another expert report in reply to Access’ October 14, 2014 letter. The December 5, 2014 letter did not offer any comment by the Consortium in reply to Access’ submission that its repertoire was underestimated in the Consortium material because of coding errors. As mentioned, the record was perfected on December 19, 2014.
[19] It is not disputed that if Circum’s calculations had been accepted, this would represent a sizeable increase in the royalties to be paid to Access, which it estimates to represent approximately $500,000.00 per year, or $3 million dollars over the two tariff periods. Before us, the Consortium did not offer any other estimate of the potential impact of these coding errors, saying that it would have to review the validity of Access’ calculations should the matter be reconsidered by the Board as its experts had not yet had the opportunity to comment on Access’ calculations.
[20] In its Reasons, the Board deals with this question as follows at paragraph 405:
[405] In a filing responding to the Board’s technical questions, Access explained some problems with these variables.261 In particular, Access claimed that using these variables to measure the volume of copying of works owned by its affiliates or authorized by bilateral agreements with RROs greatly underestimates the volume of such copying. We reject this claim for three reasons. First, Access has provided no evidence of the degree of underestimation. Second, Access has had many years to correct the underestimation but has chosen not to do so. Third, to the extent that the underestimation is related to works that were not in Access’ repertoire in 2005-2006 when the copies were made but now are, we do not want to make that correction. (Emphasis added)
[21] Footnote 261 mentioned in the quote reproduced above refers to Access’ letter dated October 14, 2014 where, as mentioned, Access only raised the issue of coding errors and how it could have a serious impact on the Board’s calculation of compensable exposures.
[22] Although the Consortium argued that this finding was based on the weight given to the evidence by the Board, a matter with which this Court should not lightly intervene, it is difficult to conclude anything other than that the Board, through oversight, overlooked the expert evidence and submissions it accepted as exhibits AC-114 and AC-114A on December 5, 2014.
[23] The Board’s clear wording that Access provided no evidence rebuts the presumption that a decision-maker has considered all the evidence before it.
[24] There is no ambiguity in the reasons expressed at paragraph 405 of the Board’s Reasons reproduced above. The Board failed to consider that expert evidence had been filed to estimate the degree of the underestimation, that Access had chosen to correct the underestimation and that it explained in detail why it had not done so before. Access’ statement that the corrections proposed by its expert did not result from an expansion of Access’ repertoire in the years subsequent to the data collected in the 2005-2006 volume study was not challenged before us. Thus, the Board’s refusal to consider whether the repertoire was underestimated is unreasonable.
[25] This Court is not in a position to assess the weight, if any, to be given to Circum’s report dated November 28, 2014. This issue is so clearly material that in my view, the Court should intervene and require the Board to assess the impact, if any, on the volume of compensable exposures. I note, however, that this is such a discrete issue that the parties may well be in a position to facilitate this exercise by jointly proposing the necessary adjustments that could then be approved by the Board.
(2) Did the Board err in restricting the “substantiality” (term used by the parties) of compensable exposure under section 3 of the Act? [26] As mentioned earlier, the first issue to be determined by the Board was the volume of compensable exposures. To do this, it had to determine if all the copying reported in the volume study reproduced “the work or any substantial part thereof” within the meaning of section 3 of the Act.
[27] The Board reviewed the principles applicable to this exercise at paragraphs 212-217 of its Reasons. Access agrees that although it had initially characterized the issue before us as a question of law, the Board properly articulated the legal principles. The Board considered the Supreme Court of Canada’s teachings in Cinar Corporation v. Robinson, 2013 SCC 73, [2013] 3 S.C.R. 1168 [Cinar], which states that a qualitative assessment is required to determine whether “a substantial portion of the author’s skill and judgment” has been copied (Cinar at para. 26). Thus, whether the Board misapplied these principles to the facts of this case is a question of mixed fact and law reviewable on the reasonableness standard.
[28] To meet its burden of establishing that the copying captured by the volume study should all be considered as coming within the ambit of section 3 of the Act, Access proposed two approaches: first, it suggested that the Board should assume that if a teacher of K-12 schools values the ability to copy certain portions of a work, even very small excerpts from a work within Access’ repertoire, it must be because the excerpts are qualitatively relevant and as such, can never be considered unimportant or unsubstantial copying. Second, Access presented some evidence supporting its view that even one or two pages of a book may constitute a substantial part of the book from which it is taken. The relevant evidence in respect of “substantiality” is set out in footnotes 42 to 44 of Access’ Memorandum of Fact and Law and in its Compendium, with the most relevant evidence having been reproduced at tabs 20 to 22 of the Compendium.
[29] Access submits that the Board failed to properly consider this evidence, including, more particularly, some samples of pages copied from books, which should have enabled the Board to confirm the qualitative value of the content, of even as few as two pages.
[30] Finally, Access argues that the Board did not give due consideration to the evidence of compound copying; the Board did not include it in its calculations. According to Access, the Board could not set any bright-line rule based on the number of pages copied. While there were instances where one or two pages from the same book were copied more than once, it did not necessarily mean that the same pages were copied.
[31] There is little to be said about Access’ argument that the Board did not consider compound copying. It is clear that the Board was aware of this argument. It devoted several paragraphs to it (see Reasons at paras. 188-202). At the hearing, when asked what evidence would have enabled the Board to estimate what adjustment should be made on that basis, Access acknowledged that there was no such evidence. I have not been convinced that the Board made a reviewable error in concluding that, based on considerations described in paragraph 202 of its Reasons, it would not make any adjustments on the basis of compound copying. This is especially so when one considers the Board’s comments at paragraph 195 that Access had acknowledged that it was not possible to empirically assess the incidence or volume of compound copying captured by the volume study. Thus, although the Board agreed that compound copying was relevant in theory, in practice, it could simply not measure this phenomenon.
[32] Given that Access raises an issue with respect to the Consortium’s burden of proof to establish the fairness of its dealing, it is worth mentioning that Access bore the legal burden of establishing that all copying in the volume study constituted potential violations of the copyright in the works of its repertoire. In theory, this means that it had to satisfy the Board that each copying event involved a substantial part of a protected work within its repertoire. As there were thousands of copying events involved, Access obviously could not produce a case by case qualitative assessment.
[33] The Board discussed the testimony of Ms. Gerrish, Access’ main (if not the only) witness on this issue (Reasons at paras. 218-220). It concluded that she provided anecdotal evidence that did not provide the Board with a reasonable basis on which to appreciate the qualitative characteristics of portions of books in Access’ repertoire that were actually copied (Reasons at para. 220). The Board refused to draw the inference proposed by Access that because in the one or two books identified by Ms. Gerrish, one or two pages could represent a substantial amount of skill, labour and judgment in certain books (only one textbook illustration was given), it should infer that this was so in respect of all the books referenced in the volume study (Reasons at paras. 217, 220). However, the Board did draw such an inference in respect of newspaper and magazine articles on the basis that these were much shorter works which, by their nature, could be treated differently (Reasons at para. 225).
[34] The Board expressly rejected the proposition that what is worth copying is prima facie worth protecting, having found that this test had been discarded long ago (see footnote 136 of the Reasons). At the hearing before us, it became quite clear that Access could not explain how the copying choices made by a teacher for the purposes of preparing a lesson are related to assessing “substantiality”. The Board came to this same conclusion when it found that copying for the purposes of meeting a student’s educational needs is not a suitable proxy for substantiality (Reasons at para. 217). This seems altogether reasonable to me in the context of tariff setting proceedings.
[35] The Board also found that the subsample produced by Access, where actual copies of the excerpts involved in the copying event were produced, was too small to serve as a basis for the qualitative assessment of all the books in the repertoire.
[36] The Board did not expressly state that little could be gained from a review of these pages without the benefit of an expert or even a lay person that would carry out a reasoned analysis. I believe that it is implicit in the Reasons that a simple review of this limited evidence would not normally be sufficient to reach an appropriate conclusion on substantiality. This is confirmed by the Board’s comments about how quantitative assessments are done when referencing Cinar where numerous expert witnesses, conflicting testimony and voluminous supporting evidence were produced to determine the qualitative part of the work taken (see Reasons at para. 222).
[37] I have not been persuaded by Access that the Board ignored any evidence produced by Access. Indeed, the Board considered it and found that it did not provide a reasonable basis to assess the qualitative nature of the thousands of copying excerpts at issue.
[38] That said, the Board had a couple of options. First, if it had applied the approach proposed by Access in respect of the Consortium’s burden of proof under the fair dealing analysis (see paragraph 81 below), the Board could have concluded that Access had not established that the exposures reported in the volume study amounted to reproduction of “a substantial part of the books” in its repertoire (this is the only genre in respect of which the finding of the Board is contested) given the lack of probative evidence produced by Access in this respect. Second, and what the Board chose to do, was to determine that because it did not have the benefit of a qualitative analysis applicable to the majority of cases, it was reasonable in the particular circumstances of the matter before it (fulfilling its statutory mandate to set a tariff where the parties only presented evidence on an aggregate basis) to infer that the copying of one or two pages from a book was not qualitatively substantial. This approach resulted in a smaller volume of copying being classified as non-substantial than if a threshold of 1% of each such work was adopted (Reasons at paras. 226-227).
[39] It is the task of a tribunal or trial court to fulfil its mandate, despite the paucity or quality of the evidence before them. Such decision-makers must determine if they are satisfied that a certain question of fact has been established. This task is at the very core of the expertise of tribunals such as the Board. Inferences, like findings of facts, are owed considerable deference.
[40] In my view, in the particular circumstances of this case, and considering the mandate of the Board under the Act, it was not unreasonable for the Board to infer that the copying of one or two pages of a book did not constitute reproduction of a “substantial part of the work” within the meaning of section 3 of the Act. It should be clear however that, in my view, such an inference would rarely be within the range of acceptable outcomes when there is evidence produced about each work at issue and would normally constitute an overriding and palpable error in the context of civil litigation proceedings where infringement is at issue.
[41] Finally, I note that considering the application of fair dealing and of section 29.4 of the Act to the exposures, most of the so called “non-substantial copying” in respect of books would have been deducted anyway from the compensable exposures (see tables 24 and 25 of Appendix A to the Reasons).
B. Fair Dealing [42] Before reviewing the parties’ arguments under this heading, a few general comments are warranted.
[43] In Alberta, Justice Abella, writing for a majority described the concept of fair dealing and the test to be applied as follows:
[12] As discussed in the companion appeal Society of Composers, Authors and Music Publishers of Canada v. Bell Canada, [2012] 2 S.C.R. 326 (SOCAN v. Bell), the concept of fair dealing allows users to engage in some activities that might otherwise amount to copyright infringement. The test for fair dealing was articulated in CCH as involving two steps. The first is to determine whether the dealing is for the allowable purpose of “research or private study” under s. 29, “criticism or review” under s. 29.1, or “news reporting” under s. 29.2 of the Act. The second step of CCH assesses whether the dealing is “fair”. The onus is on the person invoking “fair dealing” to satisfy all aspects of the test. To assist in determining whether the dealing is “fair”, this Court set out a number of fairness factors: the purpose, character, and amount of the dealing; the existence of any alternatives to the dealing; the nature of the work; and the effect of the dealing on the work.
[44] Justice Rothstein, writing in dissent, made useful and indisputable comments when he wrote:
[39] … This appeal is about fair dealing under s. 29 of the Copyright Act, R.S.C. 1985, c. C-42 (“Act”). Whether something is fair is a question of fact (CCH Canadian Ltd. v. Law Society of Upper Canada, 2004 SCC 13, [2004] 1 S.C.R. 339, at para. 52 (“CCH”)). Fair dealing is “a matter of impression” (CCH, at para. 52, citing Hubbard v. Vosper, [1972] 1 All E.R. 1023 (C.A.), at p. 1027). In CCH, this Court found that the factors proposed by Linden J.A., at the Federal Court of Appeal (2002 FCA 187, [2002] 4 F.C. 213, at para. 150), to help assess whether a dealing is fair, provided a “useful analytical framework to govern determinations of fairness in future cases” (para. 53). While useful for purposes of the fair dealing analysis, the factors are not statutory requirements. (Emphasis added)
[45] In Alberta, the Supreme Court focused on fair-dealing for the purpose of private study. The Court had to deal with the viewpoint from which fair dealing for this purpose is to be assessed – the teacher or the student, particularly when multiple copies are made for one or more classes. Shortly thereafter, the Act was amended to include “education” as another purpose in respect of which users could rely on section 29 of the Act. In my view, this addition removed the dichotomy between teachers’ or students’ viewpoints under the section 29 analysis, when education is the relevant purpose.
[46] It is also well-known and reiterated in Alberta that the factors set out in CCH originate from the decision of Lord Denning in Hubbard v. Vosper, [1972] 1 All E.R.1023 (C.A.), at 1027. A review of these last two decisions makes it abundantly clear that not all the fairness factors are relevant in all cases nor is any one factor usually determinative.
[47] To fulfill its mandate, the Board had to balance the public interest in compensating the copyright owners for the taking of substantial parts of their work against the public interest in giving certain users the right to reproduce such parts for certain purposes including education and private study.
[48] This is what the second step of the test established in CCH and applied in Alberta is meant to do.
[49] The Board dealt with fair dealing in section XIII of its Reasons (see paras. 229-351). It then described its statistical approach to fair dealing at paragraphs 418 to 457. Finally, its calculation of same can be found in Appendix B to the Reasons, which starts at page 168 of the decision.
[50] I will now turn to the first issue raised by Access in respect of the Board’s analysis of fair dealing. But first, for ease of comprehension, I will re-enumerate the recognized six fair dealing factors. They are: the purpose, character, and amount of the dealing; the existence of any alternatives to the dealing; the nature of the work; and the effect of the dealing on the work.
(1) Did the Board err in its application of the burden of proof? [51] Access states that, contrary to what it did in other decisions, the Board did not expressly refer to the burden of proof in its reasons except to recognize the general principles set out in CCH and Alberta. Access submits that there are sufficient indications in the Reasons and in the methodology used by the Board on which to conclude that the Board failed to properly apply the legal and evidential burden imposed by law on the Consortium.
[52] In Access’ view, the Consortium had to meet its burden of establishing that the first five factors tended to make its copying activities fair. Access recognizes that because it was in possession of more information than the Consortium in respect of the sixth factor (effect of the dealing), it had to provide evidence in respect of that factor. In its view, it met this burden. This issue is dealt with later on in these reasons.
[53] Based on its comments at paragraph 350 of the Reasons, that “the parties did not adequately address fair dealing,” Access argues that the Board should have concluded that the Consortium had failed to meet its burden under the second step of the test for the application of the fair dealing use and refuse to deduct any exposure from the volume of compensable exposures that are in dispute.
[54] Finally, Access states that by creating a neutral category in the statistical methodology adopted, the Board disregarded the applicable burden of proof (see outline of oral argument at para. 41). In its view, anything falling in the neutral category should be considered as not fair (i.e. in the unfair category). Access submits that even if the Board, as an administrative decision-maker, did not have to follow formal rules of evidence, it was still bound to apply the legal burden as expressed by the Supreme Court in Alberta (see Alberta at para. 12).
[55] I agree that, as a matter of law, the Board cannot ignore the burden on the Consortium to establish that it was entitled to the application of section 29 of the Act. This question is reviewable on the correctness standard. However, I have not been persuaded by Access’ arguments that the Board disregarded the burden of proof in this case.
[56] The question of who bears the legal burden is rarely relevant when reaching a conclusion based on the evidence. It is usually only determinative in cases when there is no evidence or no evidence capable of establishing a fact, or when the evidence is so evenly weighted that a decision-maker will determine an issue on that basis.
[57] In this case, to put the Board’s comments and findings in context, it is useful to review in broad terms how the parties chose to fulfill their respective burdens in respect of the second step of the fairness test.
[58] The Consortium chose to present its case using two different approaches. First, the Consortium presented evidence in respect of guidelines issued in 2012 (see Copyright Matters! Some Key Questions & Answers for Teachers, 3rd ed., Respondents’ Record [RR], Vol. 1 at Tab 4) (the Guidelines)), and the fact that they had been widely distributed to K-12 school teachers. This was presented as evidence of a general practice of the type referred to at paragraph 63 of CCH. The Consortium argued that any copy made following those Guidelines would necessarily be fair (presumably these would only be relevant for the period covered by the Second Tariff). Ultimately, the Board found that it could not rely on the Guidelines for the purpose of setting the royalty rates (Reasons at paras. 233-234).
[59] Although both parties were clearly disappointed by the fact that the Board did not offer any detailed comments on their evidence relating to those Guidelines, Access did not challenge this finding, which was based on its assessment of the weight of the evidence. This was a wise decision, for indeed, the Board’s conclusion was clearly open to it on the evidentiary record.
[60] Access argued extensively in its memorandum (not at the hearing or in its outline of oral argument) that the Board was wrong to discard the Guidelines as they were the best evidence of the behaviour to be assessed to determine the issue of fairness. This resulted, according to Access, in the rejection of what Access believed was the Consortium’s better case. Yet, in my view, Access does not indicate how the Board’s actions on this point render its analysis unreasonable.
[61] That said, contrary to Access’ submissions, the Guidelines were not the only evidence tendered by the Consortium to meet the second part of the CCH test (i.e. weighing the fairness factors).
[62] It is apparent from a review of the expert report filed by the Consortium (RR, Vol. 2 at Tab 17) that the Consortium did present a second approach based on an evaluation of the CCH factors. I note that the Consortium’s experts even offered alternative calculations, for example, in respect of “the amount of the dealing” factor, although it assumed based on the instructions received, that reproduction of 10% or less of a book would be considered fair, the said experts also calculated the impact of the Board’s finding that only the reproduction of 7%, 5%, 3% or 1% of each work would be fair (RR, Vol. 2, Tab 17 at 438).
[63] Access had, for its part and as mentioned, marshalled evidence in respect of the sixth factor given that in its view, this factor militated towards finding that the dealing was not fair because of the considerable effect that copying had on the market for those works. Its expert had also calculated what would be fair by using and adjusting the Board’s previous calculations in 2009 to account for what it considered fair based on Alberta (Applicant’s Record [AR], Vol. 1, Tab I at 887-889). Access had also taken the position that the Guidelines were flawed and indeed promoted unfairness (Reasons at para. 231). In its view, the Board could only deduct the exposures that Access had conceded met the fairness test.
[64] It is now appropriate to put paragraph 350 of the Reasons on which Access relies in context by reproducing paragraphs 340 to 350:
[340] Unless we were to accept, in their entirety, the contentions of one party or the other on fair dealing—which we do not—the calculations by the parties cannot be accepted as they are. Nor is it apparent that the data can be readily disaggregated. This poses some difficulties in the evaluation of the amount of fair dealing that we expect occurred during the 2010-2015 period.
[341] Since fair dealing is a matter of impression, one approach would be to consider the copying as a group, or in groups, the latter of which the Board did in the K-12 (2009) decision and its redetermination in the K-12 (2013) decision.
[342] In K-12 (2009), the Board identified four categories of copies that met the first step (i.e., the purpose) of the fair-dealing test. It then considered whether such copies were fair. Following the redetermination in the K-12 (2013) decision, all four categories were found to be fair dealing. In other words: all copies that were identified as having been done for a permitted fair-dealing purpose were found to be fair.
[343] In relation to genres that were compensable in K-12 (2009) (namely: books, newspapers, and magazines) the evidence does not suggest that the copying identified as Categories 1 through 4 in K-12 (2009) has characteristics that differ from copying that was not placed into a category. The only potential difference between copying that was placed into Categories 1 through 4 and that which was not, is in their purpose of the dealing and goal of the dealing.
[344] As noted above in our consideration of the purpose of the dealing, in Part XIII.C, approximately 75 per cent of all copying of books, newspapers and magazines in this matter was done for the purposes of “student instruction, assignments and class work,” and qualifies to be considered for fair dealing, but was not considered for fair dealing in the K-12 (2009) decision nor in its redetermination in K-12 (2013). As per our discussion in our consideration of the goal of the dealing,230 above, the goal of such a dealing would tend towards fairness.
[345] Given that copying identified in Categories 1 through 4 were all found to be fair, and given that copying done for the purposes of “student instruction, assignments and class work” shares the characteristics of copying identified in those categories, were copying of books, newspapers, and magazines to be considered in a group, or groups, it is likely that we would conclude, in relation to those copies made for a permitted purpose, that such dealings were fair.
[346] However, it is unlikely that it is actually the case that all copies under consideration that were made for a permitted purpose are non-infringing. In the absence of evidence of a sufficiently followed practice, and among such a large and varied number of institutions, the approximation created by such a group-based approach likely be too rough of a measure.
[347] We therefore require some means of establishing an actual measure of fair-dealing copying. In this matter, since the data adduced by the parties is at the aggregate level, we approximate the amount of fair-dealing copying by using this data.
[348] We generally proceed with our calculations in the same manner as the parties, by determining the number of copies that were made for a permitted purpose, and, of those, how many were fair. However, in order to use the aggregate information in evidence, we must make the assumption that the characteristics of copying (such as the goal of the dealing, the amount of the dealing, or nature of the work) are independent of one another. For example, whether a copy is made for one purpose or another, the amount of the work copied is not dependant on the purpose. This assumption is necessary, since the data that was adduced by the parties from the Volume Study does not let us correlate such characteristics with one another with any confidence.
[349] Given that the information in relation to consumables, which were not compensable under the K-12 (2013) decision, is also drawn from the 2006 Volume Study, and was also provided to the Board in aggregate form, we use the same method for approximating the amount of fair dealing in relation to consumables as well.
[350] The full methodology and calculations are discussed in Part XVI.E and in Appendix B. The methodology is of our own design, inspired however by submissions of the parties, particularly those of the Objectors. The calculations use data that is part of the evidence. Our assumptions and inferences are also based on the evidence. Because we are of the opinion that the parties did not adequately address fair dealing, we had no choice but to fashion a methodology of our own.
[65] On a fair reading of paragraph 350 of its Reasons, it is clear that the Board did not err as Access alleges. In my view, the Board did not find that the Consortium failed to file evidence that was capable of meeting its burden. Rather, the Board was not prepared to accept the calculations and assumptions of either party. Thus, it had to use the data produced in evidence to make its own inferences and calculations of what would in this case be fair. The Board’s purpose was to come to its own “impression” of what was fair in the best manner possible considering that both parties had agreed to use aggregate data to establish the volume of copying during the relevant period.
[66] Turning to the neutral column included by the Board in its methodology, I understand that the Board classified exposures of books in this category when the evidence adduced did not help the Board to form an “impression” either way as to the fairness of the dealing. While classifying an exposure as neutral may have affected the Board’s overall fair dealing “impression” with respect to books, I do not agree that unless an exposure is considered as tending to fairness, it can only be viewed as tending to the unfairness of the dealing. This would make each factor a statutory requirement that must be met and mean that each factor necessarily applies in the same way to all copying events. As mentioned, this is simply not the law. I cannot conclude from the simple fact t