Copyright Board Canada

Commission du droit d’auteur Canada

 

Date	2019-12-06
Citation	CB-CDA 2019-082
Regime	Copyright Act, section 70.1
Members	The Honourable Robert A. Blair Mr. Claude Majeau Mr. J. Nelson Landry
Proposed Tariffs Considered	Post-Secondary Educational Institutions (2011-2014) Post-Secondary Educational Institutions (2015-2017)

statements of royalties to be collected by access copyright for the reprographic reproduction, in canada, of works in its repertoire

Reasons for decision

I. INTRODUCTION AND HISTORY OF THE PROCEEDINGS 1

A. The Tariffs 1

B. Overview and Context 1

C. Pre-Hearing Matters 4

i. Objectors 4

ii. Interim Tariff and Subsequent Changes 4

iii. Other Relevant Procedural History 6

D. Post-Hearing Developments 8

II. EVIDENTIARY FRAMEWORK 9

A. The Licensing History 9

i. Before 2015 10

ii. After 2015 11

B. Evidence 12

i. Access 12

a. Fact Witnesses 12

Roanie Levy and Jennifer Lamantia 12

Michael Andrews 14

David Swail 14

Don LePan 15

John Degen 15

b. Expert Witnesses 16

Benoît Gauthier 16

Professor Michael Murphy 17

Michael Dobner 17

Bradley Heys 18

c. Conclusions Respecting Access’ Evidence 19

ii. Sean Maguire 19

III. PARTIES’ POSITIONS AND PROPOSED TARIFFS 19

A. Access 19

B. Sean Maguire 21

IV. LEGAL ANALYSIS 22

A. The Fair Dealing Landscape and Its Effect 22

B. Non-Affiliated Rights Holders 24

i. Affiliation Agreements and Non-Affiliates 25

ii. Analysis 26

iii. Implied Agency 26

iv. Implied Agency by Ratification is Not Established 28

C. Agency for Digital Copies and Digital Copy Deletion Provision 31

i. Proposed Tariffs 31

a. Access – Governments 32

ii. Action Taken by Access following Access – Governments 33

D. Consequences of the Absence of Educational Institutions 34

i. The Record and the Role of the Board in the Absence of Educational Institutions 35

V. ECONOMIC ANALYSIS 36

A. Introduction 36

B. The Two Proposed Tariffs 37

C. Our Proxy for 2011-2014: AUCC’s Model Licence 38

i. Similarity of the Proxy to the Target 38

ii. The Question of a Competitive Market 39

iii. Another Consideration: the Copibec Licences 40

D. Our Proxy for 2015-2017: The Premium Licence 41

i. Market “Adjustment” vs. Market “Failure” 41

ii. The Premium Licence Offering 44

a. Preliminary Considerations 44

b. The Question of Valuation Date 46

c. The Number of Signatory Institutions is Representative 46

E. The Volume-times-Value Methodology 48

F. Adjustments to the Proxies 49

i. Fair Dealing 49

ii. Insubstantial Copying 49

iii. Affiliate Adjustments 50

G. Final Rates 51

H. The College Licence 51

I. Transactional Licences 52

i. Contents of Letters Received After Hearing 52

ii. Access’ Reply 54

iii. Analysis 55

J. Who Pays the Tariff? 56

VI. ROYALTY RATES 57

VII. TARIFF WORDING 57

A. Definitions 58

i. “Copy” – definition 58

ii. “Copy” – whether activities result in a copy 58

iii. “Course collection” 59

iv. “Course of study” 59

v. “Repertoire work” 59

vi. “Secure network” 59

B. Grant of Rights 60

C. Conditions on “copying” 60

D. Non-Authorization of Circumvention of a Technological Measure 60

E. Academic or Calendar Year 61

F. Date of Payment and Report 61

G. Reporting 62

H. Audit 63

I. Subcontracting and Other Third Parties 63

J. Works in Access’ Repertoire 64

K. Payment and Interest 65

L. Other Issues 65

i. The mandatory nature of the tariff 65

ii. Royalty rates that reflect non-tariffable uses 66

iii. Licensing fees paid during the tariff period for copying nominally covered by the Tariffs 66

I. INTRODUCTION AND HISTORY OF THE PROCEEDINGS

A. The Tariffs

[1] The Canadian Copyright Licensing Agency, operating as Access Copyright, is a collective society that administers the reproduction rights in books, magazines, journals and newspapers in its repertoire for all of Canada, except Quebec. In these proceedings, the Board is asked to approve tariffs for the reproduction of those works by post-secondary educational institutions and persons acting under their authority, for the years 2011-2013 and 2014-2017.

[2] Access’ proposed rates, as of the filing of its statement of case, are $26 per full-time equivalent student (FTE) [1] for universities and $10 per FTE for other post-secondary educational institutions. For the purposes of these proposed tariffs, the reproductions in question consist primarily of:

1. paper photocopying of works in the Access repertoire for individual student study and research and for inclusion in student “course pack” materials; and
2. electronic reproduction of the works in digital form for individual study and research and for inclusion in the secure digital platforms (known as “Course Management Systems”, or “CMS”) and used by universities and colleges to deliver their courses to students.

[3] For the reasons set out below, we approve tariffs with the following royalty rates:

Table 1: Rates Fixed, per FTE per year

[4] These rates are based on two proxies: for the years 2011-2014, the model licence Access then had in place with universities and colleges; for 2015-2017, a new model licence being offered by Access to the educational institutions at a lesser rate. The rates have been adjusted to account for the copying of works whose owner of copyright is not an affiliate of Access, which we have determined Access is not entitled to license.

B. Overview and Context

[5] As described in more detail below, much of what underlies and explains these long-running proceedings has been shaped by recently evolving case law and legislation, and by advances in digital technology. In 2012, two decisions of the Supreme Court of Canada, [2] in particular, and legislative amendments to the Copyright Act [3] had the effect of expanding the scope of what is encompassed by the “fair dealing” exemption to copyright protection as it relates to the education sector.

[6] Contemporaneously with these changes, rapid developments in technology have dramatically shifted the ways in which works are reproduced. No longer dependent on a paper environment, educational institutions and their teachers are afforded ready access to literary works in digital format and the ability to deliver those courses to their students through their secure digital Course Management Systems.

[7] A significant proportion of universities and colleges responded to this evolution by altering their previous practice, under which their use of works in Access’ repertoire was governed by model licences. Instead, they developed fair dealing guidelines (“Guidelines”) which, they say, identify uses that are now covered by the fair dealing exemption and therefore no longer require a licence. Not coincidentally, perhaps, the uses identified by the Guidelines as fair dealing closely parallel the uses covered by Access’ model licences and its proposed tariffs, which Access seeks to have in place because it fears that the educational institutions will decline to enter into further licences with it for the reproduction of its repertoire works. Whether the uses outlined in the institutions’ Guideline are, or are not, exempted in the new fair dealing landscape is an issue that was litigated in the Federal Court of Canada. The judgment was rendered in July 2017, [4] and is further discussed later. This judgment is being appealed before the Federal Court of Appeal.

[8] This background has led to a further complication for the Board in dealing with the proposed tariffs. Although initially filing objections, the post-secondary school institutions all withdrew from the proceedings. In particular, the Association of Universities and Colleges of Canada (AUCC), now Universities Canada, withdrew as an objector in April 2012, during the interrogatory process. The Association of Canadian Community Colleges (ACCC), now Colleges and Institutes Canada, withdrew in October 2013, after taking part in the interrogatory process but before filing its statement of case. These two associations represent the educational institutions subject to the proposed tariffs. Their withdrawal meant that the users most directly affected by the tariffs were no longer represented at the proceedings and did not provide any evidentiary record on which the Board can rely. We will return to the added complications this posed for the Board’s deliberations later.

[9] Given the foregoing context, it is perhaps not surprising that from their inception, these proceedings have generated a near-record number of notices and rulings by the Board, some of which are explained in more detail below. The notices and rulings encompassed a wide range of contentious issues, including, for example: the fixing of an interim tariff; the winnowing of approximately one hundred initial individual and institutional objectors down to seventeen; the admissibility of certain survey evidence sought to be tendered by Access and opposed by the University of Toronto (an unsuccessful parallel application for injunctive relief by the University seeking to restrain the use of the evidence in the Board proceedings was dismissed in the Ontario Superior Court); an application by a then objector, Professor Ariel Katz, to have the Board refer a question of law to the Federal Court, which was dismissed by the Board and submitted to an unsuccessful review in the Federal Court of Appeal; requests by the Board to Access and others for further data, partly at least due to the lack of a record from education institutions; and multiple rulings relating to the interrogatory process.

[10] Access filed its proposed tariff for 2011-2013 in March 2010, and its proposed tariff for 2014-2017 in March 2013. The Board ultimately agreed to consolidate the examination of the two tariff proceedings, in July 2015. A hearing was held in January 2016. Only a single intervenor, Mr. Sean Maguire, a student, participated.

[11] Further issues arose following the hearing, however. A large number of post-secondary institutions – all of which had long before withdrawn as objectors or intervenors, and declined to participate – attempted to reinsert themselves into the process. Taking advantage of the Board’s Directive on Procedure that permits anyone to “comment” on proceedings, they wrote to the Board, reiterating previous submissions and raising new issues, on none of which they had provided any evidence for the record.

[12] With that background and context in mind, the Board has conducted its analysis of the record and the issues raised. Those include: (i) the relevance and impact of the developing “fair dealing” landscape and the Fair Dealing Guidelines adopted by the educational institutions in that context; (ii) whether Access is entitled to count their “non-affiliates” as members for purposes of tariff collection; (iii) whether Access has the capacity to authorize the making of digital copies without a requirement that these be eventually deleted; (iv) whether the former and current model licences are useful proxies on which to set a tariff; (v) whether the royalty rates need to be adjusted for either fair dealing or non-substantial copying; and (vi) whether to set any transactional licences.

[13] Before turning to the legal and economic analyses relating to those issues, however, some further explanation of the pre-hearing and post-hearing setting is in order.

C. Pre-Hearing Matters

i. Objectors

[14] The 2011-2013 proposed tariff and the 2014-2017 proposed tariff were published in the Canada Gazette on June 12, 2010, and May 18, 2013, respectively, with a notice informing potential users or their representatives of their right to object to these tariffs.

[15] About a hundred individuals and organizations objected to the 2011-2013 proposed tariff. Doubts were raised at the time as to whether some of them were “prospective users” within the meaning of the Act. On November 25, 2010, the Board ruled that 17 of these individuals or organizations could take part in the proceedings as objectors or as intervenors with full rights of participation. There was a much more limited number of objectors to the 2014-2017 tariff.

[16] The individuals and organizations objecting to one or both of the proposed tariffs included the Canadian Federation of Students (CFS), the Canadian Association of University Teachers (CAUT), Professor Ariel Katz of the University of Toronto, Sean Maguire, a student, AUCC, and ACCC. [5]

[17] Most objectors and intervenors withdrew at an early stage of the process – chiefly before interrogatories had commenced. As noted above, AUCC withdrew its objection in April 2012 (during the interrogatory process) and ACCC did so in October 2013, after taking part in the interrogatory process but before filing its statement of case. Professor Katz withdrew on December 20, 2013, as did CFS and CAUT. An application by CFS and CAUT for intervenor status with limited rights of participation in respect of the 2014-2017 tariff was later denied by the Board in June 2015.

[18] All this left Sean Maguire as the sole remaining intervenor to the proposed tariffs throughout much of the process and at the hearing.

ii. Interim Tariff and Subsequent Changes

[19] As outlined more fully below, photocopying in educational institutions has been governed since the 1990s by licences between Access and educational institutions, in the form of model licences negotiated under the auspices of AUCC and ACCC and signed by each institution. However, in 2010, Access filed its 2011-2013 proposed tariff, arguing that, in its view, it was becoming clear that the parties would not be able to agree on the terms of the new licences that would apply after the then-applicable licences would expire on December 31, 2010.

[20] In keeping with that position, Access applied to the Board on October 13, 2010, for an interim tariff in accordance with section 66.51 of the Act. Because of the urgency of the situation, and to avoid a potential legal vacuum, the Board granted the application and issued an interim tariff on December 23, 2010, “even though its provisions would benefit from fuller discussion.” [6]

[21] The interim tariff tracked the wording of AUCC’s Model Licence to the extent possible. It was to apply from January 1, 2011, until the earlier of December 31, 2013, or the date a final tariff was certified. Like the Model Licence, the tariff provided for royalties of $3.38 per FTE and $0.10 per page copied as part of a course pack. In the case of proprietary colleges, [7] these rates are $3.58 per student and $0.11 per page copied as part of a course pack.

[22] While the interim tariff was modified several times over the subsequent months, none of these modifications affected the rates payable.

[23] Finally, on September 23, 2011, in its Access Interim Tariff – Transactional Licence decision, [8] the Board denied AUCC’s application to have the Board vary the tariff to require Access to grant transactional licences on request. AUCC and the University of Manitoba then applied for judicial review of that decision. On March 20, 2012, the Federal Court of Appeal dismissed the application on the basis that there were no special circumstances that would justify the intervention of the Court at that preliminary stage of the proceedings. [9]

[24] It should be noted that AUCC and the University of Manitoba also applied for judicial review of another interlocutory decision of the Board regarding the interrogatory process. On August 18, 2011, the Board had ordered ACCC and AUCC to obtain interrogatory responses from a representative sample of institutions that had expressed their desire to opt out of the interim tariff. The Court dismissed this application as well, for the same reasons. [10]

iii. Other Relevant Procedural History

[25] We have already remarked on the number of Board interventions and rulings that the proceedings have engendered. The following will round out the picture in general terms.

[26] After AUCC withdrew its objection to the 2011-2013 proposed tariff, leaving only one institutional objector (ACCC) and one intervenor (Mr. Maguire), the Board fixed a schedule and set a date of February 12, 2014, for the hearing. On March 28, 2013, however, Access submitted its 2014-2017 proposed tariff.

[27] Access submitted its statement of case and supporting evidence with respect to it on September 13, 2013. But on November 18, 2013, after ACCC had withdrawn its objections to both tariffs, Access asked the Board to consolidate the proceedings relating to both tariffs. In its ruling of December 4, 2014, the Board initially declined to do so.

[28] The Board had received, and on December 9, 2013, dismissed an application by Professor Katz in which he asked the Board to refer a question of law to the Federal Court of Appeal, in accordance with subsections 18.3(1) and 28(2) of the Federal Courts Act. Professor Katz’ proposed question related to whether a single copying event by an educational institution would be sufficient to trigger the tariff, or “trigger” liability under it. The Board dismissed his application, holding, among other things, that the issue raised was not a proper matter for a reference because it was not a question that the Board was required to answer in order to exercise its jurisdiction and certify the proposed tariffs.

[29] Next, an issue arose that bore on the evidentiary record that would be before the Board in the proceedings. On December 13, 2013, Access applied for leave to file additional evidence in the form of an analysis by Access of data provided by the University of Toronto regarding reproductions made on the University’s secure digital platforms on which it provides its Course Management Systems. The data on which this analysis is based had been submitted by the University of Toronto in accordance with a licence entered into by the parties on January 30, 2012, and that applied retroactively from January 1, 2011, to December 31, 2013. Access submitted that the evidence was relevant because it was based on the only data available to it regarding the reproduction of works on a university’s digital platform.

[30] The University of Toronto objected to this application. It submitted that the tendered evidence was inadmissible because the University was not a party to the proposed tariff. In a parallel process, on January 9, 2014, the University applied to the Ontario Superior Court for an injunction to prevent Access from filing this evidence.

[31] In response to these developments, the Board set up a process to enable the intervenors [11] and the University to make their case against Access’ application to file additional evidence. On January 13, 2014, however, the University asked the Board to suspend the process regarding the filing of additional evidence until the Court could dispose of the injunction issue. That same day, the Board granted the request.

[32] Meanwhile, on December 19, 2013, Mr. Maguire filed his statement of case. The next day, all other intervenors withdrew. On January 17, 2014, the Board itself put a series of initial questions – for the most part, economic ones – to Access. It also adjourned the pending February 12 hearing sine die.

[33] The Board took the latter step for several reasons. First, the nature of the answers to the questions asked by the Board were such that Access could not reasonably answer them before the beginning of the hearing scheduled for February 12. Second, if the Board decided to grant leave to file Access’ additional evidence regarding the University of Toronto’s Case Management Systems, it could have questions to put to Access regarding that evidence, and more time would then be needed to answer them. Finally, the withdrawal of Professor Katz, CAUT and CFS, and of ACCC just a few weeks earlier, in addition to the withdrawal of AUCC, meant that the Board had to play a more active role and needed time to make the necessary adjustments.

[34] On February 4, 2014, the Ontario Superior Court denied the University of Toronto’s application for an interlocutory injunction. On February 6, the Board reinstated the time limits suspended on January 17, 2014, in order to receive the arguments of the parties and the University of Toronto regarding the admissibility of the evidence. On March 6, the Board granted Access’ application to file the contested evidence regarding digital copies made at the University of Toronto, finding among other things that it must ensure that it benefits from the most comprehensive evidence, in order to certify a fair and equitable tariff. The Board also stated that this issue was one of admissibility of evidence and that, as such, the University’s arguments regarding the fact that Access should have excluded copies licensed under agreements with third parties concerned the weight to be given to the evidence; for this reason, the arguments were premature.

[35] In the meantime, on January 27, 2014, Access filed its reply statement of case. On June 3, 2015, the Board asked Access to answer a number of other economic questions, some of which dealt with the new licence offers made by Access to post-secondary educational institutions from December 2014 onwards at rates lower than those of $26 and $10 per FTE under the model agreements entered into with AUCC and ACCC.

[36] On June 16, 2015, Access asked the Board to set a new hearing date. It also requested that the order of June 3 compelling it to answer questions be stayed. It alleged that several of these questions concerned years after 2013 and thus after 2011-2013, the only period under examination, given the Board’s refusal to consolidate this tariff with the one for 2014-2017. It therefore asked the Board to suspend the time limit for answering these questions. In the alternative, it asked the Board once again to consolidate the examination of the tariffs for 2011-2013 and 2014-2017, in which case it would answer the questions of June 3.

[37] On July 3, 2015, the Board consolidated the examinations of the two tariffs. It was of the opinion that the reasons for refusing to consolidate the two files in 2013 were no longer valid. First, the initial hearing had not taken place. Second, more than a year and a half later, the parties were in a better position to provide useful information regarding the 2014-2017 period. Finally, both parties agreed that consolidation would result in lower costs than separate examinations for the two periods.

[38] On July 15, 2015, the Board set a schedule of proceedings that was to culminate in a hearing on January 18, 2016.

[39] On November 3 and December 4, 2015, respectively, Access and Mr. Maguire filed their statements of case for 2011-2013 and 2014-2017. Finally, on December 17, Access filed its statement in reply. The hearing took place over four days.

[40] Finally, it should be noted that throughout the whole process, the Board asked Access questions on a variety of issues. Some questions were put to Access before the hearing, on January 17, 2014 (Access’ responses filed in evidence as exhibits AC-18, AC-19 and AC-20), February 18, 2014 (responses filed as Exhibit AC-21), June 3, 2015 (responses filed as Exhibit AC-32), December 15, 2015 (responses filed as Exhibit AC-34) and January 14, 2016 (responses filed as Exhibit AC-48), as well as after the hearing, on June 9, 2016 (Access’ responses filed as exhibits AC-50, AC-51 and AC-52).

D. Post-Hearing Developments

[41] On February 9, 2016, after the hearing had ended, York University (York) wrote to the Board to inform it that it had entered into numerous licences with third parties other than Access for the reproduction of copyrighted works, either individually or through a group of various university libraries. To avoid having to pay twice, it argued that the tariff to be certified by the Board should take these licences into account. Furthermore, in addition to the blanket licence component with a fixed per-student rate, the tariff should also include a transactional component that would allow educational institutions to pay on a per-use basis. According to York, this would take into account the fact that universities’ copying needs were in large part already covered through licences with third parties.

[42] York also noted that as of August 31, 2011, it had decided that, unlike other institutions, it would not operate on the basis of the Access Interim Tariff (2011-2013). It argued that it was not subject to the interim tariff because it had not done any copying that would have required a licence from Access. This led Access to commence proceedings against York in Federal Court claiming, among other relief sought, a declaration to the effect that York was required to pay Access the royalties set out in the interim tariff because it had reproduced copyrighted material from its repertoire that had not otherwise been cleared.

[43] Subsequently, on February 15, 2016, a consortium consisting of the following universities and Boards of Governors also sent a comment letter to the Board: University of British Columbia, University of Winnipeg, University of Lethbridge, University of Calgary, Thompson Rivers University and University of the Fraser Valley, as well as the Governing Council of the University of Toronto and the Governors of the University of Alberta.

[44] The Consortium’s letter essentially repeats the same arguments as York’s, namely that the Board must offer a range of licensing options, including a transactional licence, to account for the fact that the institutions must enter into numerous licences with third parties for materials not in Access’ repertoire, or to avoid any double payment, as is the case where licences with third parties and the Access licence permit reproduction of the same works, with Access granting licences on a non-exclusive basis.

[45] Finally, on February 17, 18, and 29 and March 25, 2016, Brock University, the University of Manitoba, Athabasca University and Mount Royal University each in turn wrote to the Board to express their support for the comments made in the letters from the Consortium and York.

[46] These post-hearing attempts by the post-secondary institutions to reinsert themselves into the proceedings utilizing the Board’s directive permitting anyone to submit “comments” on a proceeding, together with their failure to present any evidentiary record throughout the process, have had consequences for the way in which the Board has had to address the proposed tariffs and the record before it. We will return to this issue later in these reasons.

[47] The record of the proceedings was perfected for determination on October 26, 2016, after the Board had received Access’ responses to its questions posed on July 9, 2016.

II. EVIDENTIARY FRAMEWORK

A. The Licensing History

[48] Beginning in the 1990s, copyright royalty payments for the reproduction of works in the Access repertoire have been based on licensing agreements entered into between Access and the post-secondary educational institutions, or associations representing them. Before turning to a review of the evidence as a whole, we will provide an overview of those licensing arrangements in this section of the reasons, together with a brief discussion of what has happened in that regard over the past few years. Some of the agreements in question constitute the basis of Access’ tariff proposal. In addition, we will examine agreements before and after 2015 because it was at that time that Access, for the first time, granted licences at a price below what it considered to be fair.

i. Before 2015

[49] In 1994, Access and AUCC concluded their first model licence for universities. A model licence was then concluded with ACCC for colleges. The model licences were later renewed several times. Under these licences, educational institutions paid royalties to Access until December 31, 2010.

[50] The licences initially covered only paper reproductions. They generally provided for a flat fee per FTE coupled with a per-page fee for the reproductions in course packs. For royalty distribution purposes, the educational institutions provided copy reports to Access, listing the copies made for their course packs. Under the model licences that expired in December 2010, the institutions paid $3.38 per FTE and $0.10 per page included in a course pack. [12]

[51] In 2010, following unsuccessful negotiations, some institutions informed Access that they would not be renewing their licences. That same year, Access submitted its 2011-2013 proposed tariff.

[52] In January 2012, Access concluded licences with the University of Toronto and the University of Western Ontario at the rate of $27.50 per FTE, applicable from January 1, 2011, to December 31, 2013. These licences covered digital copies as well as paper copies. The parties could not agree on conditions of renewal and, after their expiry, the agreements were not renewed.

[53] During the spring of 2012, Access concluded a model licence with AUCC, and then with ACCC, which these associations encouraged their respective member universities or colleges to sign. A certain number of institutions did so, and we will address that development later. The rates provided in these model licences were $26 per FTE for universities and $10 for colleges. The agreements applied retroactively to January 1, 2011, and expired on December 31, 2015. They covered both paper and digital copies and provided that copies to be included in course packs had to be reported. Following the conclusion of the model agreement between Access and AUCC, the rates provided for in the University of Toronto and the University of Western Ontario licences were reduced to $26, pursuant to a clause in those agreements.

[54] Having agreed to a model licence, AUCC withdrew from the proceedings on April 24, 2012. AUCC correctly noted that, in accordance with section 70.191 (as it then was) of the Act, the Access Interim Tariff (2011-2013) and the final tariff that would potentially be certified would not apply to the educational institutions that had concluded an agreement with Access.

[55] In the fall of 2012, however, something else occurred that is an important step in the narrative. Relying on the case law of the Supreme Court referred to above, AUCC and ACCC adopted a fair dealing policy in the form of Guidelines. [13] Under these policies, copies of “short excerpts” which are made for one of the purposes listed in section 29 of the Act, and which involve reproducing an article in a newspaper or periodical, a chapter of a book or 10 per cent or less of a book, are considered to be fair and thus not compensable. Those limits are virtually identical to those prescribed in Access’ model licences or those provided for in Access’ proposed tariffs. [14]

ii. After 2015

[56] In 2014, most of the educational institutions that had concluded a licence with Access informed Access of their intention not to renew their licence when it expired on December 31, 2015. To meet this challenge, Access began offering educational institutions new licences towards the end of 2014, at rates lower than $26 for universities and $10 for colleges provided in the model licences in effect at the time.

[57] Access offered two types of licences: a Premium Licence, [15] which is a blanket licence with a flat rate per FTE; and a Choice Licence, [16] which combines a flat rate per FTE with a rate per page for course packs and digital copies. The latter was offered only to institutions with a centralized copyright management structure. Those licences were to apply as of July 1, 2015. The licences that some institutions had concluded under the auspices of AUCC and ACCC were expiring in December 2015, and institutions did not agree to renew them at the same rates. In this context, Access offered to replace them before they expired with new ones at reduced rates.

[58] The Premium and Choice Licences offer sliding-scale rates. For the Premium Licence, the rates per FTE for universities are $18, $15 or $12 for a one-year, three-year or five-year licence respectively. For college-level institutions, the rates are $8 and $7 for a one- or three-year licence. With respect to the Choice Licence, the rates per FTE for universities and colleges are $6 and $5 respectively for a one- or three-year licence, plus $0.12 per page for digital copies or those included in course packs. In addition, the licences provide for copy limits that are higher than those in the old licences.

[59] We turn now to a more thorough review of the evidence as presented by the various witnesses.

B. Evidence

i. Access

[60] Access called six witnesses who testified, generally, as to the licensing practices and repertoire of Access, and its proposed tariffs and the impact on the Canadian publishing industry of the current resort to digital technology in the education sector. It also called four expert witnesses who gave evidence outlining the use of digital platforms in the education sector and providing their opinions respecting trends in the quantity and methods of copying by institutions, the effect of digital technology on creators’ incomes and production, and value. A summary of this evidence follows.

a. Fact Witnesses

Roanie Levy and Jennifer Lamantia

[61] Access’ Chief Executive Officer Roanie Levy and Manager of Education Licensing Development Jennifer Lamantia described the corporate structure of Access, the composition of its Board of Directors, its mandate, the type of works it represents, the types of licences it grants and the principles governing royalty collection and distribution.

[62] They also outlined the history of licences held in the post-secondary education field from 1994 until the filing of the proposed tariffs. They then described the licences entered into by the University of Toronto and the University of Western Ontario as well as the model licences entered into in 2012 by AUCC and ACCC. They explained that there were three categories of users: those who paid royalties under a licence with Access, those who paid royalties under the proposed tariff from 2011 to 2013, [17] and those who paid no royalties.

[63] They then explained that in 2012, a few months after the model licences were entered into, several institutions adopted a fair dealing policy. Ms. Levy and Ms. Lamantia attribute the non-renewal of model licences after they expired in December 2015 to those policies as well as to other factors, such as the fact that universities enter into licences for the reproduction of periodicals directly with rights holders, that more and more institutions are moving toward open-access content, that the legislative changes and Supreme Court decisions in 2012 have created uncertainty regarding the boundaries of fair dealing, and that the digital environment makes it more difficult for rights holders to detect cases of copyright infringement.

[64] According to them, Access resigned itself to offering licences at a discounted rate to prevent the cessation of royalty payments. Despite these rates, the proportion of institutions having signed those “discounted” licences was lower than the proportion of those that had signed the model licences. They stressed that Access expected its revenues to drop considerably from 2015 to 2016. [18]

[65] In addition, they explained that Access had conducted a repertoire analysis based on the data provided by the institutions in response to interrogatories aimed at a sample of ACCC member colleges. In that regard, they gave examples of cases where course packs that had previously been subject to a payment to Access, either under a licence or an interim tariff, had been reproduced again the following year, but without paying royalties to Access. They also gave examples of reproductions where institutions had exceeded the limits that they themselves had adopted in their fair dealing policies.

[66] With respect to Access’ repertoire, Ms. Levy and Ms. Lamantia explained that Access collected royalties for works of rights holders who had signed an affiliate agreement with Access, as well as for works of members of foreign collectives with which it had concluded reciprocity agreements.

[67] They explained that Access also collected royalties for works of rights holders who had not signed an affiliate agreement, but who had authorized Access to act on their behalf through an implied agency. This would happen if Access, noting that such works have been reproduced, pays royalties to the non-affiliates, who accept the payments, which is almost always the case.

[68] Ms. Levy and Ms. Lamantia explained that Access’ repertoire was therefore made up of works of both affiliates and non-affiliates, except for those who had expressly requested that Access not represent them. In consideration of the fact that Access claims to represent non-affiliates, it includes in its licences an indemnity clause through which it undertakes to defend users who could be sued for copyright infringement by a non-affiliate.

[69] Regarding the tariff provision stipulating that copies must be destroyed once an educational institution is no longer covered by the tariff, they stressed that this was a condition set out in its previous affiliate agreements. Access had, however, obtained from its members a retroactive written waiver of that clause so that this condition could be removed from the tariff. Then, starting in 2015, Access had asked its members to sign new affiliate agreements which no longer contained that condition.

Michael Andrews

[70] Michael Andrews [19] is the Vice-President of Finance at Nelson Education Inc. Mr. Andrews explained that Nelson is the largest educational publisher in Canada. Nelson publishes textbooks, including at the post-secondary level, in paper and digital formats. Nelson also provides à la carte digital publishing services, whereby it is possible to purchase only the desired chapters and whereby a teacher can create a unique textbook made up of content created by Nelson and content developed by the teacher him- or herself. Nelson also distributes in Canada textbooks published in the United States.

[71] Mr. Andrews described Nelson’s revenue from the primary market, textbook sales, as well as from the secondary market, royalties received from Access for the licences granted to users and those resulting from the authorizations it grants directly to users. He testified that textbook sales have been dropping for the last few years and attributes this to various factors including the second-hand book market, textbook rentals, illegal distribution on the Internet and unauthorized reproduction by educational institutions. With respect to secondary revenue, he also explained that both the royalties from Access and those relative to the authorizations requested directly from Nelson have dropped dramatically since 2011, that is, since the non-renewal of Access’ licences by educational institutions and the adoption of fair dealing policies. The policies enable teachers to reproduce textbooks within limits such that the reproductions enter into competition with Nelson’s primary market.

[72] Regarding the fact that revenue related to reproduction requests made directly to Nelson has decreased, Mr. Andrews stated that this is counter-intuitive. Logically, as institutions decided to no longer be subject to an Access licence, reproduction requests made directly to Nelson should have increased. [20] Finally, he described how a repertoire study based on ACCC’s answers to interrogatories by Access [21] made it possible to demonstrate that several of the titles published by Nelson were reproduced by educational institutions when they were not subject to any licence and paid no royalties.

David Swail

[73] David Swail was the President and Chief Executive Officer of McGraw-Hill Ryerson Ltd. (MHR) until 2014. Since 2015, he has been the Executive Director of the Canadian Publishers’ Council. He described the changes to MHR’s corporate structure that took place in 2014 and the fact that the parent company, McGraw-Hill Education, based in the United States, reduced the scope of MHR’s Canadian activities. MHR mostly publishes textbooks for the educational sector, at the primary and secondary levels, as well as at the post-secondary level, in both paper and digital formats. Although most of MHR’s revenue comes from print publications, revenue from digital publications is climbing steadily.

[74] Mr. Swail explained that developing digital resources is expensive and described the variety of digital products on offer, including the possibility of creating personalized textbooks. He also explained the significance of the royalties paid by Access and described how these amounts have decreased from 2011 to 2014, for various reasons, but mainly because of the fair dealing policies and of the uncertainty surrounding the status of copyright in Canada. According to him, since Canada is a relatively small market, in time, the drop-in revenue associated with the educational institutions’ fair dealing policies may result in reduced investment in the creation of Canadian content by publishers such as MHR, thus eroding the diversity of textbooks available.

Don LePan

[75] Don LePan is the Founder, President and Chief Executive Officer of Broadview Press, a Canadian publisher that mostly publishes poetry, anthologies and fictional works mainly targeting literature students at the post-secondary level. Mr. LePan described the significance of royalties received from Access, which for a small independent publisher can make the difference between making a profit and reporting losses.

[76] Mr. LePan also compared the revenues from the sales of nine titles published by Broadview Press from 2011 to 2015. During that period, revenues dropped steadily and considerably. He attributed the drop to the fair dealing policies adopted in 2012 by several educational institutions. By allowing poems to be reproduced in their entirety and included in a course pack, these policies make it possible to create poetry anthologies without paying royalties, thus competing with Broadview Press’s niche market. In addition, Mr. LePan gave examples of specific cases where reproducing 10 per cent of a work, as permitted by these policies, is sufficient to reproduce dozens of poems or an entire introduction to an anthology, which is undoubtedly the most important part of an anthology.

John Degen

[77] John Degen is the Executive Director of The Writers’ Union of Canada (TWUC). Mr. Degen described the findings of a study carried out in 2015 on the revenues of its approximately 2,000 members. According to that study, the average annual revenue of its members was $12,879. In addition, half of the respondents indicated that their revenues had decreased in the last few years by about 25 per cent. The TWUC also conducted a survey in 2013 regarding its members’ perceptions of the fair dealing policies adopted by several educational institutions. According to Mr. Degen, the study showed that almost all of its members, although in favour of fair dealing, consider the policies to be an erroneous and unfair interpretation of this concept and anticipate that it will cause a drop in their revenues, which are already low.

b. Expert Witnesses

Benoît Gauthier

[78] Benoît Gauthier, President of Circum Network Inc., presented the findings of several analyses he had conducted at Access’ request.

[79] He first described the results of an analysis of answers to questions provided by a random sample of 16 ACCC member institutions not party to a licence with Access, as ordered by the Board in connection with the interrogatory process. [22] The institutions had to provide, among other things, a list of all the works reproduced for a selection of courses in the 2011-2012 school year. Mr. Gauthier explained that, based on these data, he estimated that 986 compensable copies had been made per FTE per year. He also explained that the institutions had made five times more digital copies than copies for inclusion in a paper course pack.

[80] In addition, Mr. Gauthier presented the results of an analysis of reproductions made by post-secondary institutions in relation to paper course packs for the period from 2005 to 2014. [23]

[81] According to his findings, the volume of reproductions of works in Access’ repertoire made for inclusion in course packs is 168.7 pages per FTE per year for AUCC member institutions, 36.2 pages for ACCC members and 43.7 pages for proprietary colleges. Mr. Gauthier also explained that he used only the data for 2005 to 2010 to calculate the number of copies. According to him, the data for 2011 and later, which show a drop in the number of copies, underestimate the number of copies and should therefore not be taken into account.

[82] Initially, from 2005 to 2010, all the educational institutions had a licence with Access. The licences provided that the institutions pay royalties by page for copies inserted into course packs. Consequently, those copies were reported to Access. However, when the licences in place up to that point expired at the end of 2010, several institutions decided not to renew their licence with Access and not to operate under the interim tariff, which had the same structure as the licences in place until then, that is, a flat rate per FTE plus a per-page rate for copies in course packs. Accordingly, they stopped reporting the copies they made. This does not mean, however, that they stopped making copies. Moreover, Mr. Gauthier said that, starting in 2011, digital copies progressively replaced course pack copies.

[83] Mr. Gauthier also analyzed the number of digital and paper copies made at the University of Toronto. [24] With respect to digital copies, it should be noted that the data this analysis is based on were given to Access by the University of Toronto as stipulated in the licence in effect between the parties from 2011 to 2013. The University of Toronto had then unsuccessfully disputed the filing in evidence of these data as mentioned above.

[84] Based on his analysis, Mr. Gauthier concluded that, assuming that there are as many digital copies as there are students enrolled in courses, the number of reproductions on the university’s digital platform was 344 copies per FTE per year, and that there were six times more digital copies than paper copies used in course packs.

[85] Finally, Mr. Gauthier described the results of a survey he conducted with the authors who were members of Access. The goal of that study was to determine the potential impact of a decrease or elimination of revenues paid by Access in relation to the adoption of fair dealing policies by educational institutions.

[86] Mr. Gauthier first described the allocation of revenues collected by authors in relation to their literary creation activities. In that respect, 39 per cent of their revenues came from payments made by their publisher and 21 per cent from royalties paid by Access. Regarding the significance of the revenues paid by Access, 64 per cent stated that these revenues are at least moderately significant for them, and 34 per cent said that they are not very or not at all significant. Finally, regarding the authors’ wish to continue creating works for the post-secondary market, half of them stated that a drop in or cessation of revenues would have no effect on them, while about a third would reduce the amount of time they spend on creating for this market.

Professor Michael Murphy

[87] Professor Michael Murphy explained what the digital platforms used in educational institutions are. He also described how they work, and in particular, how the institution uploads content onto them and how students access them, namely, by downloading the content onto a device, printing it or consulting it. Finally, he explained the number of copies that these uses involve, as well as the types of copies made (permanent, temporary, and ephemeral).

Michael Dobner

[88] Mr. Michael Dobner, Partner at PricewaterhouseCoopers LLP, presented the findings of his analysis of impacts that the implementation of fair dealing policies adopted by the majority of educational institutions in 2012 would have on the market.

[89] He explained that adopting these policies risks an almost total elimination of secondary revenues, that is, the royalties paid to the rights holders by Access and the revenues from the licences granted directly by publishers to the educational institutions. These royalties represent a significant portion of the authors’ and publishers’ revenues. He also testified that adopting these policies would cause a drop in textbook sales; as these policies make it possible to reproduce works without authorization, with similar limits to those that existed under Access’ licences, royalty-free copying will supplant textbook sales.

[90] In Mr. Dobner’s opinion, the drop in authors’ revenues will reduce their interest in creating works and will in the end result in a drop in the number of works created by them. As for publishers, these decreases in revenue will result in a decrease in investment, especially with respect to the costly transition from paper to digital. This will particularly affect small publishers, which will likely have to close down, lay off employees or direct their efforts somewhere other than the post-secondary education market. Finally, a drop in the publication of specifically Canadian titles will follow, as well as a decrease in the quality and diversity of titles published.

Bradley Heys

[91] Finally, Access asked Mr. Bradley Heys, Vice-President of NERA Economic Consulting Inc., to provide an estimate of the value of the authorizations granted by each of the two proposed tariffs.

[92] Mr. Heys explained that the best way to assess the value of these licences was to use benchmark licences, that is, the model licences concluded in 2012 with AUCC, ACCC and proprietary colleges, as well as licences concluded in 2012 with the University of Toronto and University of Western Ontario. According to him, those licences represent an appropriate benchmark since they are based on rates that were freely negotiated by the parties for granting authorizations similar to those granted by the proposed tariffs. For these reasons, the rates provided in the benchmark licences, namely, $26 for universities and $10 for other institutions, represent the fair market value of the copies made during the 2011-2013 period.

[93] Mr. Heys submits that those licences also constitute the appropriate benchmark with respect to the 2014-2017 period and that the rates of $26 for universities and $10 for other institutions represent the fair market value of the copies made during that period. He states that the lower fees provided in the Premium and Choice licences do not represent the fair market value of the copies made during the 2014-2017 period.

[94] Indeed, those rates were offered by Access in a context where many educational institutions decided that they had no need for a licence; institutions began interpreting the concept of fair dealing very liberally, and a majority had adopted fair dealing policies providing for very generous limits for copies not subject to payments. In addition, institutions were apparently informed that the Board’s tariffs were voluntary and that damages for copyright violations were, in any event, rather low. [25] Added to this is the withdrawal of AUCC and ACCC, and the fact that it is becoming more and more difficult for rights holders to enforce their rights in a context of digital copies where copies are now found on servers more often than in course packs. Finally, there is nothing to suggest that there has been a drop in the number of copies made in educational institutions – quite the contrary, in fact – or that more licences are concluded with third parties other than Access.

c. Conclusions Respecting Access’ Evidence

[95] Because of the absence of objecting parties, none of the foregoing fact evidence was contested or contradicted. We accept the facts as set out in the testimony of the fact witnesses and the exhibits presented through them. With one caveat, we generally accept the opinion evidence of the expert witnesses.

[96] The caveat is with respect to the testimony of Mr. Heys. While we accept his evidence that the model licences provide suitable proxies for assessing value, we do not accept his opinion that the Premium Licences and the Choice Licences are not a reflection of the market. For the reasons outlined below, we conclude that the Premium Licence is an appropriate proxy for the 2015-2017 period because it considers the reality of the current uncertainty in the market with respect to the boundaries of the fair dealing exception in the education sector.

ii. Sean Maguire

[97] Mr. Maguire did not present any evidence or call any witnesses.

III. PARTIES’ POSITIONS AND PROPOSED TARIFFS

A. Access

[98] For 2011-2013, Access proposed a rate of $45 per FTE for universities and $35 per FTE for other educational institutions (“other institutions” or “college-level institutions”). For 2014-2017, it proposed a rate of $35 per FTE for universities and $25 per FTE for other educational institutions.

[99] In response to a question from the Board on September 30, 2016, Access explained that the requested rate in its proposed tariff for 2014-2017 was lower than the one requested for 2011-2013 because, by the time the second proposed tariff was filed, Access had entered into model agreements with AUCC and AUCC at lower rates. It therefore wanted to move closer to the negotiated rates in those agreements ($26 and $10).

[100] The two proposed tariffs are similar. Both cover digital as well as paper copies. The restrictions are also identical: both proposed tariffs permit the copying of up to 10 per cent of a work, or 20 per cent in the case of chapters. They both cover reproductions made in connection with posting “a link or hyperlink” to a digital copy.

[101] The proposed tariffs also contain a provision to the effect that copies cannot be “stored or indexed with the intention or result of creating a library of published works.” [26] Finally, they provide that when an educational institution is no longer covered by this tariff, the institution and authorized persons must cease to use the copies made under the tariff and delete them from the hard drives, servers and networks used in the institutions. [27]

[102] There are some differences between the two proposed tariffs. For example, the 2014-2017 proposed tariff adds a reference to the making available right, unlike the 2011-2013 proposed tariff. Access explains that although the making available right and the reproduction right are actually two distinct rights, the fact that it represents these two rights and bundles them in the licences it grants means that the Board does not have to attribute a distinct price to each of them.

[103] In its statement of case, Access reduced its rates to $26 per FTE for universities and $10 per FTE for college-level institutions, for both proposed tariffs. These rates match those of the model licences entered into with AUCC, ACCC and the proprietary colleges in 2012, and which were in force retroactively from January 1, 2011, to either December 31, 2015, in the case of the universities, or to December 31, 2013, in the case of colleges.

[104] The 2.6 to 1 ratio between the rates of $26 and $10 stems from the fact that before 2011, when all the institutions held a licence from Access consisting of a flat rate per FTE and a per-page rate for course packs, the royalties paid by the universities were about 2.6 times higher than those paid by the other institutions. The negotiated rates in the 2012 model licences reproduce this ratio. [28]

[105] Access submits that these licences, the terms of which are comparable to those of the 2011-2013 proposed tariff, are an appropriate benchmark for this tariff, and that the rates set out in it reflect the fair market value [29] of the copies made during the 2011-2013 period.

[106] Access further submits that these licences are also the appropriate benchmark for 2014-2017, and that the rates set out in them also represent the fair market value of the copies made during that period, that is, $26 for universities and $10 for colleges.

[107] Indeed, even though Access began offering Premium and Choice licences at the end of 2014, at rates below the $26 and $10 rates in the benchmark licences (for example, under a three-year Premium Licence, the rates offered are $15 for universities and $7 for colleges), Access submits that these licences should not be taken into consideration, as they do not reflect the fair market value of the authorizations granted by Access to institutions for 2014-2017.

[108] According to Access, it became clear in 2014 that most of the educational institutions that had entered into licences at the $26 or $10 per FTE rate were not going to renew their licences when they expired on December 31, 2015. This was apparently due to the educational institutions’ interpretation of the addition of the word “education” to section 29 of the Act and of the Supreme Court’s decision in Alberta, [30] which, the educational institutions say, suggested that most of the copies made in these institutions would fall under fair dealing. It may also have been due to the educational institutions’ adoption of their Fair Dealing Guidelines in 2012, according to which copying within limits similar to those provided in Access’ licences are claimed to be fair dealing, and to the difficulties faced by rights holders, especially in a digital environment, in trying to detect cases of copyright infringement and enforce their rights.

[109] Access submits that in this context, where a growing number of institutions have taken the position that an Access licence is no longer necessary, it decided to offer licences at discounted rates. The purpose of these licences was to encourage institutions to enter into licences with Access, albeit at lower rates, thereby averting the loss of royalties that would have resulted from the licences not being renewed. These licences are the consequence of a market failure, Access argues, and do not reflect the fair market value of the authorizations granted by Access to educational institutions for 2014-2017.

[110] Finally, Access submits that the volume of compensable copies is such that if the rates were to be set by taking into account the volume of copying multiplied by a per-page rate, as opposed to the rates set out in the benchmark licences, the royalties would be much higher. According to Access, this shows that the proposed rates of $26 and $10 are reasonable.

B. Sean Maguire

[111] Mr. Maguire states that in his experience, when licences consisted of a flat rate per FTE along with a per-page rate for course packs, the flat-fee portion of the royalties was borne by the educational institutions, while the portion related to the use of course packs was borne directly by students. Mr. Maguire believes that the new tariff structure proposed by Access, namely, an overall rate covering all copies, will transfer the costs of royalties directly to students, most likely in the form of fees. This would result in an increased financial burden on students.

[112] Mr. Maguire submits that the rates proposed by Access are too high. Although he did not specify what the royalty rates should be, he argues that the addition of the word “education” to section 29 of the Act concerning fair dealing, as well as the decision in Alberta likely make a significant proportion of the copies made in educational institutions fair. These changes should therefore result in a reduction in the rates. Moreover, the fact that, following these changes, the institutions chose not to renew the licences they had entered into at rates of $26 and $10 shows that the value of Access’ licences has fallen considerably. Not only that, but the fact that the proportion of institutions having signed Premium and Choice licences with Access is lower than the proportion of institutions that had previously entered into licences at the higher rates of $26 and $10 suggests that the value of Access’ licences is in fact lower than the reduced rates provided for in the Premium and Choice licences.

[113] Finally, regarding the wording of the tariff, Mr. Maguire has expressed concerns about subsection 4(3), which provides that copies “shall not be stored or indexed with the intention or result of creating a library of published works.”

[114] He asserts that any significant research done at the post-secondary level relies on multiple bibliographical sources, and that any diligent student would organize and index them for ease of reference. He fears that this prohibition applies to this sort of activity. If such is the case, he argues that a tariff cannot prohibit an activity that is such an engrained part of post-secondary education. He therefore asks that this subsection be amended or deleted.

IV. LEGAL ANALYSIS

[115] In this section we consider a number of legal issues that must be addressed in determining these proceedings and that will be applied in conjunction with our economic analysis which follows. Most have been alluded to earlier in these reasons. We turn to them in more detail now. They include:

1. The legal landscape respecting the notion of fair dealing as it applies to the education sector;

2. Access’ claim to represent non-affiliated rights holders through implied agency;

3. Agency in relation to digital copies and the digital copy deletion provisions; and

4. The implications flowing from the absence of educational institutions in the proceedings.

A. The Fair Dealing Landscape and Its Effect

[116] Access argued strenuously that the educational institutions have misinterpreted the effect of the Supreme Court of Canada’s decisions in Alberta and Bell, and of the amendment to the Act specifically adding “education” as an allowable purpose of “fair dealing” under s. 29. They submit that the educational institutions have gone too far in their implementation of their Fair Dealing Guidelines.

[117] In the end, we conclude that it is unnecessary for us to determine in these proceedings whether, or to what extent, the Fair Dealing Guidelines or the proposed tariffs encompass uses that are exempted from copyright protection under the fair dealing provisions of s. 29 of the Act. As explained in the economic analysis section of these reasons, the AUCC Model Licence and the Premium Licence provide useful proxies that permit us to set a price for the proposed tariffs. Those licences already incorporate a market-generated allowance for the current instability surrounding the fair dealing landscape and it is therefore unnecessary to conduct a fair dealing analysis in that regard.

[118] However, because the fair dealing debate underlies much of what these proceedings have been about, and has contributed to their length, a brief outline of how that current instability has evolved may be helpful.

[119] As early as 2004, in the seminal case of CCH, [31] the Supreme Court of Canada began to take a more expansive view of the concept of fair dealing. It held in that case that fair dealing was to be given a large and expansive interpretation, was not confined to non-commercial or private situations, and was to be interpreted taking into consideration the perspective of the users. It set out a framework for carrying out the appropriate analysis.

[120] Eight years later, in 2012, the Court revisited the fair dealing concept in Alberta [32] and Bell [33] – two decisions involving judicial review of tariffs approved by the Board. Again, it took an expansive view of “research” and “private study” – the allowable fair dealing purposes that were then permitted by s. 29 of the Act – and, again, the Court did so by approaching the analysis from the perspective of the user.

[121] At approximately the same time, the Copyright Modernization Act came into effect, specifically adding “education” as an allowable purpose under s. 29.

[122] The Alberta and Bell decisions, together with the amendment to s. 29, signalled an increasingly broad approach to the concept of fair dealing and, as described above, have had significant effects on the use of the Access repertoire in the education sector. These included the adoption by most post-secondary educational institutions of Fair Dealing Guidelines, the refusal of many of the institutions to renew or enter into further licences with Access, and Access’ attempt to address what it refers to as the resulting “market failure” by offering its Premium and Choice licences in an effort to prevent the royalties paid by education institutions from dwindling to virtually nothing.

[123] However, the fact that institutions have adopted Fair Dealing Guidelines does not mean that the Board must rely on these. Indeed, in Access – K-12 (2005-2009) the Board concluded that the limited rules in place at the institutions were not sufficient to establish a general fair dealing practice. [34] Instead, the Board relied on information contained in logging stickers collected during a study to categorize particular copying events as fair dealing or not. [35]

[124] In a later case involving tariffs in that field, Access – K-12 (2010-2015), the Board also declined to enter into an analysis of the Fair Dealing Guidelines on the grounds that there was not a generally uniform practice that could have been evaluated in the manner contemplated by CCH. [36] Instead, the Board relied on evidence supporting a different approach to fixing the relevant royalty; the Federal Court of Appeal upheld the Board’s conclusion in that regard. [37]

[125] Similarly, we do not examine the institutions’ Fair Dealing Guidelines in these proceedings. It is unnecessary for us to do so. We rely instead on proxies which implicitly determine the amount of copying by the institutions that is fair dealing. We explain why in more detail in the economic analysis portion of these reasons.

B. Non-Affiliated Rights Holders

[126] As a collective society, Access enters into “affiliation agreements” with copyright owners in varying capacities. It also claims to represent what it refers to as non-affiliated rights holders, in the circumstances described below, and asserts that it is entitled to do so on the basis of an implied agency by ratification.

[127] This issue bears on the size of Access’ repertoire for the purposes of establishing the tariffs and on Access’ legitimacy in collecting royalties on the part of its non-affiliates. We examine it further now and, for the reasons that follow, reject Access’ claim in relation to these proposed tariffs. There can be no implied agency by ratification when Access cannot demonstrate that monies have been paid to, and accepted by, non-affiliates for the reproduction of works within the Access repertoire. We find that to be the case here.

i. Affiliation Agreements and Non-Affiliates

[128] Its affiliation agreements allow Access to grant licences to users authorizing them to reproduce the works of rights holders who are Access members. Access refers to owners with whom it has signed an affiliation agreement as “affiliated rights holders” or simply “affiliates”. [38]

[129] In practice, Access’ affiliation agreements are generally entered into with publishers. Access explained that in the publishing world, publishers obtain either an assignment or a licence from authors to use the work. Because they are the copyright owners (via assignment) or hold a licence for the reproduction rights, it is from the publishers that Access obtains the right to manage the right to reproduce the works, through the affiliation agreement. [39]

[130] Access also enters into bilateral agreements with foreign collective societies, in which the latter give Access a mandate to grant licences authorizing users to reproduce the works in their repertoires. [40]

[131] Finally, Access claims to represent the works of rights holders who have not signed affiliation agreements with it or whose works are not represented by a foreign collective society with which it has entered into a bilateral agreement. Access refers to these rights holders as “non-affiliated rights holders” or “non-affiliates”.

[132] When Access has information that works published by non-affiliated publishers have been reproduced, its practice is to pay royalties to these publishers, via a cheque accompanied by an explanatory letter. [41] Access does not pay non-affiliated authors directly. It is up to the publishers to pay the portion of the royalties owing to each rights holder, if applicable. [42] When these cheques are “cashed”, Access claims that this creates an agency by ratification, which Access also calls a retroactive implied agency, [43] authorizing Access after the fact to act as the agent of the non-affiliated rights holder for the purposes of copies made by a user with an Access licence. [44]

[133] Access has always operated this way. Its evidence is that sometimes non-affiliates even receive royalties from Access for years without Access ever asking them to become members. Access makes such efforts on a case-by-case basis, generally according to the size of the repertoire of a non-affiliated publisher to which it has paid royalties. [45] In response to a question from Vice-Chair Majeau, Access advised that it considers copies made in this context to be part of its repertoire. [46]

[134] At the hearing, Access explained that it is through mandatory reporting of copies or bibliographic surveys that it learns about the reproductions of works of non-affiliated rights holders having been made. [47] The licences entered into between Access and the educational institutions state that certain types of copies must be itemized in detailed reports. For the purposes of these proposed tariffs, the relevant reportable copies are those in course packs or reproduced on digital networks. [48] However, copies distributed in an ad hoc manner, in class or by email, do not need to be reported. Access therefore has no information about the latter group of copies.

[135] With respect to copies that are not reported, and for which Access has no information, Access explained that it only pays royalties to its affiliated members. They are shared among Access’ affiliates in accordance with distribution rules developed by its Board of Directors without regard for the fact that some of the royalties might relate to unreported works of non-affiliates that were actually copied. Accordingly, rights holders who have not signed an affiliation agreement with Access receive none of the royalties collected for copies about which Access holds no information.

[136] The proportionate number of copies falling into the foregoing category is quite significant in this case. In response to a Board request for information regarding the ratio of copies not included in course packs to copies made for inclusion in course packs, Access advised that the ratio was around 6.2:1, based on data from the University of Toronto. [49]

ii. Analysis

[137] We do not accept Access’ theory of implied licence with retroactive ratification, in the circumstances of these proceedings.

iii. Implied Agency

[138] A situation of implied agency or authority may arise, in a given factual context, where “one party has conducted [itself] towards another in such a way that it is reasonable for that other to infer from that conduct assent to an agency relationship.” [50] In a judgment of the Ontario Court of Appeal, [51] the notion was summarized as follows:

While agency is often created by an express contract, setting out the scope of the agent’s authority, the creation of an agency relationship may be implied from the conduct or situation of the parties. Whether an agency relationship exists is ultimately a question of fact, to be determined in the light of the surrounding circumstances.

[139] An agency relationship may be formed retrospectively through ratification where the principal accepts and adopts the act of an agent done prior to receiving the authority to act. By adopting the agent’s act, the principal ratifies the relationship and the previously unappointed agent is treated as having been authorized at the time in question. This ratification can arise through subsequent express consent or by implication through the conduct and situation of the parties. [52]

[140] A finding of implied agency is not to be too easily made. As the Alberta Court of Appeal has put it, implied authority rests on “clear unequivocal evidence that demonstrates that a principal has in fact consented to the agent’s having authority to act on his or her behalf.” [53] The burden lies on the party asserting the implied agency claim. [54]

[141] Here, Access argues that, as soon as the non-affiliated rights holders cash the royalties Access paid them, an implied agency is formed, retroactively legitimizing the fact that Access is acting as its agent and thus able to authorize users to reproduce the works of these non-affiliated rights holders. In support of its theory, Access relies on the definition of “collective society” in section 2 of the Act: [55]

“collective society” means a society, association or corporation that carries on the business of collective administration of copyright or of the remuneration right conferred by section 19 or 81 for the benefit of those who, by assignment, grant of licence, appointment of it as their agent or otherwise, authorize it to act on their behalf in relation to that collective administration […] [Our emphasis]

[142] Access explains how the agency takes form as follows:

[…] Access Copyright represents the works of rightsholders who have, by implied agency, authorized Access Copyright to act on their behalf. Such circumstances include payments to publishers for whom Access Copyright has evidence or reporting indicating that their works were copied by licensees (“Non-Affiliated. Rightsholders”). When Access Copyright is notified that a work of a Non-Affiliated Rightsholder has been copied under licence, Access Copyright distributes the associated royalties to the Non-Affiliated Rightsholder. When such Non-Affiliated Rightsholder cashes the cheque or accepts the payment, Access Copyright acts as that rightsholder’s agent. The acceptance by the rightsholder of the payment from Access Copyright ratifies the transaction retroactive to the date on which Access Copyright licensed the work. Access Copyright’s policy has always been to represent all rightsholders and to compensate all rightsholders for the reproduction of their works. [56]

[143] The Board has accepted this argument in a previous decision, Access – K-12 (2005-2009), finding that:

[133] Non-affiliated rights holders who cash the cheque they received as a result of the distribution of royalties based on the volume study, retroactively and implicitly grant to Access the power to act on their behalf in respect of copies captured by the study. They cannot take proceedings for infringement of copyright against the person who made the copy. […] The existence of an implied agency relationship, arising from the cashing of the cheque and limited to only those copies that were captured in the study, is sufficient to lead us to include these copies in the calculation of remuneration.

[134] […] Here, by cashing the royalty cheque, non-affiliated rights holders performed an act confirming that Access had the right to authorize reproduction of the rights holder’s work for the sole purpose of the copy captured by the volume study.

[144] Access raises a second argument as well in support of its ratification theory. Access claims to represent all non-affiliated rights holders except those who specifically expressed that they did not want Access to represent them. It bases this claim on the fact that in almost all cases the rights holders to whom it did pay royalties cashed their payment cheques, arguing that this creates the basis for a broader state of retroactive implied agency. We return to this submission later.

iv. Implied Agency by Ratification is Not Established

[145] The difficulty with Access’ position in these proceedings is that the evidence does not support a finding of either payment to or acceptance of payment by Access’ principals – the non-affiliated rights holders. There can therefore be no implied agency by ratification on this record.

[146] As noted above, Access has no information about the vast bulk of copies that are the subject of this argument because the copying is not reported to it. Moreover, as Access explained, it only pays royalties respecting such copies to its affiliated members; no royalties are paid to non-affiliated rights holders whose works may have been reproduced but that are unknown to Access. There being no royalties paid to them, non-affiliates have no cheques to cash and, accordingly, there can be no ratification in favour of Access by those non-affiliates whose works were reproduced but not reported.

[147] This conclusion is consistent with earlier Board decisions regarding Access and the education sector.

[148] As noted above, the Board accepted an implied agency argument in Access – K-12 (2005-2009). In that proceeding, however, the Board had clear evidence of copying – based upon a volume study prepared by the parties – and of the payment of royalties to non-affiliates and the cashing of related cheques. It limited the implied agency relationship to those copies captured by the volume study and included those copies as part of Access’ repertoire for purposes of calculating compensation.

[149] In a decision with respect to copies made by governments, [57] the Board came to a different conclusion. In that proceeding, the evidence was that no payments had been made by Access for non-affiliate copies identified in the relevant volume study. It followed that there could have been no cashing or acceptance of royalty payments, and therefore no formation of an agency relationship between Access and the non-affiliated rights holders on that basis. The record in that proceeding also demonstrated that Access’ distribution model was not based on copies that had actually been made; instead, it distributed “repertoire payments” among its affiliates irrespective of the number of copies actually made and even though some of those copies were undoubtedly copies of works by non-affiliates. The Board held that the acceptance by rights holders of royalties that were not linked to actual copying events cannot serve as a basis for a finding of implied agency by ratification. [58]

[150] Similarly, in Access – K-12 (2010-2015), the Board declined to include copies of works of non-affiliates in Access’ repertoire for purpose of calculating royalties, on the basis that implied agency had not been established. Although the Board had evidence of the same volume study it had considered in Access – K-12 (2005-2009), it also had much more complete evidence regarding Access’ method of distributing royalties.

[151] For example, when Access made payments in relation to non-affiliates, it made the payments to publishers and not to the authors directly, leaving it to the publishers to distribute the authors’ share. The record did not show whether the publishers owned the copyright in the copied work, however, and no agency could be formed where the publisher – not being the owner of the copyright – had no power to ratify the copying act. In addition, Access could not say which works captured in the 2006 volume study had continued to be copied after 2006, and it had no information about the works of non-affiliates that had been copied but not included in the volume study.

[152] We referred earlier to Access’ additional argument that it is entitled to represent all the non-affiliated rights holders except those who choose to opt out and specifically express that they do not want Access to represent them. This issue arose in Access – Governments as well. There are two reasons why the argument must fail, however. First, it is based on the premise that in almost all cases, rights holders to whom it pays royalties cash their cheques, thereby creating a retroactive implied agency. The record does not establish any such royalty payments to non-affiliates here. Second, the theory approaches the issue from the wrong angle: the reproduction right accorded to authors under the Act is an exclusive right, and it is therefore the responsibility of Access to seek and obtain the authorizations required to manage reproduction rights and not the responsibility of the rights holders to advise Access they do not wish it to manage those rights.

[153] The foregoing decisions of the Board serve to illustrate how a finding of implied agency is rooted very firmly in the factual context in which it arises.

[154] We do not say, therefore, that agency can never be implied or retroactive. Nonetheless, it is well-established that a collective society must establish that it has valid authorizations regarding the rights holders it is claiming to represent. [59] In the circumstances of these proceedings, given Access’ practices and distribution methodology, as described above, we are not satisfied that that a relationship of implied agency as between Access and its non-affiliates has been established to support its authorization to represent the non-affiliates for purposes of the collective administration of their works.

[155] In light of the above, we will make an adjustment in the economic analysis section below to exclude from the rate calculation the proportion of works by non-affiliated rights holders reproduced by educational institutions. We will also amend the definition of “repertoire work” as indicated in the Tariff Wording section below.

C. Agency for Digital Copies and Digital Copy Deletion Provision

i. Proposed Tariffs

[156] The proposed 2011-2013 and 2014-2017 tariffs submitted by Access include digital copies, just as the licences it has concluded since 2012 do. With respect to such copies, each contains a digital copy deletion provision requiring the educational institutions to delete digital copies upon expiration of the tariff.

[157] On this point, the proposed tariffs state the following at subsection 5(4), under Additional Conditions Regarding Digital Copies of Repertoire Works:

Where the Educational Institution is no longer covered by this tariff, the Educational Institution and all Authorized Persons shall immediately cease to use all Digital Copies of Repertoire Works, delete from their hard drives, servers and networks, and make reasonable efforts to delete from any other device or medium capable of storing Digital Copies, those Digital Copies and upon written request from Access Copyright shall certify that it has done so.

[158] On February 18, 2014, the Board asked Access to explain the reasons underlying the inclusion of a digital copy deletion provision in the proposed tariffs.

[159] In its response, [60] Access explained that it was a standard clause used in reproduction agreements in the digital publishing industry and that its purpose was to address the ease with which digital copies can be disseminated, for example through digital platforms or email. With paper copies, this ease of dissemination would be a less-pressing concern. Access also stated that the 2012 Model Licence contained a similar clause and that this did not prevent several educational institutions from agreeing to this licence. [61]

[160] Moreover, Access explained that the agencies granted by its members for digital copies do not authorize it to license any digital copying of its affiliates’ works without the corresponding requirement to delete such digital files upon termination of the licence. [62]

[161] This issue is more than a question of semantics in the wording of the tariff; it involves the issue – a fundamental one – of the agency relationship Access has with its members with regard to digital reproduction.

a. Access – Governments

[162] The issue of the digital copy deletion provision is not new. It arose for the first time in Access – Governments, rendered in 2015, after the foregoing questions had been posed to Access in these proceedings.

[163] In Access – Governments the proposed tariff included a digital copy deletion provision identical to the ones here. After the hearing, during discussions on the administrative provisions and the wording of the tariff, the Board asked for the parties’ submissions on the issue of whether the tariff should contain such a clause. In this regard, objectors opposed the inclusion of such a clause in the tariff on the ground that this condition had no basis in law: reproductions made in accordance with a licence or tariff were duly authorized, and the fact that the licence or tariff has expired does not change the fact that these are lawful copies. In this context, nothing would prevent a user from keeping the lawfully made digital copies. They submitted that the Act protects the reproduction right, not the right to keep or read a copy. [63]

[164] As in these proceedings, the evidence showed that the reason Access had included this clause in its proposed tariff was that its affiliates, through Access’ standard affiliation agreement, would not grant Access the right to authorize digital copies unless a clause requiring the destruction of this type of copy upon termination of the licence was included in the licence between Access and all users.

[165] The Board agreed with the objectors and determined that it could not include such a clause in the tariff. It then asked the parties for their input on the effect of not including the digital copy deletion provision in the tariff.

[166] The objectors submitted that since Access could not authorize digital copies of the works of its affiliates without the requirement to destroy these files upon termination of the licence, the non-inclusion of this provision in the tariff would exclude digital copies from the application of the tariff. It followed that the copies should not be considered in the calculation of the rate. [64] For its part, Access informed the Board that it had obtained authorization from its Board of Directors to grant licences for digital copies despite the absence of a deletion provision. [65]

[167] The Board ruled as follows:

It is clear from this response that Access did not, at the time it filed the proposed tariffs, and still does not, have the authorization from all – or perhaps any – of its affiliates to license the making of digital copies without such a deletion requirement. Whether the Board of Directors has authorized Access to do something or not is not sufficient. Access’ authority to license the copying if a work flows from a licence granted by the owner of copyright; where it does not have such a licence, it has no authority to license the use itself. Access’ Board of Directors cannot grant to Access rights which owners of copyright did not themselves grant to it. [66]

[168] The Board therefore excluded digital copies from the calculation of royalty compensation in Access – Governments; this represented a significant proportion of the copies captured by the volume study.

[169] Access applied to the Federal Court of Appeal for judicial review of that decision. However, it did not request a review of the Board’s findings regarding the exclusion of digital copies. [67] The Court dismissed this application for judicial review in March 2018. [68]

ii. Action Taken by Access following Access – Governments

[170] Although Access continues to maintain that the digital copy deletion provision is a standard clause in the industry and should be included in the tariffs, it now advises that it is willing to have the Board remove this provision from the tariffs. It says that it now has the authority required to allow digital copies despite the lack of a digital copy deletion provision in its licences, thereby avoiding a result similar to that in Access − Governments, namely, the exclusion of digital copies from the tariffs. [69]

[171] In this respect, Access was able to demonstrate that it had its members sign a waiver, retroactive to January 1, 2010, regarding the removal of digital copy deletion provision that was in its 2002 standard affiliation agreement, [70] or sign a new affiliation agreement in 2015 which did not include this condition. Thus, as of December 17, 2015, a few weeks before the hearing, close to 91 per cent of publishers with Access had signed the waiver and/or the new 2015 affiliation agreement [71] which did not include the clause requiring Access’ agreement to include the digital copy deletion provision in the licences entered into with users. [72]

[172] What, then, is the scope of Access’ agency with regard to digital copies? And what is the impact of the fact that approximately 10 per cent of publishers with Access did not sign either the waiver of the requirement to destroy copies or the new affiliation agreement that does not contain this condition?

[173] As regards the first question, considering the actions Access took, we are of the opinion that, contrary to the situation in Access – Governments, Access has agency to authorize digital copies with no conditions. As for the second question, this impact is reasonably insignificant. These figures date back to December 2015. [73] It is reasonable to believe that after 2015, Access’ rights holders continued to migrate towards the new affiliation agreement such that it is likely Access has now obtained from all of its rights holders the rights that would allow it to grant licences without the digital copy deletion provision. [74]

[174] For the reasons above, we conclude that the record supports Access’ position. We hold that Access may legitimately authorize digital copies, with no conditions.

[175] Lastly, Access submits that insofar as the Board deems that there are gaps in the capacity of Access to authorize digital copies without conditions, section 30.02 of the Act would fill these gaps as of November 7, 2012, when it came into force. This section, which only affects educational institutions, provides that rights holders who have authorized Access to enter into agreements with educational institutions for paper copies are deemed to have also authorized Access to enter into agreements for digital copies.

[176] Access is correct when it states that even if there were a problem with its digital agency, this problem would have been resolved as of November 7, 2012. [75] In this instance, we have found that Access has the required authority to authorize digital copies with no conditions, however; section 30.02 therefore need not apply.

D. Consequences of the Absence of Educational Institutions

[177] The absence of institutional objectors and the associations representing the educational institutions, together with their re-emergence through the means of “comment” letters after the hearing and closing arguments, have affected these proceedings. These features of the proceedings raise issues with respect to the completion of the record, the Board’s role in the absence of any institutional objectors, and the weight to be given to the comment letters. Not surprisingly, they have contributed to the length of the proceedings as well.

[178] We address those issues now.

i. The Record and the Role of the Board in the Absence of Educational Institutions

[179] In Access’ submission, the Board’s role is to certify a fair and equitable tariff based on the evidence before it. The fact that a user subject to a proposed tariff has decided to withdraw its objection and not participate – and as such not file evidence – has no impact whatsoever on this role. [76]

[180] Accordingly, Access says, the Board’s role is to assess the evidence on the record in an impartial manner, bearing in mind the public interest, of course, but without trying to mitigate the consequences of the absence of a user who has decided not to take part in the proceedings. In other words, the educational institutions must take responsibility for having chosen not to participate. In the end, only Access filed any evidence. The Board must weigh this evidence, and its decision must be based on that evidence and it alone.

[181] While there are general aspects of Access’ submission with which we agree, we would describe the role the Board must play in this case more accurately as follows.

[182] Since it acts as guardian of the public interest, [77] the Board may – and sometimes must – ask questions to ensure that it has a proper understanding of the issues in a case. It has broad powers to do so and can also take steps to complement a partially defective record. [78] The Board is not restricted to the parties’ submissions and may, on its own initiative, seek out evidence, subject always to questions of procedural fairness. [79]

[183] Whether or not there are any objectors, the Board’s role on the whole remains the same. That role is to assess the evidence before it and to base its decision on that evidence. Should the Board render a decision that is not based on the evidence, but on unsupported speculation, its decision would be, as the Federal Court of Appeal noted in Bell, arbitrary and unreasonable. [80] Although the Board has broad powers, these powers nonetheless do not permit the Board to stand in for an absent party, nor do they give rise to “a duty on the part of the Board to create [a record].” [81]

[184] Here, the lack of participation by institutional objectors has left us without the benefit of an opposing party’s evidence, analysis and arguments to counterbalance those of Access and its evidence. [82] Because we are required to approve a tariff that is fair and equitable, [83] and because we also considered the public interest, we were required to examine the record with utmost care, and even to ask questions to supplement it where appropriate. This we have done. However, that is where our role ends. From a procedural standpoint, Access should not have to bear a heavier burden than it normally would have, had the educational institutions decided to participate, nor should the Board have to assume the mantel of counsel for the educational institutions. In the end, it is the institutional users that chose not to take part.

V. ECONOMIC ANALYSIS

A. Introduction

[185] Access proposed that for universities, the Board certify a royalty rate of $26 per FTE, based on the Model Licence it signed with AUCC, [84] for all the years covered by the two proposed tariffs, i.e., 2011 to 2017. Access described the Model Licence as a “reasonable proxy”. [85]

[186] Similarly, it proposed that the rate of $10 per FTE be certified for colleges for the same years, based on the Model Licence with ACCC. [86]

[187] We use the term “target” for the price (or tariff) the Board is tasked with setting. We use the term “proxy” for the price in a market related to the target. A proxy is said to be good if the proxy is determined in a pseudo-competitive market [87] and the proxy market is “similar” to the target market. [88] To the extent that these two conditions are not met, the proxy may have to be adjusted to determine a fair and equitable target tariff. The more uncertain the adjustments necessary to determine a fair and equitable tariff are, the less the proxy is usable. If the adjustments are subject to a great deal of uncertainty, the proxy may not be usable at all.

[188] For the reasons that follow, we conclude that the AUCC Model Licence and the ACCC Model Licence are “good” proxies for purposes of setting the respective tariffs for the years 2011 to 2014. In our view, they were entered into in a market that is sufficiently competitive and similar to the target market (with some slight adjustments) to qualify for that purpose. We do not treat the Model Licences as valid proxies for the years 2015 to 2017, however. Instead, based on a similar analysis, we adopt the Premium Licence for that purpose.

[189] For purposes of simplicity, we will focus for now only on the proposed tariffs for universities and deal with the tariff for colleges later in these reasons. The analysis is similar for each.

B. The Two Proposed Tariffs

[190] When it initially filed its proposed tariffs with the Board, Access proposed different rates for two periods: $45 per FTE for 2011-2013 and $35 per FTE for 2014-2017. When it filed its evidence, Access revised these rates to $26 per FTE for both periods.

[191] On June 9, 2016, the Board posed a number of questions to Access, [89] including the following:

Explain why there has been a decrease in the rates sought by Access Copyright in the proposed tariff for 2014-2017 ($35 for universities and $25 for other educational institutions) as compared to the proposed tariff for 2011-2013 ($45 for universities and $35 for other educational institutions).

[192] On September 30, 2016, Access responded as follows: [90]

Access Copyright filed its proposed tariff for 2014-2017 at the lower rate of $35 for universities and $25 for other educational institutions because by the time of filing in March 2013, it had negotiated new comprehensive licences with the Association of Universities and Colleges of Canada (AUCC, now Universities Canada) at a rate of $26 per FTE and the Association of Canadian Community Colleges (ACCC, now CiCan) at a rate of $10 per FTE. Prior to that time, there were no negotiated licences in place and thus no appropriate benchmarks for the proposed valuation. Accordingly, Access Copyright adjusted the proposed rates downward from the rates it proposed for the 2011-2013 tariff to more accurately reflect the going market rate at the time of filing.

As is typical of most tariff proposals, at the time of filing, Access Copyright’s valuation expert had not yet assessed the value of the Proposed Tariffs. Once Mr. Heys had the information available to him, the rates sought by Access Copyright were adjusted downward to reflect his opinion.

[193] In effect, Access contends that the tariff should reflect the market rate. However, while it is appropriate for Access to reflect the market rate at the time of filing (namely, March 31 of the year prior to the commencement of the tariff), our task is to reflect (at least to some extent) the market rate evaluated using the information available to us at the time of the decision itself.

[194] As stated above, Access began offering its new licences towards the end of 2014. We find that this corresponds to the date when the market environment changed, with lower rates characterizing such market environment. Accordingly, we will set a rate for the years 2011 to 2014, and a different rate for the years 2015 to 2017.

C. Our Proxy for 2011-2014: AUCC’s Model Licence

i. Similarity of the Proxy to the Target

[195] Access led evidence on the question of similarity, stating that:

The Benchmark Licences are arm’s-length negotiated agreements for substantially the same rights as would be granted under the terms of the Proposed Tariff and apply to the 2011 to 2013 period covered by the Proposed Tariff. In particular, each of the Benchmark Licences and the Proposed Tariff:

1. Provide the right to make paper and digital copies of excerpts of published works contained in the Repertoire;
2. Place essentially the same restrictions on the permitted copying;
3. Require that each institution collect certain information about the copying of “Course Collections” and contemplate that Access Copyright might gather bibliographic and volume information about copying through surveys of institutions; and
4. Provide that institutions will pay annual royalties based on the number of FTEs. [91]

[196] The four points of similarity raised by Access are accurate. They are also important parts of the terms and conditions of the Model Licence and the proposed tariffs. [92] This supports our choice of the Model Licence as a proxy.

[197] The greatest area of difference among these four points relates to the survey of institutions, but this difference is largely a matter of tone. While the Model Licence provides that the survey should minimize administrative burden and comply with privacy laws and policies of the licensees, [93] the proposed tariff provides that “[i]f an Educational Institution unreasonably refuses to participate in the survey or otherwise does not comply with this section, the licences in sections 3 and 10 will cease to be in effect.” [94] We do not find this difference significant.

[198] There is one further point of difference between the Model Licence and the proposed tariff. The Model Licence contains an indemnity clause (Clause 13). That clause is now irrelevant, given our findings above relating to non-affiliates and the scope of Access’ repertoire.

ii. The Question of a Competitive Market

[199] To be a valid proxy for a target tariff, the comparator need not necessarily be reflective of a perfectly competitive market.

[200] On at least two occasions, the Board has considered the question of the relationship of its mandate – the setting of fair and equitable tariffs – to the act of setting tariffs that “mimic competitive markets”. It first set out its understanding of the relationship in a 1993 decision setting various SOCAN tariffs:

The Board regulates the balance of market power between copyright owners and users; this does not mean that the royalties must be set by recourse to a freely negotiated price in a non-competitive or even a competitive market. The Federal Court [Performing Rights Organization of Canada Ltd v CBC (1986), 64 N.R. 330 (FCA) at para 30] has already enunciated the principle that a “market price” is only one of several possible rational bases for setting a tariff. [95]

[201] More recently, the Board echoed its earlier understanding of its role, particularly in comparison to the role played by rate-setters in the United States:

American rates are subject to a detailed set of statutory instructions. They are intended to reflect competitive market outcomes and to account for the promotional value of air play. In Canada, the only overreaching requirement is that the rates be fair. A market price is but one possible rationale for setting a tariff; the Board is not bound to set tariffs only on that basis. Also, the Board has repeatedly refused to take into account the promotional value of air play. [96] [Emphasis added]

[202] In economic terms, it is nearly impossible for agreements between a collective society and a user to be competitive in the strict sense. Collective societies exercise a certain amount of market power. This implies they charge a price that exceeds the marginal cost of reproduction, in part to cover the fixed cost of creating the work in the first place.

[203] A theoretical, perfectly competitive market has many buyers and many sellers. In the case where a market is created by a collective society offering for sale on a take-it-or-leave-it basis a bundle of rights in regards of a repertoire of works, the market is not competitive. While prices from competitive markets are prima facie good proxies, the price in the non-competitive market created by the collective society could still constitute a good proxy if the take-up of the offer is reasonably high.

[204] In the case of the 2012 AUCC Model Licence, we have the following data. As of September 11, 2013, 35 institutions had signed the Model Licence, the Universities of Toronto and Western Ontario has signed almost identical licences, [97] and 20 institutions that were members of AUCC had not signed a licence. [98] As of November 2, 2015, 35 institutions were licensed, of which 24 were licensed under the Model Licence and 11 had converted to the new Premium Licence (discussed in greater detail below). The remaining 22 institutions were unlicensed.

[205] There is no obvious threshold for a monopolistic market with competitive features. In this case, about 60 per cent of the market was licensed during the tariff period. In addition, this licence continued throughout 2014. In our opinion, this is a market with enough competitive features to serve as a proxy.

[206] We conclude, therefore, that the Access Model Licences are a good proxy for the tariffs: the prices from those licences are thus a good proxy for the rates in the tariffs.

iii. Another Consideration: the Copibec Licences

[207] The Board reviewed three agreements between Copibec and universities in Quebec. Broadly speaking, these agreements have the same rates and terms, and cover identical rights and repertoire.

[208] The Board has in the past used the word “comfort” to refer to a second piece of evidence that corroborates a primary model. Having reviewed the Copibec licences, we view them in that same fashion for purposes of these proceedings. While we do not utilize them as proxies themselves, they provide us with comfort for our conclusion that the Model Licences are good proxies.

[209] This observation applies both for the 2012 Model Licence (compared to the 2007-2012 Copibec licence) and the 2015 Premium Licence (compared to the 2014-2017 Copibec licence).

D. Our Proxy for 2015-2017: The Premium Licence

i. Market “Adjustment” vs. Market “Failure”

[210] Access submits that the $15 rate in the Premium Licence does not represent a market rate. It believes this rate reflects instead a market failure. [99] It states that it was limited to offering licences at this rate because educational institutions had deemed, following Alberta and legislative changes in 2012, that they no longer needed a licence. As a result, the Board should certify the $26 rate set out in the 2012 model licences for the entire duration of the tariff; unlike the $15 rate, this $26 rate would result from negotiations in a level-playing field.

[211] Contrary to Access’ submission, we are of the opinion that the $15 rate, far from constituting a market failure, is an illustration of the way in which the market adjusted in favour of the expansion of the concept of fair dealing that came about in 2012 and the uncertainty accompanying that development.

[212] What Access describes as a market failure is in fact a drop in demand that corresponds to a change in the perception of the concept of fair dealing. Whether this perception is correct or not is not relevant to the concept of market failure. Additionally, the fact that universities had purchased this Premium Licence at different prices ($12, $15 and $18) is an indication there is no market failure. If there were a true market failure, no user would have purchased this licence, whatever the price Access could have offered.

[213] Since fair dealing is a qualitative concept, it is not easy to quantify it. As the Supreme Court stated in CCH, citing Lord Denning, “it must be a matter of impression”. [100] It is for this reason that its interpretation is often contentious. Each new decision results in a new position for the parties and creates some uncertainty in the licensing market. This was seen in the preamble of the licences entered into between Access and the educational institutions we referenced above. It was also seen in Access – Governments, where Access explained that its relationship with the federal government had changed following CCH in 2004. [101]

[214] Put in place following the decisions in Alberta and Bell, the Premium Licence, itself, reflects this uncertain environment:

The parties have entered into this agreement to permit the reproduction by the Licensee of Repertoire Works at a time of uncertainty, disagreement and ongoing litigation about the scope and interpretation of the rights, obligations and exceptions, including fair dealing, under the Copyright Act. The parties acknowledge and agree that the Royalties payable under this agreement reflect this uncertainty and are not an admission of the actual value of the rights licensed or the volume of Repertoire Works Copied for any proceeding unrelated to a breach of this agreement. [102]

[215] As to the question of how Access decided on $15, Access explained that it was simply a rate it believed the institutions would accept in light of their interpretation of fair dealing and the instability of the market as a result of the 2012 changes. It felt that this rate was low enough that the institutions would enter into an agreement. [103] This corresponds exactly to the testing process offerors must go through in a market to find a new equilibrium price when there is an initial change in the demand.

[216] When Access states that it was the educational institutions’ interpretation of the concept of fair dealing that expanded, [104] it is therefore not entirely wrong. Indeed, the educational institutions interpreted the concept of fair dealing from Alberta and Bell in a certain way. They adopted a new policy regarding fair dealing; [105] they did not want to renew their licence with Access at the same rate as 2012. In response, Access offered the licences at a lower rate, as did Copibec.

[217] Our concern here is not to decide whether educational institutions went too far in their interpretations of the decisions. In any event, there is no evidence to explain how their policies are applied and implemented. [106] Although it is not germane for our purposes, we observe that in York, [107] the Federal court raised many concerns in this regard.

[218] We conclude that an analysis of the Guideline policies in the context of fair dealing is unnecessary here, because we agree with Access that the rates should be based on agreements entered into between Access and the educational institutions using the proxy approach. The proxy agreements themselves incorporate a discount for fair dealing and the existence of the Guidelines.

[219] This approach is consistent with that taken by the Board in Access – K-12 (2010-2015). In that proceeding, the opponents had presented two approaches to determine the applicable rates. The first was based on the guidelines adopted by the educational institutions, and the second, on the volume of compensable copies. Upon judicial review of this decision, both parties were critical of the Board for not having provided any detailed comments on their evidence relating to the Guidelines. However, the Federal Court of Appeal [108] determined that the Board, having given greater weight to the evidence on the volume of copies, was not required to rule on the validity of the guidelines adopted by the elementary and secondary schools. In the present proceedings, there is even greater reason for dispensing with the analysis of these policies, in our view, because no evidence on this issue was presented by the educational institutions.

[220] Far from showing a market failure, what the developments we have just described actually indicate, in our view, is that the market has adapted.

[221] In submitting why the Premium Licence rate should not be considered a valid baseline for the value of licences Access offered in 2015, Access argued that:

[…] [T]he December 2014 Licence Offers do not reflect […] a fair and reasonably royalty for the 2014 Tariff because they were made solely as a consequence of the failure of the market, in which copying without a licence was unilaterally determined by the educational sector to be a viable preferred alternative to copying with a licence. [109] [Emphasis added]

[222] For this reason, Access says that it was forced to offer very low rates, which are not market rates, because the institutions had deemed that a licence was no longer necessary. [110] However, the facts do not support this finding. While close to two thirds of universities had signed the 2012 Model Licence at the rate of $26, another 23 per cent signed the Premium Licence, or 11 universities out of 57, representing 8 per cent of students. [111] Excluding the 20 universities that would not sign a licence at any price, the acceptance rate is around 30 per cent, or 11 universities out of 37. Yes, there was a reduction in the acceptance rate, but it was not reduced to zero. It is therefore incorrect to state that all the institutions had decided that they no longer needed a licence. To repeat, this demonstrates a market adjustment rather than a market failure, in our opinion.

[223] We turn, then, to our analysis of the Premium Licences as a valid proxy.

ii. The Premium Licence Offering

a. Preliminary Considerations

[224] In December 2014, Access offered two families of licences to its university customers (and prospective customers) to replace the 2012 Model Licence – Premium and Choice. [112] We focus our analysis on the Premium Licence, primarily for two reasons. First, the Premium Licence is expressed as a per-FTE price without a per-page supplement, just like the proposed tariffs. Second, as explained by Ms. Levy, the Choice Licence required institutions to have a centralized copyright management system that could track, report, and collect payment for use of the licence. [113]

[225] Table 2 shows the features of the Premium Licence, as presented in Access’ marketing material. [114]

Table 2: Features of the Premium Licence

Table 2: Features of the Premium Licence

Item	Information
Agreement Type	Comprehensive agreement with a flat fee per student (FTE) providing institution-wide coverage for paper and digital copying
One Year Price	$18/FTE
Three Year Price	$15/FTE
Five Year Price	$12/FTE
Pay Per Use	Copying 20% to 25% of a work, $0.12 per page
Coverage	Up to 20% of a work, and 20% to 25% on a pay-per-use basis
Reporting	Logging of all course pack uses and any use from 20% to 25% of a work
Indemnity	Paper and digital copying within licence limits

[226] To simplify our analysis, we focus on the three-year licence, priced at $15/FTE, ignoring both the longer and shorter licences, as well as the per-page costs for exceeding 20 per cent of a work. We do so because the $15 licence is the median-priced licence. It is also the (unweighted) average-priced licence and very close to the weighted, average-priced licence, equal to $15.54.

[227] In conducting our analysis, we have taken into account several features of the Premium Licence.

[228] First, the Premium Licence contains a provision for a marginal cost rate of $0.12 per page when exceeding the copying limits. The tariff proposed by Access does not contain a marginal cost component. This is true both of the tariff originally proposed by Access for 2014-2017, at $35/FTE, and the revised proposal, at $26/FTE. As explained by Access, copying beyond the limits of the licence was always subject to a transactional licence with a marginal cost. The only difference for the Premium Licence is that the magnitude of such marginal cost was made explicit for copies that exceeded the licence limits by a fixed amount. [115]

[229] Second, the price per FTE is to be increased according to “all-items” inflation, beginning on September 1, 2017. [116] This is not relevant here, since the Board typically sets royalties for the January to December year.

[230] Third, the Premium Licence includes reporting requirements identical to those in the 2012 Model Licence plus one additional reporting requirement: the course code and the name of the instructor associated with whom the course pack copying is being made. [117] This is not relevant here, since this additional reporting requirement does not add additional value to the licence for Access or add any material additional cost to complying with the licence.

[231] Fourth, it may be argued that extending the copying limit to 20 per cent of a work has a considerable amount of value to the licensee and we should account for this in some way. In general, we agree. Whenever a licence is used as a proxy for a tariff, the terms of the proxy need to be compared to those of the tariff. Where they differ in a material way, the Board has two options. It can make modification to the proxy to account for differences between the proxy and the tariff. Or it can reject the proxy altogether. We adopt the first option here.

[232] Since the additional copying limit has value, there are again two possible approaches. We could modify the copying limits in the final tariff to match those in the proxy and keep the proxy price in the tariff. Or, we could adjust the proxy price to reflect the differences between the proxy and the tariff and keep the copying limits as in the proposed tariff. We prefer the former approach for the following reasons.

[233] First, the copying limits are precise – 10 per cent in the proposed tariff and 20 per cent in the Premium Licence. Any change to the price would involve a complicated calculation that necessarily introduces imprecision to the mapping between the proxy and the tariff. The first step for the calculation is to select a data set. There are several such data sets available, including course pack copying or digital copying (or both), covering different periods. All of these data sets suffer from one major flaw – they do not contain information on the number of pages in the books copied. Since that information is not easily obtained through automatic means, manual entry of the data requires selecting a sampling strategy. All of these choices introduce imprecision into the tariff.

[234] Second, preliminary and rough calculations, applied with common-sense assumptions suggest that a price adjustment would likely be small. The amount of effort that would need to be expended to get a more precise calculation might not be justified, given that the ultimate adjustment would not likely alter the royalties collected by a significant amount.

[235] The issue of the limits raises, as a collateral matter, one additional issue – the question of the years of the tariff. We dealt with that question separately earlier.

b. The Question of Valuation Date

[236] Mr. Heys testified in his Report, and orally, that the Premium and Choice licences were unsuitable as proxies because they were entered into after the beginning of the proposed tariff period – January 1, 2014 – and therefore did not provide an appropriate indicator of FMV royalty rates for the proposed periods. [118] While we agree that FMV is defined in relation to a valuation date, we do not agree that the first day of the term of the proposed tariff is necessarily the applicable valuation date. Accordingly, we do not accept Mr. Heys’ approach in this respect.

[237] It is not self-evident that the valuation date for a tariff from January 1, 2014, need be the beginning date of that tariff. From an economic perspective, other possibilities include January 1, 2016 (the midpoint of the tariff period), December 31, 2017 (the endpoint of the tariff period), or even January 18, 2016 (the start of the oral hearing). Indeed, on at least one prior occasion, the Board has adopted a valuation date that did not correspond to the beginning of the proposed tariff period, choosing instead the midpoint of the tariff. [119]

[238] Accordingly, we do not view the fact that the Premium Licence was entered into after the beginning of the proposed tariff proceeding to be an impediment to its usefulness as a proxy.

c. The Number of Signatory Institutions is Representative

[239] In his final report, Mr. Heys gave evidence about the number of institutions that signed various licence offerings. [120] There are 57 AUCC member institutions. Of these institutions, 35 signed the 2012 model licence. By October 19, 2015, 11 institutions had transitioned to the Premium Licences and 24 remained under the 2012 model licence that expired December 31, 2015.

[240] We do not have evidence about what happened between October 19 and December 31 of 2015; there are several possibilities. Some of the 24 institutions may also have transitioned to the Premium Licence. Given the narrative as to what has happened in the market in response to the evolving fair dealing landscape, we think it more likely, however, that most, if not all of them allowed their licence to expire. Measuring the take-up of the Premium Licence requires having a view about a certain group of 20 institutions. These institutions did not pay under the interim tariff and did not sign an AUCC Model Licence. Their public statements suggest that they were not willing to be licensed by Access at any price. In our view, it is a reasonable inference that these institutions “left the market”. Accordingly, we treat the Access-AUCC market as consisting of 37 institutions, not 57. Given this analysis and the evidence in the preceding paragraph, we proceed on the basis that there are 11 licensed institutions out of 37 (or just under 30 per cent).

[241] We observed above that a market in which 60 per cent of potential participants purchase the licence at a given price is functioning well and the given price is a market price. In the case of the Premium Licence, there are about 30 per cent of potential participants purchasing the licence at the given price. [121] We are satisfied, in the circumstances here, that this, too, is reflective of a functioning market and market price.

[242] Mr. Heys explained that “Whether or not a majority of the institutions have signed the model licence is not relevant to FMV.” [122] We agree.

[243] The relevant question is whether or not the signatories to the Premium Licence are representative. These 11 institutions come from five provinces, so they can reasonably be said to be geographically representative. They are relatively small institutions, with an average enrollment of about 4,000 FTEs. [123] In this respect, these institutions are not representative.

[244] However, we are less concerned with size-representativeness than we might be in other tariffs. Because the proposed tariff is structured as a payment per FTE, there is a natural scaling process. It may be argued that larger institutions are better able to structure their licences to avoid obtaining a licence through Access; if this is the case, these institutions would opt out of any licence offered by Access or tariff approved by the Board.

[245] We are satisfied that these 11 licensees are representative in the sense ascribed to that term by Mr. Heys.

E. The Volume-times-Value Methodology

[246] The Board has used the volume-times-value (VtV) methodology in decisions involving Access in the past. However, we are not inclined to do so in this proceeding.

[247] In Access – Governments, the Board described that methodology as follows:

[T]he Board established the royalty rates per FTE on the basis of the volume of compensable copies of published works, multiplied by the estimated per-page value of each genre of copied works, divided by the total number of FTE students. [124]

[248] Neither Access nor Mr. Maguire asked the Board to adopt the VtV methodology. Access had requested its expert, Mr. Heys, to provide a calculation based upon it, and he did so, although he himself expressed reservations about the “substantial limitations” in the data upon which the volume estimates were based. [125]

[249] Given the Board’s use of the VtV methodology in previous decisions, and Mr. Heys’ contribution on the record, we nonetheless considered this option but are not inclined to adopt it in these proceedings.

[250] We agree with Mr. Heys’ reservations about the limitations in the relevant data. In addition, we have our own reservations respecting several questionable assumptions made by him in calculating the royalties for universities, ranging from $133.18 to $146.23 per FTE per annum.

[251] First, he assumes that per-page pricing by genre of work copied can be transposed from K-12 schools to post-secondary institutions, without adjustment. Second, the data he uses for the volume portion of the calculation imply that copying at colleges exceeds that at universities, which contradicts other data filed in this proceeding (Exhibit AC-5, for example). Third, Mr. Heys assumes the same value for digital copies as for paper copies. Finally, he assumes all copies are compensable.

[252] We do not accept the soundness of these assumptions in the circumstances.

[253] Taking all of these factors into consideration – particularly the fact that neither party recommended it, and Mr. Heys’ reservations about its application – we find that the VtV methodology, as presented by Mr. Heys, is not useful for setting a royalty rate in this matter.

F. Adjustments to the Proxies

[254] In general, the Board considers the process of choosing and adjusting a proxy as an iterative one. If the chosen proxy requires so many or such extensive adjustments as to render the original proxy unrecognizable, the chosen proxy is not reasonable and should be discarded in favour of a proxy requiring fewer adjustments, if possible. Here, three potential adjustments are relevant: fair dealing, insubstantial copying, and non-affiliate copying. We examine each in turn and, in the end, conclude that none affects our decision to use the Premium Licence as a proxy. We do adjust for non-affiliate copying, however.

i. Fair Dealing

[255] In its responses, Access made the following points. First, publication of a fair dealing policy without tendering evidence that the policy is enforced is insufficient to find that copies made pursuant to the policy are fair. [126] Second, the rates proposed by Access include a fair-dealing discount. As Access explains:

Access Copyright has proposed to the Board a tariff rate that incorporates a discount for fair dealing for educational purposes, as well as other exceptions under the Copyright Act. The Benchmark Licences are the result of arms’ length negotiated licences for the bundle of rights licensed by the Proposed Tariff, including all exceptions available to the users of the tariff. [127]

[256] We agree. Indeed, as noted above, the 2012 AUCC Model Licence and the Premium Licence confirm this to be the case.

[257] Considering the evidence and arguments tendered by Access, we see no reason to discount further the proxies on account of fair dealing.

ii. Insubstantial Copying

[258] In the most recent case relating to copying in elementary and secondary schools, the Board used the following approximations: any copying from an article, whether from a newspaper, magazine, or journal, was presumed to be substantial; copying of one or two pages from a book (or consumable) was presumed not to be substantial; any other such copying is presumed to be substantial. [128]

[259] The Federal Court of Appeal accepted this approach as not unreasonable.

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. [129]

[260] In the present matter, we have no evidence that would lead us to conclude that the prices in the Model Licence do not already reflect the amount of substantial and non-substantial copying that takes place. As such, no further adjustments are required.

iii. Affiliate Adjustments

[261] As explained above, we need to exclude from the calculation of the rate the copying of works whose authors or publishers are not affiliated with Access Copyright.

[262] In this respect, on June 9, 2016, the Board sent a number of questions to Access. [130] One of these questions concerned a calculation of the fraction of course-pack copying that was of non-affiliated rights holders. The Board had calculations and asked Access to confirm or correct them.

[263] In its response, Access made the following points. First, the Board used the wrong dataset. [131] Second, it is appropriate to analyze data from 2011-2014, since these data cover the period of the proposed tariffs. [132] Third, the Board’s calculations are incorrect. [133]

[264] Access then filed the following calculations:

For the correct calculations of repertoire status and share of non-affiliated rights holders, please see section 2 of the Gauthier Report. Table 1 provides accurate calculations of the repertoire status of the 2011-2014 datasets. After reapportioning the volume associated with the “Unknown” category, the results are as follows:

· 44.4% of the volume is classified as “Affiliate”;

· 50.9% of the volume is classified as “Bilateral”;

· 3% of the volume is classified as “Agency” (i.e., the non-affiliated rights holder share);

· 1.4% of the volume is classified as “Excluded or Public Domain”; and

· 0.2% of the volume is classified as “Foreign – no bilateral”. [134]

[265] While we have some concerns that these figures may overstate the volume associated with the categories “affiliate” and “bilateral”, thereby understating the volume associated with the three remaining categories, we accept them for the purposes of our analysis.

[266] Access’ repertoire is comprised of the “affiliate” and “bilateral” categories. The remaining three categories are not part of Access’ repertoire. We have explained earlier in these reasons why Access’ non-affiliates are not included in its repertoire or relevant for the calculation of royalties. Works classified as “Excluded or Public Domain” or as “Foreign – no bilateral” fall into the same category.

[267] The sum of the percentages for the latter three categories is 4.6 per cent. Accordingly, the proxies are adjusted downwards by 4.6 per cent, to account for the difference between Access’ repertoire as it thought it was when it offered the 2012 Model Licence and the 2015 Premium Licence and Access’ repertoire as it actually is.

G. Final Rates

[268] Given the absence of a true competitive market that the Board could consider in fixing royalties, we are of the view that the adjusted proxies are reasonably indicative of the prices that would have been agreed upon between a willing buyer and a willing seller acting in a competitive market with all relevant information, at arm’s length and free of external constraints. The rate we thus set for universities for the years 2011 to 2014 is $24.80 per FTE per year. This rate was obtained by applying a deduction of 4.6 per cent to the rate of $26 to account for the copying of works by non-affiliates. Similarly, the rate we set for the years 2015 to 2017 is $14.31, resulting from a 4.6 per cent deduction of the rate of $15.

H. The College Licence

[269] To this point, the Decision has focused only on the royalty rates for universities, for one main reason. The evidence shows that there is much more copying occurring at universities that at colleges. We find it more appropriate, then, to set the college rates as a function of the university rates than vice-versa. It remains to do so.

[270] We consider the rates we set for universities to be the best proxy for the college tariff. The evidence shows that the amount paid in 2005-2010 for copying at universities was about 2.6 times as much as was paid for copying at colleges in the same period. It was on that basis that Access set the rates at $26 and $10, respectively. As such, we simply divide the university royalty rate by 2.6 to obtain the college rate. We have no reason to expect that this ratio is different for the two tariff periods. We also see no reason to believe that the adjustment of 4.6 per cent for non-affiliates is different between colleges and universities.

[271] Accordingly, the (before-adjustment) rates we fix for colleges are $10 for the years 2011 to 2014 and $5.77 for the years 2015 to 2017. After adjustment, the rates we set for colleges are $9.54 per FTE per year for 2011-2014 and $5.50 for 2015-2017.

I. Transactional Licences

i. Contents of Letters Received After Hearing

[272] On February 9, 2016, after the hearing, York University wrote to the Board to inform it that it was party to many licences with third parties other than Access for the reproduction of copyrighted works, be it individually or as part of a group of university libraries.

[273] It submitted that, for this reason, the tariff to be certified by the Board should, in addition to the flat-rate blanket licence, include a transactional component that would allow it to pay per use. This would reflect the fact that most of its needs are covered by the licences concluded with third parties. With a transactional licence, it would avoid having to pay twice for the reproduction of literary works, which it would inevitably have to do if the institutions had to pay royalties according to a flat rate (blanket licence) even though reproduction of the work is already authorized by a licence concluded with a third party.

[274] York stated that it was making these comments because it felt that Access had not sufficiently addressed this issue:

We note that Access Copyright’s evidence and oral submissions at the hearing did not address in any meaningful way […] alternate licensing arrangements […]

[…]

One way to structure the tariff to address the concerns noted above is to include a pay-per-use licensing option and York requests that the Board do so. [135]

[275] York also noted that it did not intend to file any evidence to support its statements:

In these comments, York makes reference to certain evidence and materials in the record before the Board, but does not seek to introduce evidence. [136]

[276] In support of its request, York stated that the Board had recognized the existence of the licences entered into with third parties in its reasons of March 16, 2011, in respect of Access Interim Tariff (2011-2013), in which it allowed Access’ application for an interim tariff for 2011-2013. York cites the following excerpts from the decision:

[45] […] A tariff applies only to those who need the licence; those who do not, need not pay. Under the general regime, which applies in this instance, users whose consumption patterns justify different rates remain free to secure, from Access or from others, transactional or other licences that will trump the tariff.

[…]

[84] Some Objectors did offer indications of possible shifts in compensable uses. New approaches are emerging to making published works available within Institutions. Significant amounts are being spent in acquiring the licences needed to use those approaches. Some of these expenses may well duplicate what was paid until now to Access. Were the Objectors to offer some evidence of the existence and extent of some forms of double-dipping, we may consider reviewing the FTE rate.

[277] York mentioned that the Board also recognized the existence of such licences in its Access Interim Tariff – Transactional Licence decision, in which it had refused the educational institutions’ request that Access be compelled to enter into transactional licences:

[29] Furthermore, if more competitive and vibrant models for lawfully accessing works are thriving and growing, then all the institutions have to do is to avail themselves of these models and ignore the Access repertoire altogether.

[278] Following the letter from York, as mentioned above, the Board received a letter co-signed by the following universities and Boards of Governors: University of British Columbia, University of Winnipeg, University of Lethbridge, University of Calgary, Thompson Rivers University and University of the Fraser Valley, as well as the Governing Council of the University of Toronto and the Governors of the University of Alberta (“Consortium’s letter”).

[279] The Consortium’s letter essentially makes the same arguments as York, namely that the Board has to offer a variety of licensing options, including a transactional licence, in order to account for the many licences entered into by these universities with third parties. It also states that:

The undersigned have adopted copyright policies and practices that […] give effect to fair dealing, transactional and other licences, and other forms of permitted use of materials subject to copyright […] Therefore, a range of licensing arrangements must be considered by the Board.

[280] Finally, and also as mentioned before, Brock University, the University of Manitoba, Athabasca University and Mount Royal University also wrote to the Board to express their support for the observations made in the Consortium’s and York’s letters.

ii. Access’ Reply

[281] The Board afforded Access the opportunity to reply to the universities’ letters of comment. Unsurprisingly, it objected to the Board giving any weight to the letters on the ground that the claims in the letters were unsupported by any evidence:

The comments recently received from the Universities, after the close of the hearing, amount to little more than recycled claims that relate to issues that were in play since Access Copyright filed its first tariff. Those claims are wholly unsupported by any evidence of record before the Board.

[282] Access reiterated, among other things, the difficulties it had had at the interrogatories stage in getting the universities to reply to questions, some of which specifically concerned the licences entered into with third parties, despite the Board’s repeated orders. [137] An interlocutory decision by the Board on this issue had even been the subject of an application for judicial review by AUCC before the Federal Court of Appeal, which the Court had dismissed. [138]

[283] The lack of evidence Access refers to concerns as much the licences entered into with third parties as the alleged double payments and the institutions’ ability to set up a transactional licensing regime and enforce it.

[284] In this regard, Access points out that, in Access Interim Tariff – Transactional Licence, on AUCC’s application for a transactional licence, the Board had noted the difficulties posed by a transactional licensing regime. While such a regime implies the existence of measures (audits, detailed reporting of copies, monitoring, etc.) to ensure that rights holders are properly paid, the Board concluded that the educational institutions had not established that these conditions were met; hence its refusal to implement a pay-per-use licence. [139] There being no evidence showing that the institutions’ situation has changed since this decision, Access alleges that the same outcome is inevitable: the transactional licence should be denied.

[285] Indeed, Access finds it ironic that these universities are requesting a pay-per-use licence given that most of them did not pay any royalties for all or part of the period covered by the interim tariff despite its structure being that of a transactional licence, at least in part, since copies for course packs were subject to a per-page rate. The same applies to the Choice Licence offered by Access as of 2015, which also includes a transactional component. None of the universities chose this type of licence. [140] How then can the application for a transactional licence be considered to be credible when several universities felt in 2011 that they did not make any reproductions requiring a licence from Access?

[286] Lastly, regarding the statement made in the Consortium’s letter according to which the universities had adopted fair dealing policies, Access repeats its submission that there is nothing in the record concerning the fair nature of these policies or their implementation.

iii. Analysis

[287] Section A.2 of the Board’s Directive on Procedure provides as follows:

Comments

Anyone may comment in writing on any aspect of the proceedings. As a general rule, comments received later than the date by which participants must present or file oral or written arguments, will not be considered. In due course, the Board will forward these comments to participants.

[288] As a procedural matter, we could have decided not to consider the universities’ letters of comment given that they were sent after the hearing had ended. We have chosen to consider them nonetheless, given the particular circumstances of these proceedings.

[289] Since there are no institutional objectors in this instance and because these tariffs are of general application, the letters of comment assist us in considering the public interest by permitting us to take into account the views of those who are not before us but who will be affected by our decision.

[290] The Federal Court of Appeal addressed a related issue in Netflix v. SOCAN:

Since tariffs certified by the Board are of general application, the interests that must be considered are those of an industry as opposed to those of an individual or an entity. [141]

[291] Even though this statement was made in relation to the issue of whether the parties asking the Board to approve a particular tariff represented the interests of all prospective users, it is nevertheless instructive here. We will therefore examine the arguments made by the universities that submitted requests to the Board.

[292] Access submits that the Board cannot grant the universities’ requests to certify a pay-per-use licence on the ground that the universities did not submit any evidence regarding the licences entered into with third parties. We agree.

[293] Based on the record before us, there is good reason for the Board to believe that the market for the reproduction of literary works in post-secondary institutions is probably different from what it was several years ago. Access does not disagree, and has never denied the existence of these licences, but points out that it has no way of knowing how to measure the size and conditions of these licences. [142]

[294] As we noted above, the universities chose not to participate in the proceeding. Had they chosen to participate, they would have had to file evidence in support of their arguments. In the absence of evidence, it is not possible for the Board to approve a tariff with a transactional licence, as requested by the universities, or to calculate an arbitrary reduction in royalties to address the licences entered into with third parties (as requested by the University of Toronto in 2014). [143]

[295] The lack of evidence is fatal to the universities’ comment that the Board should incorporate a transactional licence provision into the tariffs. The Board has to make a decision on the basis of the record before it. Consequently, the Board rejects the request of the above-mentioned universities to approve a tariff with a transactional licence.

J. Who Pays the Tariff?

[296] One of the important points raised by Mr. Maguire relates to the question of who pays the tariff. Mr. Maguire submitted that, according to his experience, most institutions paid the $3.38 per-FTE fee out of general revenues and charged the per-page fee to the purchasers of the course pack. Now that there is no per-page fee, the entire per-FTE fee is paid by students as part of their ancillary fees. [144] Access disputed this claim. Ms. Levy explained that some institutions paid the per-FTE fee out of library budgets or ancillary fees, and some augmented the cost of course packs to cover the per-FTE fee. [145]

[297] We acknowledge Mr. Maguire’s point that, if AUCC negotiates a Model Licence with Access, and an institution signs that model licence, ultimately charging the royalties back to the students as a special levy, AUCC is far removed from the ultimate payer of the tariff. Two points bear mention respecting this, however. First, even if AUCC is far removed, AUCC, the institutions, and the students have common objectives: to minimize costs and maximize access to works in Access’ repertoire. Incentives are thus aligned. Second, while the arrangement between the students and the institutions is somewhat different than those which prevail in other tariffs approved by the Board, it has the desirable feature that the user and the payer are often the same person.

[298] Ultimately, the choice of how to fund the tariff does not bear significantly on these proceedings, in our view. An institution with 50,000 FTEs and 2,000 faculty members could choose to charge each student the per-FTE amount of the tariff or each faculty member 25 times that amount, or however they choose to apportion the amount, if that is what they choose to do. Or, it could use funds from general revenues. Neither affects the appropriateness of the tariff.

VI. ROYALTY RATES

[299] In accordance with the preceding analysis, we fix the following royalty rates:

Table 3: Rates Fixed, per FTE per year

Table 3: Rates Fixed, per FTE per year

Educational Institution	Rates for the years 2011-2014	Rates for the years 2015-2017
Colleges	$9.54	$5.50
Universities	$24.80	$14.31

VII. TARIFF WORDING

[300] Because we set the royalty rates by using the 2012 Model Licence and 2015 (Three-Year) Premium Licence as proxies, we also use these licences as the starting point for the terms and conditions of the 2011-2014 Tariff and 2015-2017 Tariff, respectively.

[301] Notably, this includes using the grant of rights (and copying limits) of the Proxy Licences, as opposed to those set out in the Proposed Tariffs, since the proxy prices reflect these rights and limits. We have no reliable means of adjusting the proxy prices to reflect the differences between those proxies and the Proposed Tariffs. It is considerably easier to use the provisions of the Proxy Licences than to try to estimate new prices resulting from such differences.

[302] Starting with these Proxy Licences, we made several alterations to the terms and conditions (dealing primarily with retroactive effects and transitional provisions), and provided these as Draft Tariffs to the parties on February 6, 2019, in Notice 2019-007 for their comments on the feasibility and clarity of the terms of the tariffs.

[303] Following comments from the parties, the Board permitted affected persons to comment on the feasibility and clarity of the terms of the Draft Tariffs (Notice 2019-013, March 6, 2019). The parties were subsequently provided with an opportunity to respond to comments from affected persons (Notice 2019-021, April 24, 2019).

[304] We have considered the parties’ submissions, as well as the comments of the affected persons, and have amended the Tariffs accordingly. We address below some of the more substantial issues.

A. Definitions

i. “Copy” – definition

[305] The definition of “copy” for 2015-2017 in the Draft Tariffs is problematic. The term is defined as both an object (a reproduction) and an act (making available). “Copying” being defined as the making of a “copy”, therefore resolves to the convoluted “making of a making available to the public.” Later uses of the term “copy” may also be difficult to interpret because of this definition.

[306] That being said, this language is present in the 2015 Premium Licence, and no Party or affected person commented on this aspect of the Draft Tariff. It is therefore likely that despite the theoretical interpretative difficulties, it has not caused a problem in practice.

[307] As such, we do not disturb this language but caution that the Board, in a future tariff, may not approve this language as a matter of course.

ii. “Copy” – whether activities result in a copy

[308] Some affected persons submitted that some of the activities enumerated in the definition of “copy” do not result in copies. It is important to note that the definition defines “copy” as a reproduction that is made “by, as a consequence of, or for the purpose of” those activities.

[309] For example, if the “projecting an image using a computer or other device” does not result in a copy, and no copy was made for that purpose, then there is no “copy” for the purpose of the Tariffs. Furthermore, even if a reproduction is made, but does not require authorization, or has been authorized, then there will still be no “copy” for the purpose of the Tariffs. As such, we conclude that no modification is required in this respect.

iii. “Course collection”

[310] Certain affected persons submit that this term, as defined in the 2011-2014 Draft Tariff should not refer to “digital copies […] that are emailed, linked or hyperlinked.”

[311] Access does not object to this change, and we make this adjustment in the 2011-2014 Tariff.

iv. “Course of study”

[312] Access supports the submission of certain affected persons that the definition of this term be clarified to make clear that each section of the same course is a separate “course of study”. This would ensure that different sections of a course are not caught by the same copying limit (e.g., Fall section and then Winter section).

[313] We make this adjustment, referring to it as a “course section.”

v. “Repertoire work”

[314] Access submitted that the definition of “repertoire work” be modified to reflect the definition of “collective society” in the Act. In particular, it proposes the addition of the phrase “by assignment, grant of licence, appointment of it as their agent or otherwise authorize it […]”

[315] This does not appear to be a necessary change. The definition of repertoire in the Draft Tariffs is – if anything – less restricted than that being proposed by Access. Furthermore, the definition in the Draft Tariffs avoids the suggestion that Access has, in fact, obtained authorization through all of the listed methods; something on which, given the manner in which royalties in this matter are determined, we do not need to make a conclusive determination.

[316] For the same reason, we also remove the phrase “or by another collective management organization.” We agree with Mr. Katz that this is a redundancy.

vi. “Secure network”

[317] Access supports Universities Canada’s submission that the definition of “secure network” in the 2015-2017 Tariff include the qualifier that these are “operated by” or “for and subject to the control of” educational institutions.

[318] We make this adjustment and harmonize the definition for the two tariff periods.

B. Grant of Rights

[319] Mr. Katz submits that the Draft Tariffs, as formulated, appear to permit copying of works which may not be in Access’ repertoire. In particular, the Draft Tariffs appeared to authorize the “copying” of works contained in other works (such as an essay in a compilation) even though the second work is not explicitly required to be in Access’ repertoire.

[320] We agree. We include a clarification to the wording of the Tariffs to ensure that they only authorize the “copying” of works within Access’ repertoire.

C. Conditions on “copying”

[321] Access submits that the Board should include conditions that i) require that reproductions are faithful and accurate, and ii) credits are included where reasonable. Access argues that these conditions “are important to respect the moral rights of our author affiliates.”

[322] First, we observe that it is not possible for users to comply retroactively with these conditions. For all acts in the past, either these provisions have been complied with, or they have not. To the extent they have not, the inclusion of such a condition would suggest that copies made that do not meet those conditions do not benefit from the licence of the Tariffs. This would, in turn, suggest that the royalties payable should be adjusted for the decrease in the Tariffs’ effective scope.

[323] In this case, the most practical solution is not to include the condition for copies made prior to the approval of the Tariffs. We have no evidence before us on the extent to which these provisions have been met, we have not adjusted royalties on this basis, and we therefore do not include these conditions for copies made in the past.

[324] Second, we observe that the Tariffs do not impact the moral rights of the author: they remain unwaived. As such, an infringement of the moral right of an author committed while performing an act permitted by these Tariffs remains an infringement of those moral rights. Moreover, Access’ authority comes from the owner of copyright (as opposed to the author), and given that the moral right remains with the author, the appropriateness of such provisions is questionable.

[325] For these two reasons, we do not include such conditions.

D. Non-Authorization of Circumvention of a Technological Measure

[326] Access seeks the inclusion of a statement that the Tariffs do not authorize the circumvention of a technological measure. We decline to do so.

[327] The Board avoids restating the law in a tariff where it is not necessary to do so for the understanding of a tariff. [146] We agree with this approach. Since there is nothing in the Tariffs that suggests that such acts would be authorized thereby, we conclude that such a clarification is not appropriate.

E. Academic or Calendar Year

[328] Access submits that royalties should be calculated and paid on a calendar-year basis, as opposed to an academic-year basis. It argues that this would make the Tariffs “simpler and streamlined and there will be no need to adjust the rate for portions of Academic Years that fall outside the tariff periods and it will also make the interest calculations simpler.”

[329] Those affected persons that commented on this issue submitted that “paying royalties on a calendar-year basis rather than an academic year is inconsistent with the practices of academic institutions.”

[330] FTEs are calculated across an academic year. Applying a royalty rate based on FTEs for an entire calendar year would almost certainly result in overpayments or underpayments with respect to the “Fall/Winter” session. While it may be possible to introduce various adjustment mechanisms, we prefer to avoid more complex back-and-forth mechanisms. We therefore set royalties on an academic-year basis.

F. Date of Payment and Report

[331] Access submits that under the Draft Tariffs “royalties for copying (and the corresponding FTE reporting) are not due until eleven and a half months into a calendar year licensed by the tariff. […] Since the purpose of the tariff is to provide a licence to users prior to their usage, it should operate prospectively, and the royalties and FTE reporting should be due at the beginning of the calendar year.”

[332] On the other hand, educational institutions such as Simon Fraser University argue that “it would be completely impossible […] to know [the educational institution’s] approximate FTE numbers until well into July of the academic year, and final numbers would not be readily available until the Fall for the previous academic year.”

[333] Where the royalty structure requires the measurement of some quantity, such as the number of downloads sold, or the number of students enrolled, it is more practical that payment be made once the quantity on which royalties are based is known. For example, in Online Music Services Tariff (2010-2013), [147] a person who operates a permanent download service only makes payments in relation to those downloads after the end of the quarter in which those downloads occurred.

[334] While it may be possible to create a mechanism whereby educational institutions attempt to forecast their FTEs, make payments accordingly, and adjustments (additional payment or reimbursement) occur once final FTE counts are obtained, we prefer not to introduce such a back-and-forth mechanism. Instead, we fix the reporting of FTEs for an academic year, and the associated payment, for November 15 of the subsequent academic year.

G. Reporting

[335] Certain affected persons argue that reporting requirements be limited to the reporting of FTEs, as this is the only data required to determine royalties payable.

[336] It is correct that the royalties payable under these Tariffs are calculated on the basis of FTEs – and are not at all affected by the presence or absence of the additional information sought. This does not mean, however, that requiring the collection and reporting of “copying” information will automatically be inappropriate. Rather, we consider the practicality and proportionality of the collection and reporting of that information, and how necessary that information is to achieve the goal of distribution of royalties. This is in-line with previous decisions of the Board.

[337] A survey of previous decisions of the Board shows that – in general – for-profit entities whose main activities relate directly to the copyrighted subject matter at question typically have more extensive reporting requirements than non-commercial entities, or commercial entities whose main activities are not so directly related to the copyrighted subject matter. The more central the subject matter was to an entity’s activity, and the greater its capacity to carry out that reporting (either financial, technical, or otherwise) the more extensive such requirements tend to be. However, where a collective society could not demonstrate how certain usage information was to be used for distribution, the Board has declined to include reporting for that information in the approved tariff.

[338] In this case, the users in question are not for-profit, nor do their main activities relate directly to the copyrighted matter in issue in these Tariffs.

[339] Furthermore, according to the Canadian Internet Policy and Public Interest Clinic (CIPPIC), the lack of transparency as to the content of Access Copyright’s Repertoire makes compliance with such a provision difficult, if not impossible. This lack of clarity places educational institutions in a difficult position, walking a fine line between over-reporting and under-reporting, without knowing where the line might be.

[340] It is true that there is no definitive, readily-available source that would provide users with a list of works in Access’ repertoire, as that term has been described in this decision as well as in Access – Governments and Access – K-12 (2010-2015).

[341] We are also aware that lack of this information may reduce Access’ ability to distribute royalties to the appropriate rights holders. However, there are two mitigating factors to this. First, Access does not base its distributions solely on the copying information received from such institutions. Second, to the extent such reporting requirements are feasible, Access will have received this class of information from educational institutions with whom it has entered into a licensing agreement. As such, there would be readily-available sources from which to proxy the copying behaviour of educational institutions subject to this Tariff.

[342] We do not include a requirement that “educational institutions” report bibliographic and related information for “course collections” for the following reasons:

• the royalty structure of the Tariffs does not require the reporting of “copying” to determine royalties;

• “educational institutions” do not have a readily-available and reliable manner of determining to which “copies” reporting requirements would apply;

• Access Copyright does not distribute royalties exclusively on the basis of copying reports; and

• to the extent the reporting provisions in 2012 Model Licence and 2015 (Three-Year) Premium Licence are actually being complied with by those “educational institutions” signatories to such licences, Access Copyright will have information from which to proxy the copying behaviour of “educational institutions” subject to these Tariffs.

H. Audit

[343] Some affected persons submitted that the 5 business-days’ notice would be overly burdensome, and may not even be feasible. Access submitted that it is willing to accept a 10-day notice period. We adjust the Tariffs accordingly.

I. Subcontracting and Other Third Parties

[344] Access submits that

[u]sers may have in fact relied on the proposed tariff to authorize the subcontracting of the copying during the tariff time periods. If the Board removes these uses from the tariffs now for copying that occurred before certification, that would mean that all the copying subcontracted by Educational Institutions who were relying on the proposed tariffs, would turn into infringing uses as of certification. This would be an unfortunate result.

[345] Universities Canada agrees that for the 2011-2014 Tariff, a subcontracting provision should be added, and should also apply to copies made prior to approval.

[346] Access agreed with this proposition, and put forward the following provision:

In the event a Subcontractor has paid royalties to Access Copyright for paper course packs on behalf of an Educational Institution, the Educational Institution is entitled to a credit for the amount of royalties paid by the Subcontractor to Access Copyright against any royalties to be paid by the Educational Institution under the tariff provided that Access Copyright is provided with a copy of the subcontracting agreement as well as full reporting and accounting to assess the credit.

[347] We generally agree with Access’ proposal. We add a provision permitting subcontracting, and require that the agreement and accounting or other information that demonstrates the amount of royalties the subcontractor has previously paid be submitted in order to obtain credit.

J. Works in Access’ Repertoire

[348] Many affected persons indicated that they do not have means of determining whether a work is in Access’ repertoire, and the scope of authorization granted by the rights holder with respect to a work.

[349] The Canadian Association of Research Libraries (CARL), for example, asked that the Board

include a provision in the tariffs mandating that Access Copyright provide a publicly accessible list of its affiliate copyright owners, including essential bibliographic information concerning the author and publisher and ISBN or ISSN identifier of the works it purports to licence. This list should include any distinctions between its ability to authorize paper and digital copies.

[350] We are also aware of the June 2019 Report of the Standing Committee on Industry, Science and Technology on the Statutory Review of the Copyright Act, wherein it recommended that

the Copyright Board of Canada review whether provisions of the Copyright Act empower the Board to increase the transparency of collective rights management to the benefit of rights-holders and users through the tariff-setting process […]

[351] In this context, we acknowledge the challenges educational institutions face in determining whether Access has been authorized to authorize a particular act with respect to a given work. However, the imposition of this, or a similar requirement, would be relatively novel in a tariff approved by the Board. We would prefer that the practicality of such a requirement, along with any benefits and disadvantages thereof, be more fully addressed before establishing such a requirement.

K. Payment and Interest

[352] While some affected persons submitted that there should be no interest payable on amounts owing, we do not agree. The interest rates we fix simply reflect the time-value of money. This is a common Board practice, and ensures that the person paid is (approximately) as well off as if she had been paid at the time the amount was initially due.

[353] Any amount owing, either by or to educational institutions, shall bear interest.

[354] Furthermore, we ensure that interest does not accrue on unpaid federal or provincial taxes. We have no evidence to suggest that Access Copyright has already remitted these taxes on behalf of the educational institutions subject to these tariffs – particularly since it does not yet know the amounts due.

L. Other Issues

[355] On April 24, 2019, the Board asked the Parties to reply to the comments received from affected persons in response to Notice 2019-007 (Notice 2019-021). The Board noted that

[g]iven the breadth of issues raised by the comments, the Board […] is of the view that many issues go beyond “feasibility and clarity.” The Board does not intend to rely on those comments and does not expect the Parties to respond thereto. This includes comments on issues of the mandatory nature of the tariffs, the retroactive application of the tariffs, the structure of the royalties (i.e. FTE vs transactional) and the scope of acts covered by the tariffs. It is also not the case that these issues are new: they were either explicitly made an issue by the proposed tariffs, or their possibility should have been reasonably contemplated (e.g., that a certified tariff may have retroactive effect).

[356] While we are of the view that the three frequently-raised issues we have identified below are beyond the scope of the information sought from affected persons, we nevertheless address them for the benefit of the affected persons.

i. The mandatory nature of the tariff

[357] The Tariffs are silent on whether compliance with a tariff is mandatory for users who do not seek to benefit from the licence offered thereby. We are aware that related issues have been raised in recent judicial proceedings [148] and it is not necessary for us to opine on the issue at this point.

[358] To the extent it might be appropriate for a tariff to include wording whereby its benefits and obligations would only apply on an opt-in basis, we would appreciate a more complete record before including such a provision, and invite affected persons to participate in the proceedings on the next occasion the Board considers proposed tariffs for these users.

ii. Royalty rates that reflect non-tariffable uses

[359] Some affected persons ask the Board to fix rates that take into account uses that do not require the authorization of the Tariffs. As discussed above, the rates we set, being based on a market proxy, should incorporate the considerations to which affected parties make reference – the fact that certain acts may be otherwise licensed, or that certain acts do not require the authorization of the owner of copyright – as the parties to those market proxies understood them at the time of agreement.

iii. Licensing fees paid during the tariff period for copying nominally covered by the Tariffs

[360] In Notice 2019-021, the Board sought responses from parties on the issue of previous payments made for the licensing of material during the tariff period. The Board stated that it

is of the preliminary view that it does not have the evidence in this proceeding necessary to establish a mechanism in the tariffs that would adjust royalties based on other licensing activities of the user.

[361] Access agreed with the Board’s preliminary view and Mr. Maguire did not respond to the Notice.

[362] We therefore conclude that we do not have the evidence in this proceeding necessary to establish a mechanism in the tariffs that would adjust royalties based on other licensing activities of the user. We hope to have more and better evidence the next time on this question.

[363] We wish to thank Access’ counsel for the clear and well-presented submissions on the administrative provisions of the Tariffs. We also wish to thank Mr. Maguire for his contribution to this proceeding.

Lara Taylor

Reasons certified by the Secretary General

APPENDIX / ANNEXE

List of Objectors and Intervenors / Liste des opposants et des intervenants

Objectors / Opposants

• Association of Canadian Community Colleges / Association des collèges communautaires du Canada
• Association of Universities and Colleges of Canada / Association des universités et des collèges du Canada
• Athabasca University / Université d’Athabasca
• British Columbia Association of Institutes and Universities

Intervenors / Intervenants

• Ambrose University / Université Ambrose
• Ariel Katz
• Canadian Alliance of Student Associations / Alliance canadienne des associations étudiantes Canadian Association of University Teachers / Association canadienne des professeures et professeurs d’université
• Canadian Federation of Students / Fédération canadienne des étudiantes et étudiants Government of Alberta / Gouvernement de l’Alberta
• Jay Rahn
• Mark McCutcheon
• Meera Nair
• Nancy Pardoe
• Sean Hunt
• Sean Maguire
• St-Mary’s University College / Collège universitaire St-Mary’s
• Yellowhead Tribal College / Collège tribal de Yellowhead