| Copyright Board |
|
Commission du droit d’auteur |
| Date
|
2015-05-22
|
| Citation
|
File: Reproduction of literary works, 2005-2014
|
| Regime
|
Collective Administration in Relation to Rights Under Sections 3, 15, 18 and 21
Copyright Act, subsection 70.15(1)
|
| Members
|
The Honourable William J. Vancise
Mr. Claude Majeau
Mr. J. Nelson Landry
|
| Proposed Tariffs Considered
|
[Provincial and Territorial Governments – 2005-2014]
|
Statement of Royalties to be collected by Access Copyright for the Reprographic Reproduction, in Canada, of Works in its Repertoire
Reasons for decision
III. LICENCES WITH GOVERNMENTS IN CANADA
E. Government of British Columbia
IV. POSITION OF THE PARTIES AND PROPOSED RATES
B. Results of the Study as analyzed by Access
C. Results of the Study as analyzed by the Consortium
D. Compensability re-analysis and copying event data
E. Second calculation of per-FTE volume
VII. INTERPRETING THE VOLUME STUDY DATA
B. Page-count of the copied work
C. Other issues with non-affiliate copyright owners
B. The effect of non-inclusion of the Deletion Provision
X. ACTS OUTSIDE OF THE SCOPE OF THE TARIFF
A. Amount of work copied greater than permitted by the Tariff
B. Where reproduction is not a substantial part of the copied work
ii. Quantitative evidence and burden of proof
iii. The work to be considered
XI. ACCESS TO INFORMATION ACT AND PRIVACY ACT EXCEPTIONS
i. Purpose (Step 1) - Who is the user? Which purposes should be considered?
ii. Interpreting the scope of purposes
iii. Purpose (Step 1) - Is the dominant purpose a permitted purpose?
D. Fair dealing – Step 2: Fairness
a. Goal - The fairness of various goals
c. Goal - Ulterior motives of the governments
d. Goal of the dealing - Conclusion
a. Character of the dealing - Width of distribution
b. Character of the dealing - Destruction of the copies
c. Character of the dealing - Conclusion
a. Amount used - How to measure?
b. Amount used - Does the context affect the analysis?
c. Amount used - Should and can the amount used be evaluated qualitatively?
d. Amount of the dealing - Conclusion
iv. Alternatives to the dealing
a. Would have copied the work anyway
b. Purchase one or more copies of the work
c. Acquire a one-time right to make a copy of the work
d. Made do without copying the work
e. Refer others to the work without sending a copy
f. Use part of a different work that was not protected by copyright
vi. Effect of the dealing on the work
B. Use of works in legal proceedings
A. Two different approaches to determining royalty rates
i. Introduction of the two approaches
v. VTV: Should we use the data from Phase I or Phase II?
i. Whose calculations are more adequate
D. Calculating the royalty rates
E. Total amount of royalties generated from the Tariff
ii. Interest on retroactive payments
A. Section 2 – Definitions – “Copy” – Printing from an electronic file
i. Whether printing from an electronic file is a form of reprography
ii. Whether printing from an electronic file is “reproducing by a machine that makes a Digital Copy”
B. Section 2 – Definitions – “Copy”
C. Section 2 – Definitions – “FTE”
D. Section 2 – Definitions – “Published Work”
F. Paragraph 4(h) – Moral rights
I. Section 6 – Notification of the terms and conditions of copying
K. Section 11 – Records and audits
I.
INTRODUCTION
[1] The Canadian Copyright Licensing Agency, operating as Access Copyright (Access), is a collective society that administers the reproduction right in published books, magazines, journals, and newspapers, for all of Canada except Quebec.
[2] On March 31, 2004 and March 31, 2009, Access filed, pursuant to subsection 70.13(1) of the Copyright Act
[1]
(the “Act”), proposed statements of royalties to be collected for the reproduction in Canada (excluding Quebec) of works in its repertoire by employees of provincial and territorial governments for the years 2005 to 2009 (“2005 Proposed Tariff”)
[2]
and 2010 to 2014 (“2010 Proposed Tariff”),
[3]
respectively. The 2005 Proposed Tariff targets paper copies; the 2010 Proposed Tariff targets both paper and digital copies.
[3] The proposed statements of royalties were published in the Canada Gazette on April 24, 2004 and May 9, 2009, with a notice informing potential users or their representatives of their right to object to the statements. These tariffs are inaugural.
[4] The governments of British Columbia (BC), [4] Ontario, Alberta, Saskatchewan, Manitoba, Yukon, Nunavut, Nova Scotia, Prince Edward Island and Newfoundland and Labrador filed timely objections to the 2005 Proposed Tariff. The same governments objected to the 2010 Proposed Tariff, as well as the governments of the Northwest Territories and New Brunswick. On March 1, 2011, after it concluded a licence with Access, the Government of Ontario withdrew from the proceedings. On March 31, 2011, the Government of the Northwest Territories also withdrew from the proceedings.
[5] The governments of Alberta, Saskatchewan, Manitoba, Yukon, Nunavut, Nova Scotia, Prince Edward Island, New Brunswick and Newfoundland and Labrador elected to be jointly represented. They are referred to as the Consortium. The Government of BC elected to be represented on its own. The Government of BC and the Consortium are, collectively, the Objectors.
[6] The Board consolidated the examination of both proposed tariffs. The hearing began on October 23, 2012 and lasted 8 days, including final arguments. The record was not perfected until August 28, 2014 by reason that the Board required Access and the Objectors (together, the “Parties”) to provide further information and perform additional analysis and calculations. A detailed description of the demands for further particulars is set out in part II.B, below.
II.
THE PROCESS
A.
Pre-hearing process
i.
Crown immunity
[7] On January 15, 2010, the issue of whether the Act was binding on the Crown in right of the provinces was raised by some of the Consortium members as a preliminary matter. The matter was a direct challenge on the Board’s jurisdiction to certify a tariff for the reproduction of works by employees of provincial and territorial governments. The preliminary issue on the Board’s jurisdiction was heard on September 27, 2011. On January 5, 2012, the Board dismissed the claim for Crown immunity.
[4]
[8] The decision of the Board was upheld by the Federal Court of Appeal on April 3, 2013.
[5]
ii.
The Volume Study
[9] In 2010, Access and the Objectors agreed to conduct a study of the volume and nature of published works reproduced by government employees (“Volume Study” or “Study”), the details of which are set out in part VI below. On December 17, 2010, the Board ordered that a pre-test be carried out in order to assess the overall reliability of the Study. The pre-test took place between May 6, 2011 and June 20, 2011 and, as discussed below, was considered satisfactory to the Parties.
[10] The Volume Study was conducted in two phases, Phase I and Phase II, which were completed by December 2011.
B.
Post-hearing process
[11] On December 4, 2012, the Board issued two notices. The first asked the Parties to provide a legal brief to address issues related to exceptions to copyright, and to comment on the Board’s statements regarding Access’ repertoire. The second notice asked the Parties to make various calculations, submissions on value, and to clarify certain statements made by Access in previous submissions.
[12] The responses to the second notice were completed by February 1, 2013, and those to the first notice by February 22, 2013.
[13] The Parties had completed making their submissions regarding administrative provisions by April 26, 2013. All were of the view that a single tariff could be certified for the entire 2005-2014 period (the “Tariff”).
[14] On March 14, 2013, the Board issued a notice in which it asked the Parties to comment on calculations made by the Board’s staff with respect to the appropriateness of using certain data from the Volume Study. It also expressed a preliminary view on the events captured in the Volume Study that should be used in the calculation of a royalty rate, and directed the Parties to submit additional data relating to those events. Those submissions were received by April 30, 2013.
[15] On May 6, 2013, the Board ordered Access and the Objectors to prepare data samples and to analyze the data from these samples with respect to compensability and repertoire.
[16] On May 27, 2013, the Consortium wrote to the Board informing it that there remained a discrepancy of four copying events between the sample prepared by it and by Access.
[17] During the time this discrepancy was being resolved, the Board suspended the timetable for the analysis of the sample. Subsequently, the Parties filed an agreement in which they agreed that the sample consisted of 291 copying events. The analyses were completed by August 6, 2013.
[18] On May 6, 2014, the Board ordered the Parties to make submissions on the effect of the requirement in the 2010 Proposed Tariff that digital copies be destroyed once a licensee is no longer covered by a tariff. Those responses were received by June 13, 2014.
[19] On July 21, 2014, the Board issued a notice in which it set out its preliminary conclusions as to which events from the Volume Study were compensable for the purposes of determining a royalty rate for the Tariff and directed the Parties to calculate the average number of compensable copies made by each government employee in a year. The Board received the responses by 28 August, 2014.
III.
LICENCES WITH GOVERNMENTS IN CANADA
[20] Access and the Government of Canada, and some provincial and territorial governments of Canada, have previously entered into licensing agreements. Some of these agreements form the basis of Access’ proposal and were the subject of intense submissions in the evidence and at the hearing. It is therefore necessary to provide an overview of these licensing agreements.
[6]
A.
Government of Canada
[21] Access
[7]
and Copibec
[8]
, the collective society that administers, in Quebec, the reproduction right in published works, entered into a joint licensing agreement with the Government of Canada in May of 1994 (“Initial Canada Licence”
[9]
). Although effective as of April 1, 1994 until March 31, 2001, the licence provided for the compensability of copies carried out from April 1, 1991 to March 31, 1994.
[22] The royalty rate for the Initial Canada Licence was based on the results of a joint study that was conducted between June of 1992 and May of 1993 (“Goss Gilroy Study”
[10]
). The Goss Gilroy Study concluded that federal government employees made 57.04 million pages of copies (or “exposures”) of licensable materials annually. The parties agreed to discount the estimated volume by 20 per cent (to 45.63 million exposures) to account for the assumed repertoire coverage of the collectives at the time (80 per cent). That volume was then multiplied by a blended page rate of $0.03 to arrive at an annual royalty of $1,368,960.
[23] The parties also signed a separate agreement covering press clippings, which was effective from April 1, 1994 to March 31, 2001 (“Press Clipping Licence”
[11]
). Under this licence, the Government of Canada made an annual royalty payment of $500,000. Thus, as of April 1, 1994 the Government of Canada was paying Access and Copibec $1,868,960 annually for the Initial Canada Licence and the Press Clipping Licence.
[24] In January of 2000, the parties signed an amending agreement,
[12]
effective from April 1, 1999 to March 31, 2001, which consolidated the Press Clipping Licence and the Initial Canada Licence into one agreement. There was no change to the combined annual royalty rate. This amending agreement was extended repeatedly until a new licence was signed on December 22, 2005 (“2004 Canada Licence”
[13]
).
[25] During the negotiations of this agreement, the parties agreed to conduct a joint study to measure the volume of photocopying by federal government employees. The study (“2003 Canada Study”
[14]
), conducted by Circum, took place during the course of 2003 and 2004.
[26] The 2003 Canada Study concluded that the Government of Canada employees made approximately 54 million copies of published works each year (not including press clippings for the purpose of media monitoring). The parties disagreed on the interpretation of the study results as well as on quantification issues. The most contentious issue was the appropriate discount to be applied for fair dealing. At the time, the Supreme Court of Canada had just released its decision in CCH Canadian v. Law Society of Upper Canada
[15]
(CCH) and the Government of Canada argued that a 50 per cent discount to the volume should be applied as a result of that decision. Access argued that a 6 per cent discount to the volume should be applied for fair dealing.
[27] The parties eventually settled on an annual royalty of $2,500,000.
[28] The 2004 Canada Licence covered the period from April 1, 2004 to March 31, 2013. The annual royalty rate payable under the joint licence was at $2,500,000 for 2004, and increased at a rate of 2 per cent annually. At the time of the hearing, the rate for 2012-2013 was thus $2,929,148.44.
[16]
The licence only covered paper reproductions. The 2004 Canada Licence included an indemnity provision pursuant to which Access and Copibec would indemnify the licensee for any claim arising from the copying of published works under the terms of the licence, irrespective of whether the published work was within the collectives’ repertoire.
B.
Government of Ontario
[29] On June 16, 1995, Access and the Government of Ontario entered into a licensing agreement,
[17]
effective from April 1, 1995 to March 31, 1998 at a rate of $2.50 per full-time equivalent employee (FTE). The licence covered reprographic reproductions of published works, excluding newspapers.
[30] On September 18, 1995, this licence was amended to include departmental newspaper copying and media-monitoring uses, and the royalty was increased to $3.12 per FTE.
[18]
[31] The parties agreed to renew the licence at the rate of $3.12 (“1998 Ontario Licence”
[19]
). The 1998 Ontario Licence came into force on April 1, 1998 and extended until March 31, 2001. The licence was renewed annually from March 31, 2001 to March 31, 2010 at the same rate of $3.12 per FTE.
[32] The parties again entered into a licensing agreement on February 10, 2011 (“2010 Ontario Licence”
[20]
) for the period April 1, 2010 to March 31, 2015.
[33] The licence provided for a rate of $7.50 per FTE for 2010. For subsequent years, the rates increase yearly according to the Consumer Price Index. The licence contained an indemnity provision pursuant to which Access would indemnify the Government of Ontario for liabilities it may incur by copying works as authorized under the licence.
[34] This licence was the only one in force between Access and the government of a province or territory at the time of the hearing.
C.
Government of Alberta
[35] In 1996, Access and the Government of Alberta signed an initial licence,
[21]
effective from October 1, 1995 until March 31, 1999 at a rate of $2.80 per FTE. The rate was a negotiated rate; no survey on the volume of copying was conducted. This licence covered reprographic copying of published works, including departmental copying of newspapers.
[36] In 2000, a new licence (“1999 Alberta Licence”
[22]
) was signed to cover the period April 1, 1999 to March 31, 2002 at the rate of $2.80 per FTE. No survey had been conducted.
[37] The 1999 Alberta Licence was renewed annually at a rate of $2.80 per FTE until March 31, 2010.
D.
Government of Saskatchewan
[38] In 1998, Access and the Government of Saskatchewan entered into a licence agreement at a rate of $2.80 per FTE (“Saskatchewan Licence”
[23]
). No survey had been conducted.
[39] The licence was effective from April 1, 1998 until March 31, 2000, and was automatically renewed annually, remaining in force until December 16, 2009.
E.
Government of British Columbia
[40] In 2003, Access and the Government of BC jointly designed a copying survey methodology. The survey was conducted by an independent contractor in 2003.
[41] A draft report analyzing the survey results was written by BC Stats.
[24]
However, the report was never finalized because Access and the Government of BC were unable to agree on the analysis of the results and corresponding FTE rate. BC Stats estimated that the applicable FTE rate should be $3.17, whereas Access believed that the applicable FTE rate should range between $8.46 and $19.95. The Government of BC offered to enter into an interim two-year licence for $2.80 per FTE, as that was the rate paid by Alberta and Saskatchewan at the time. Access rejected the proposal and no licence was entered into.
[42] BC also conducted another survey, on its own, in 2006; however, no further licensing agreement with Access was entered into.
F.
Government of Quebec
[43] The first licensing agreement that was entered into between Copibec and the Government of Quebec covered the period from 1996 to 2001. The term of that licence was subsequently extended until June 30, 2003.
[25]
[44] A further agreement covered the period from July 1, 2003 to March 31, 2008 (“2003 Quebec Licence”
[26]
). The rate ranged from $8.80 per FTE in 2003 to $12.80 per FTE in 2007. This licence was extended until October 31, 2008.
[27]
[45] A further licence covered the period November 1, 2008 to March 31, 2011 (“2008 Quebec Licence”
[28]
). The rate ranged from $14 per FTE in 2008-2009 to $14.25 per FTE in 2009-2011. The parties agreed that the firm L’Observateur would conduct a study to estimate the volume of copies made by employees of the Government of Quebec.
[29]
[46] Both the 2003 and 2008 Quebec licences provide for additional royalties to be paid by the Ministère de l’Éducation, du Loisir et du Sport (MELS) for ministry exams.
[47] In 2011, Copibec and the Government of Quebec entered into a licensing agreement for the period May 3, 2011 to March 31, 2014 (“2011 Quebec Licence”
[30]
). (For the period April 1, 2011 to May 2, 2011, the preceding licence was extended) The licence provided for a rate of $14.55 per FTE for the period April 1, 2011 to March 31, 2012, a rate of $14.85 for the period April 1, 2012 to March 31, 2013 and a rate of $15.15 for the remaining term of the licence. Digital copies were not allowed under the licence, with the exception of copies made by the MELS for use in ministry exams, for which additional royalties were payable.
[31]
The licence provided the Government of Quebec with an indemnity provision covering the copying of works that are not on Copibec’s exclusions list.
IV.
POSITION OF THE PARTIES AND PROPOSED RATES
A.
Access
[48] Access argues that the data from Phase I of the Volume Study are more reliable than the data from Phase II. However, even though Phase I data are more reliable, Access contends that they nevertheless suffer from many limitations and rely on too many assumptions; as such, they are not reliable enough to be used in a volume times value calculation.
[49] Access instead proposes to establish a fair market value by examining the royalty rates determined by actual arm’s-length negotiations for the licensing of similar rights to those of the proposed tariffs, a “market comparable” approach.
[50] Such an approach is based on agreements entered into by either Access or Copibec with the provinces or the Government of Canada (“Benchmark Licences”). In Access’ opinion, the market comparable approach is the most appropriate method to estimate the fair market value of the rates under the proposed tariffs because rates agreed upon under arm’s-length negotiations provide the best information available.
[51] However, in applying the market comparable approach, Access contends that some of the agreements into which Access entered reflect values below fair market value, and should therefore not be considered in this analysis. It argues that some of these agreements were negotiated without the benefit of information on copying in governments, as governments were not interested in carrying out a survey, and that Access may have made some of those agreements while under a compulsion to act. Access argues that such agreements cannot be representative of “fair market value.”
[52] For the 2005 Proposed Tariff, Access initially proposed a rate of $15 per FTE. For the 2010 Proposed Tariff, it proposed a rate of $24 per FTE. While the 2005 Proposed Tariff covers the making of copies from paper sources, the 2010 Proposed Tariff also covers the making of copies from digital sources.
[53] In its statement of case, Access reduced the rates it sought. It proposed that the rate for the 2005 Proposed Tariff be $10.10 per FTE, with annual adjustments for inflation, or $10.50 per FTE if fixed for the five-year term of the tariff. For the 2010 Proposed Tariff, it proposed a rate of $11.30 per FTE, with annual adjustments for inflation, or $11.70 per FTE if fixed for the five-year term of the tariff.
[32]
According to Access, of the Benchmark Licences that were in force during the 2005-2009 period, the best proxy is the 2003 Quebec Licence. This is the licence on which the proposed rates in the statement of case are based.
[33]
[54] Alternatively, if the Board were to conclude that the 2003 Quebec Licence is not indicative of a fair market value royalty, then the royalties should be based on the rates provided for in the 2004 Canada Licence, namely $8.40 per FTE, after adjustment. Finally, if the Board were to conclude that a rate cannot be derived from the 2004 Canada Licence, then the Board should set the rate at $5.56, which is the mid-point between the 2004 Canada Licence and the rate provided for in the 1999 Alberta Licence and the Saskatchewan Licence.
[34]
[55] Similarly, for the 2010-2014 portion of the Tariff, Access is of the opinion that the best proxy is again the 2003 Quebec Licence. Alternatively, if the Board were to conclude that this licence is not indicative of a fair market value royalty, then the Board should set a rate based on the 2004 Canada Licence, which was in 2010 at an equivalent of $9.54 per FTE.
[35]
Finally, if the Board were to conclude that a rate cannot be derived from the 2004 Canada Licence, then the Board should set the rate at the mid-point between the 2004 Canada Licence and the 2010 Ontario Licence, at the rateof $8.45 per FTE for the five-year term.
[36]
B.
The Objectors
[56] The Objectors submit that the Board should adopt the same volume-times-value methodology that the Board used in its 2009 decision
[37]
relating to the reproduction of published works in Access’ repertoire by elementary and secondary schools outside Quebec. In that decision, the 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 (“volume-times-value methodology”).
[57] The Objectors disagree with Access’ premise that the Board cannot adopt this methodology for lack of reliable data. On the contrary, the Volume Study, which Access, the Consortium, and the Government of BC jointly designed and implemented, provides ample and reliable data upon which to rely in applying an empirical, volume-based methodology. They argue that there is no reason why the Board should abandon the methodology it adopted in 2009; there should be tariff-to-tariff consistency in terms of how Access’ tariffs are established by the Board, whether the users of its repertoire are elementary and secondary schools or provincial and territorial governments.
[58] Furthermore, the Consortium submits that the data from Phase II of the Volume Study are reliable.
[59] The Objectors contend that the Study indicates that there is in fact a very limited amount of compensable copying taking place in provincial and territorial governments. Using the estimated volume of compensable copies derived from the Volume Study, the FTE rate should range between $0.01 (Phase II results) and $0.22 (Phase I results) for the 2005-2009 Tariff period, and $0.07 (Phase II results) and $0.73 (Phase I results) for the 2010-2014 Tariff period.
[38]
V.
EVIDENCE
A.
Access
[60] Ms. Maureen Cavan, Executive Director of Access at the time of the hearing, and Ms. Roanie Levy, then General Counsel and Director, Policy and External Affairs, described the collective society’s organizational structure, mandate and activities; they explained how Access acquires its repertoire and the principles governing royalty collection and distribution.
[61] Ms. Cavan and Ms. Levy also provided a description of the terms and conditions of the past and current licences in place in Canada in the government sector. They described the history of negotiations with the provincial, territorial, and federal governments from 1993 up to the filing of the tariffs and Access’ attempts to negotiate licences with provincial governments.
[62] They explained that until the proposed tariffs were filed, and except for the Government of BC, the Objectors had refused to agree to a survey that would have provided information on the nature and volume of copying in governments. The only study that had been used to set a rate in Canada was the Goss Gilroy Study conducted within the Government of Canada in 1993. Any other surveys that have been conducted in the government sector (2003 BC Study, 2003 Canada Study and the 2007 Quebec Study) have not been used to establish a licence rate. Rather, because the parties were never able to agree on the results of those surveys, they simply acted as a background for negotiations.
[63] Finally, Ms. Cavan and Ms. Levy described the repertoire analysis carried out by Access on the Volume Study data to determine what copies were subject to compensation under the proposed tariffs.
[64] Mr. Benoît Gauthier, President of Circum Network Inc., was a member of the joint steering committee that designed and implemented the Volume Study with respect to the copying carried out by the Objectors. The steering committee was comprised of two representatives of the Objectors and three representatives of Access, including Mr. Gauthier. Mr. Gauthier described what was contained in Phase I and Phase II of the Volume Study. He explained why, in his opinion, the Phase I data is the most reliable and why he therefore based his analysis on the Phase I data only. He then described how the data were processed, and presented the results of the Study. As instructed by Access, he applied three different fair-dealing rules to account for copies that ought to be excluded from the volume of compensable copies because they appeared to be fair. Depending on which rule was applied, he estimated the volume of compensable exposures per year to be 283.7, 250.8, or 237.1 per FTE.
[39]
[65] Mr. Bradley A. Heys, Vice-President of NERA Economic Consulting, was asked by Access to provide an estimate of the value of the royalty rate relevant to each of the proposed tariffs.
[66] Mr. Heys explained that Benchmark Licences provided the best methodology for valuing the proposed tariffs. He took into consideration the particular circumstances that surrounded the negotiations of those licences and concluded that the 2003 and the 2008 Quebec licences best represented the fair market value for the royalty rates to be certified.
[40]
As such, he proposed that the rate for the 2005-2009 Tariff period as of January 1, 2005 be set at $10.10 per FTE, with annual adjustments for inflation, or $10.50 per FTE if fixed for the five-year term. For the 2010-2014 Tariff period, he proposed that the rate be set as of January 1, 2010 at $11.30 per FTE, with annual adjustments for inflation, or $11.70 per FTE if fixed for the five-year term.
[41]
[67] In Mr. Heys’ opinion, in order to use the Objectors’ proposed volume-times-value methodology, the Board would have to make a number of findings on issues on which the Parties do not agree, such as the interpretation of the evidence and the impact of fair dealing. In his opinion, the Benchmark Licences approach is the least likely to result in error because it requires no assumptions about volume, no adjustments to determine page value and because it relies on the opinion of the Parties to those licences taking into account the value of the fair-dealing exceptions.
[68] Finally, Dr. Michael Murphy, of Michael J. Murphy Consulting, described the number of copies that are created when documents are posted on an intranet and when they are communicated via e-mail. In his opinion, posting documents online or sending them via e-mail results in the creation of much more than one copy, and is likely to create as many as seven to ten copies of the work.
B.
Consortium
[69] Mr. Stephen McGrath is team leader at the litigation section of the Nova Scotia Department of Justice. He sits on the steering committee established by the Consortium members to oversee their common opposition to the proposed tariffs and their joint participation in the hearing. He described the membership of the Consortium, composed of seven provincial and two territorial governments. He explained how these governments cooperated in reaching a common position with respect to Access’ proposed tariffs through the filing of a single statement of case, as well as the production of joint evidence.
[70] Dr. Paul C. Whitehead and Dr. Piotr Wilk, both professors at the University of Western Ontario, described their participation in the design and implementation of the joint Volume Study of copying in provincial and territorial governments. They explained why, unlike Access’ expert Mr. Gauthier, they consider that Phase II data is reliable and moreover, that it provides information of a higher quality than that of Phase I.
[71] Dr. Whitehead and Dr. Wilk also described how their analysis led them to conclude that government employees overall engage in very little compensable copying. They explained that the differences between their estimates of compensable copies and that of Circum’s were a function of the fact that they used both Phase I and Phase II data, the manner in which they analyzed the data, the assumptions that they made in carrying out those analyses, and the adjustments they performed. Those adjustments included the way to account for documents posted on an intranet, fair dealing, insubstantial copying, and users’ rights.
[72] Finally, they explained that their analysis indicated that Phase I and Phase II data, respectively, resulted in estimates that government employees make, on average, 11.80 and 1.15, compensable exposures per year. In their view, these are the numbers that should be used to establish the annual FTE rates.
[42]
[73] Mr. Sean Smith, Senior Manager, Valuation Services and Dr. Muris Dujsic, Partner, National Transfer Pricing Group, at Deloitte & Touche LLP presented a critique of the market comparable approach put forward by Access’ expert Bradley Heys. They explained why the Board should reject Mr. Heys’ approach and instead set the tariff based on the volume-times-value methodology used, according to them, by the Board in its K-12 decision. They argue that such an approach bears many advantages: it is derived from data and it specifically considers the two primary factors which influence the value of the licence from the perspective of both the licensor and licensee, namely the volume of compensable copying and the per-page value of such copying.
[74] Finally, Mr. Smith and Dr. Dujsic submit that the approach avoids any issues regarding the relative comparability of licences negotiated between Access and/or Copibec and different levels of government; in their view, these licences are unrelated to any of the Consortium members and, as such, are bad comparators. Moreover, the number and extent of the adjustments that would be necessary to develop any sort of concordance between the terms of these agreements and the rights covered by the proposed tariffs are so great as to render them completely unreliable for the purposes of this proceeding.
[75] Mr. Dustin Chodorowicz, a partner with Nordicity Group, was asked to estimate the fair market value of the rates to be certified for the 2005-2009 and 2010-2014 periods. He explained that the best way for the Board to establish the FTE rates in this proceeding was to apply the same methodology that was used in the K-12 decision, subject to some adjustments. He also described the per-page value that should be given to copying a page from a book, a newspaper, a journal or a magazine article.
[76] Based on the estimated volume of compensable copies made by provincial and territorial governments, his own assessed per-page valuation rates for each of the four genres of reproduced works and the total number of FTEs, he concluded that the annual FTE rate should range from $0.01 to $0.22 for the 2005-2009 period, and from $0.07 to $0.73 for the 2010-2014 period.
[43]
[77] Finally, Mr. Chodorowicz described what he perceives as flaws in the market comparable approach put forward by Mr. Heys and the fact that such an approach is based on neither an estimated volume of compensable copies nor the genre composition of copying by government employees.
C.
British Columbia
[78] The Government of BC relied on the testimonies and reports of the Consortium’s expert witnesses, namely Drs. Whitehead, Wilk, and Dujsic, and Messrs. Smith and Chodorowicz. In addition, BC called two witnesses, Ms. Victoria Lester and Mr. Don McRae.
[79] Ms. Victoria Lester is Director of the Intellectual Property Program at the Government of BC. She explained the role of this program. She provided the Board with the history of the province’s negotiations with Access and described the volume studies of photocopying by government employees which were conducted in 2003, jointly with Access, and in 2006, by BC alone. She also explained that, as BC and Access were not able to agree on the amount of copying in the Government and were thus unable to negotiate a licence agreement, the Government of BC took steps to reduce unauthorized copying within the government. Ms. Lester described what those steps were.
[80] Finally, Ms. Lester set out the Government of BC’s proposed changes to the proposed tariffs.
[81] Mr. Don McRae is a former Executive Director at BC Stats, the organization that managed the 2003 and 2006 BC studies on behalf of the Government of BC. Mr. McRae provided the Board with the background of the volume study conducted in 2003. He also explained what he considered were errors in the analysis of the results of that study by Access’ survey expert. Finally, he explained the differences in the approach taken in the 2006 BC Study with a view to correct previous methodological errors and to account for the decision of the Supreme Court of Canada in CCH.
VI.
THE VOLUME STUDY
A.
Design issues
[82] As discussed above, Access and the Objectors agreed to conduct a study relating to the volume and nature of published works reproduced by government employees. A memorandum of understanding (MOU) was entered into by the Parties and a steering committee was established to design and implement the survey.
[44]
[83] The steering committee was comprised of two representatives of the Objectors and three of Access. The MOU provided, among other things, that the Parties agreed not to challenge the wording of the questions, but retained the right to provide divergent interpretations of the results and take differing positions with respect to the relevance of the survey questions.
[84] The Study was designed to be conducted in two phases. Broadly speaking, Phase I consisted of government employees completing an online questionnaire. The questionnaire asked about the respondents’ copying patterns and gathered information on the most recent copying event. Phase II consisted of the completion of logs by government employees of their copying activities over a period of 14 consecutive days.
[85] There were initial disagreements among the Parties regarding some aspects of the Study methodology. The Board staff participated in discussions dealing with the Study. As noted above, the Board ordered that a pre-test, testing different versions of the two phases of the proposed Study on a small sample, be done in order to assess the overall reliability of the survey.
[86] The pre-test was conducted between May 6, 2011 and June 20, 2011. For the pre-test, a simple random sample of 1,149 names and their e-mail addresses was selected from a list of all employees collated from submissions from each jurisdiction. A total of 493 of the people so selected completed Phase I of the pre-test, and 49 completed Phase II. The pre-test resolved the design problems; some minor changes in wording to the questions were made.
[87] The Objectors then supplied lists of employees to constitute the sampling frame for the main survey. There were 135,726 employees available for surveying. By agreement among the Parties, Phase I would have at least 9,000 completed questionnaires, for which 27,000 individuals would be invited to participate. The sampling of these individuals was also simple random sampling. Furthermore, the individuals chosen were randomly split into ten groups with staggered start dates for the survey.
[88] The fieldwork, conducted by R.A. Malatest and Associates, began on September 19, 2011 and ended on December 11, 2011. There were 9,844 completed Phase I questionnaires. Of these, 3,993 respondents were eligible to complete the Phase II questionnaire by virtue of having made at least one copy of a published document during the two months before completing the Phase I questionnaire. From these eligible respondents, 1,285 respondents agreed to complete the Phase II questionnaire. Finally, 794 respondents actually completed the Phase II questionnaire.
[89] It is worth mentioning two differences between this Volume Study and the study in the K-12 decision. (“K-12 Study”) In the K-12 Study, only photocopying was measured. In this Volume Study, eight types of copying were measured: photocopying, scanning, printing, e-mailing, faxing, posting on an intranet, posting on the Internet, and saving an electronic file. In the K-12 Study, there was an assistant beside each logged photocopier to monitor copying activity. In the Volume Study, there was no monitoring; instead, all reports of copying were based on recollection (Phase I) or self-logging (Phase II).
B.
Results of the Study as analyzed by Access
[90] Access computed the amount of compensable copying from Phase I of the Volume Study. This amounted to 44.7 exposures per FTE from books, 144.2 from journals, 15.6 from magazines, 55.2 from newspapers, and 24 from other genres of works. In total, this amounted to 283.7 compensable exposures per FTE without accounting for fair dealing or other users’ rights. Access then calculated two other versions of these numbers, based on two alternative approaches to fair dealing. This reduced the total compensable exposures to 250.8 per FTE or 237.1 per FTE, depending on which approach was used.
[45]
[91] In its reply, Access filed the amount of compensable exposures from Phase II of the Volume Study. This amounted to 16.4 exposures per FTE from books, 25.8 from journals, 11.5 from magazines, 12.2 from newspapers, and 38.9 from other genres of works without accounting for fair dealing or other users’ rights. In total, this amounted to 104.8 compensable exposures per FTE. Access then calculated two additional versions, based on its two proposed approaches to fair dealing. This reduced the total compensable exposures to 91.1 per FTE or 84.9 per FTE, depending on which approach was used.
[46]
C.
Results of the Study as analyzed by the Consortium
[92] The Consortium calculated the amount of compensable copying from Phase I of the Volume Study. It determined the amount of “potentially compensable” copying to be 23.95 exposures per FTE from books, 72.26 from journals, 10.42 from magazines, and 16.29 from newspapers. This amounted to 122.92 potentially compensable pages copied per FTE in total. The Consortium then calculated actual compensable exposures (taking into account fair dealing and other users’ rights). This reduced the total compensable exposures to 11.80 per FTE.
[47]
[93] Finally, the Consortium calculated the amount of compensable copying from Phase II of the Volume Study. It calculated the amount of “potentially compensable” copying to be 2.59 exposures per FTE from books, 8.49 from journals, 1.17 from magazines, and 1.25 from newspapers. This amounted to 13.49 potentially compensable exposures per FTE in total. The Consortium then calculated actual compensable exposures (taking into account fair dealing and other users’ rights). This reduced the total compensable exposures to 1.15 per FTE.
[48]
D.
Compensability re-analysis and copying event data
[94] On March 14, 2013, the Board issued a notice informing the Parties that it was of the preliminary opinion that the compensability analyses performed by them were inadequate and ordered that an analysis be performed for each qualifying copying event, logged in Phase II, and that all survey data be provided for these events.
[95] In this notice, the Board stated that Phase II consisted of 1,466 copying events. This figure was not contested by the Parties. The Parties made further submissions, following the Board’s additional instructions on how to proceed. In particular, on May 6, 2013, the Board ordered that the copying events performed by a participant whose copying log was incomplete, as determined both by Mr. Gauthier and by R.A. Malatest, would be excluded from the analysis.
[96] On May 30, 2013 Access submitted that the events should be partitioned as follows:
Of the 1,466 events, 768 involved making a copy for the person making the copy (as opposed to for someone else).
Of these 768 events, 483 events involved copying a published document.
Of these 483 events, Access identified 311 where the work copied was in its repertoire.
Of these 311 events, Access claims compensability for 291. [49]
[97] On May 31, 2013, the Consortium advised the Board that it agreed with the figure of 291 as the number of qualifying events that should be analyzed.
[50]
These figures are reproduced in the Annex, Table 1.
[98] On June 17, 2013, the Consortium submitted its analysis, and details of the 291 copying events on behalf of itself and the Government of BC. Access made its submission on July 22, 2013. The submissions included spreadsheets which provided, among other things, identifying information on the work copied in each of the 291 events, and the answers that employees provided in the Phase II questionnaire in relation to those events. This included the number of pages copied, the number of sets made, the number of persons to which copied sets were distributed, and the purposes for which the copy was made. Lastly, the spreadsheets included the Parties’ submissions on whether each of the 291 events is compensable for the purposes of this Tariff.
E.
Second calculation of per-FTE volume
[99] Given the design of the Volume Study, the copying events therein are not equally representative. Therefore, calculating the total number of compensable pages copied is not a simple matter of multiplying the sum of the pages from compensable events by some constant value.
[100] As a result, on July 21, 2014, the Board issued a further notice with its preliminary conclusions as to which events from the Volume Study were compensable for the purposes of determining a royalty rate for the Tariff. Twenty-six events were identified in the notice, along with the genre of the work copied, and the number of compensable pages for each of the two proposed tariffs. The Parties were asked to calculate the weighted total number of pages copied, per FTE, per year, for each of the following genres of works: book, newspaper article, and magazine article. The calculations were to be done twice, once for 2005-2009 and once for 2010-2014, and the weighted calculations were to use the same weighting described in section 2.5 of Exhibit AC-4 or section 3.5 of Exhibit Consortium-3.
[101] The Parties provided their responses on August 28, 2014. In its response, Access calculated the number of compensable exposures to be 0.68 per FTE for the 2005-2009 period, and 0.75 per FTE for the 2010-2014 period.
[51]
The Consortium calculated that these exposure amounts are 1.3768 per FTE for the 2005-2009 period, and 1.5179 per FTE for the 2010-2014 period.
[52]
VII.
INTERPRETING THE VOLUME STUDY DATA
A.
Introduction
[102] As discussed below in part XIV, we use the “Volume times Value” methodology to determine the royalty rate for this Tariff, where volume is the number of compensable copies made (in pages) and where the value is the rate payable for each copied page made.
[103] The Parties agreed that there are 291 events in the Volume Study that are potentially compensable for the purposes of establishing a royalty rate for this Tariff.
[53]
For each of these 291 events, the Parties made submissions and filed data, such as the length of the work copied and the number of pages copied, as well as on the compensability of the events that is, whether the event represents copying for which a licence would be required from Access under the Tariff.
[54]
[104] We discuss the page-count, pages copied, and genre of the works copied in the Volume Study in parts VII.B, VII.C, and VII.D, below. We then determine the compensability of the events for the purposes of this Tariff in parts VIII-XII. The number of events captured by each step in the consideration of the events’ compensability is summarized in the Annex, Table 2.
[105] When considering the compensability of an event from the Volume Study, a finding in relation to a copying event is not a finding in relation to the actual copying event that occurred in 2011, when the Volume Study was carried out. Rather, it is a finding that the event, as described by the data captured in the Volume Study, represents compensable or non-compensable copying events that will have occurred during the period of the Tariff.
[106] As discussed in part XV, below, the Parties agreed that a single tariff could be certified for the 2005-2014 period, and that its administrative provisions should be based on the wording of the 2010 Proposed Tariff. As such, we generally refer to the wording of the 2010 Proposed Tariff during our analysis.
B.
Page-count of the copied work
[107] The length, or page-count, of the copied work is important primarily for determining what percentage of the work was copied. Additionally, the page-count also serves to provide additional context where the data from the Volume Study is ambiguous, such as when certain values are missing (e.g., the number of pages copied is marked as unknown), or where certain values are incorrect (e.g., where the number of pages copied is greater than the page-count of the work).
[108] Both Access and the Objectors submitted values for the page-count of the works copied. However, the page-counts for compilation works, such as newspapers, journals, and magazines that were submitted by the Objectors were almost always an average page-length for the entire compilation, not for the individual work copied. As described below in part X.B.3, the most relevant page-count for our analysis is that of the individual work, not of the compilation in which the work is included.
[109] We therefore determine the page-count of the copied work as follows:
Where Access provided a page-count for the work, this was used. Access is usually in the best position to provide the page-count of a work.
For works other than books, the Objectors provided page-counts that appear to be the average page-count of the collection in which the work appeared, instead of the length of the work itself (e.g., the page-count of a newspaper, as opposed to the newspaper article that was copied). Therefore, where Access did not know the page-count of a work, the Objector’s page-count was used only in instances of works that did not appear as part of a collection of works (usually books).
Where neither Access nor the Objectors could provide a valid page-count, the employee’s page-count was used. However, where the page-count the employee provided appeared to be incorrect (e.g., it appears to be the number of copies made multiplied by the number of pages copied), the page-count of the work was marked as unknown.
Where the work copied was a newspaper, and, given the copying methods used on it, the source appeared to be in electronic format, the work was assigned a page-count of 1, in accordance with the testimony of Prof. Wilk, which was relied on by Access. [55]
C.
Number of pages copied
[110] For each copying event, the employee making the copy was asked to enter the number of pages copied. She could not enter a number less than one, she could not enter a non-whole number (e.g., 2.5), nor could she provide a value with a unit other than number of pages (e.g., a person could not indicate that they copied a certain number of paragraphs). For this reason, our considerations of issues relating to the length of the work copied, or the amount that was copied, use the number of pages as a measure. The effect of this limitation on our consideration of whether a copy was a substantial amount of the work is discussed in part X.B, below.
[111] For 13 events in the Volume Study, the surveyed employee entered a value for the number of source pages copied that is greater than the page-length of the entire work. While some of these events appear to be minor errors (e.g., entering 59 pages instead of 56), others are more significant (e.g., entering 23 source pages from a 6-page work). It is possible that some employees entered the total copied pages made, instead of the number of pages in the work from which the copies were made. In these instances, we reduced the number of pages copied to the number of pages in the work.
D.
Work genre
[112] Where Access’ assessment of a work’s genre was the same as that entered by the employee, we accepted this as the genre. For those events where Access’ assessment did not match that of the employee, we examined the other available information, such as the length and name of the publication, in order to determine the genre of the work copied.
VIII.
REPERTOIRE
A.
Government-owned copyright
[113] In its evidence,
[56]
as well as in public statements, such as on its website,
[57]
Access has stated that it does not license the copying of works whose copyright is owned by a government – with the exception of Quebec and Australia.
[114] We note that to the extent that such works appear in a compilation, the copyright in that compilation, if any subsists, may belong to a person other than a government (e.g., as in the case of a compilation of statutes). However, in order for a substantial part of a compilation to be copied, a part of the compilation that represents a substantial portion of the author’s skill and judgment expressed therein would have to be copied.
[58]
Where the skill and judgement expressed by the author in the compilation are not contained in the portion copied, such as where only one work, or less, is copied from such a compilation, only the underlying work is copied, and not the compilation in which that work is contained. Furthermore, even if the maker of the compilation adds factual information, or makes mechanical changes to the underlying work, such as changing the font or correcting grammatical or spelling errors, these are insufficient to warrant copyright protection.
[59]
[115] Therefore, for the purposes of this Tariff, the relevant owner of copyright in such cases is the owner of copyright in the immediate work copied, not the owner of copyright, if any, in the collection in which the work appears.
[116] Given the evidence before us, we conclude that it is more likely than not that in 4 of the copying events the copyright in the work copied belongs to a government other than the Government of Quebec or Government of Australia, and are therefore works the copying for which Access does not issue a licence. This being the case, we exclude them from compensability for the purposes of this Tariff, leaving 287 of the 291 events as being potentially compensable.
B.
Non-affiliates
[117] Access enters into agreements with owners of copyright which permit Access to license the reproduction of the rights holders’ works. Owners of copyright that have entered into such an agreement with Access are often referred to as affiliated rights holders, or “affiliates.” Those that have not entered into such an agreement with Access are referred to as non-affiliated rights holders, or “non-affiliates.”
[118] At the hearing, Ms. Levy testified that Access’s repertoire consists of all published
[60]
works that are not specifically excluded by the exclusions list,
[61]
a list that Access’ witnesses described as “a list of rights holders who have excluded their work(s) from coverage by Access Copyright or another collective.”
[62]
Specifically, Access’ witnesses stated that Access’ repertoire includes
[a]ny published work in print form or that has a print equivalent (a “Print Work”), that has been issued to the public with the consent or acquiescence of a rightsholder that is either resident or domiciled in or a citizen of or incorporated in Canada and/or in a jurisdiction with which Access Copyright has a bilateral agreement in place, that has not been excluded by the rightsholder. [63]
[119] In short, Access considers in its repertoire almost all published works, without regard as to whether there is any relationship between the rights holder and Access.
[120] In its 2010 Proposed Tariff, Access defined the term Repertoire as
those Published Works published in or outside of Canada by any author or publisher, estate of an author or publisher or other person with a copyright interest in the Published Works who, by assignment, grant of licence or by appointment as an agent or otherwise, has authorized Access Copyright to collectively administer reproduction rights in Published Works and those Published Works published in or outside of Canada by other copyright owners where an agreement between Access Copyright and another reproduction rights collective society authorizes Access Copyright to represent such other copyright owners. [64] [emphasis added]
[121] In an attempt to reconcile this proposed definition with the testimony of Access’ witnesses, on December 4, 2012, the Board issued a notice, asking the Parties to agree or disagree with several propositions.
[122] The Consortium and the Government of BC agreed with the proposition
[65]
that
the owner of a non-repertoire work that is captured during a distribution survey and who cashes the royalty cheque received from Access does not bring that work into the repertoire. However, by virtue of agency by ratification, the owner of the non-repertoire work no longer can complain of copyright violation by the user on account of the copy associated to the cashed cheque. [66]
[123] Access disagreed with the first portion of this proposition, submitting that works where an unaffiliated copyright owner has accepted payment are in its Repertoire, as defined in the 2010 Proposed Tariff.
[67]
[124] In its response, the Consortium further submitted that the repertoire from which the royalty rate will be derived cannot include the published works of rights holders who have not, within the Act’s definition of a collective society, “by assignment, grant of licence, appointment of it as their agent or otherwise,”
[68]
authorized Access to act on their behalf for the purposes of the proposed tariff. Access may, however, choose to share a portion of these royalties with copyright owners of works that are not in its repertoire on a purely voluntary basis.
[69]
[125] In a similar vein, the Government of BC submitted that the definition of “Repertoire” in the 2010 Proposed Tariff does not include “works that are captured during a distribution survey, where the owner accepts the royalty payment received from Access Copyright, for the specific use by the particular user.”
[70]
[126] In its K-12 decision,
[71]
the Board held that the cashing of a royalty cheque by a copyright owner, issued in relation to one or more copying activities, had the effect that those activities were thereby legitimized, such that “[the copyright owner] cannot take proceedings for infringement of copyright against the person who made the copy.”
[72]
The Board went on to hold that
[t]he 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. [73]
[127] In the matter before us, payments have not been made by Access in relation to the copying events captured in the Volume Study, including to those with whom Access does not have an affiliate agreement. Since no payments have been made, no agency relationship could have arisen between the relevant owner of copyright and Access. The argument the Board accepted in K-12 for including works of non-affiliated copyright owners in the determination of the royalty rate is thus inapplicable in this matter.
[128] Indeed, Access acknowledged that “[a] work-by-work analysis of what is in Access Copyright’s repertoire as a result of affiliation or claim for agency will not yield […] useful results because those acts have not yet been ratified.”
[74]
It argued instead that the Board should disregard the current non-existence of these agency relationships, rely instead on Access’ practice of distributing royalties to non-affiliates, and thereby infer that agency relationships will eventually be formed.
[75]
The eventual acceptance by the rights holder of the payment from Access, it submitted, ratifies the transaction retroactive to the date on which Access licensed the work.
[76]
[129] For the reasons that follow, we find that this is not sufficient for us to include copying events where the owner of copyright was not affiliated with Access as compensable for the purposes of determining a royalty rate in this Tariff.
[130] Firstly, Access’ current distribution model for royalties collected for copying by governments does not appear to be based on any title-specific information. Access stated that for publishers, 5 per cent is paid on the basis of how many works they bring into Access’ repertoire, and 95 per cent is based on a model which attempts to approximate copying patterns. For authors, Access distributes 40 per cent of the funds to affiliated authors equally without regard to the usage of their works, and the remaining 60 per cent based on a model that takes into account the genre, date, and amount of works that the author published.
[77]
[131] The witnesses also detailed the distribution of royalties that Access makes to non-affiliated rights holders:
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. In its 2005 distribution, Access Copyright distributed $1,338,909.34 to 1,529 Non-Affiliated Rightsholders. This represented 6.5% of the total distributions made for that same period. […] In its 2010 distribution, Access Copyright distributed $1,079,843.95 to 916 Non-Affiliated Rightsholders. This represented 4.63% of the total distributions made for that same period. [78]
[132] In K-12, the Board held that the acceptance by rights holders of payments of royalties based on measured copying granted Access the power to act on their behalf in respect of those measured copies.
[79]
Acceptance by rights holders of distributions not based on actual copying cannot serve as the basis for the kind of retroactive ratification contemplated in the K-12 decision.
[133] Secondly, and most importantly, the vast majority of copying of non-affiliated works will never be reported to Access. As Access stated, and as is only logical, Access only sends cheques in relation to copying events for which it has information. Access only claims, and can only claim, agency relationships in cases where a rights holder has actually cashed a cheque she received from Access.
[80]
[134] While Access pointed to the percentage of rights holders that cash cheques that are issued, this is not the measure that is important. Rather, it is the number of copying events for which such cheques will actually be issued that is important. In most situations, Access will only have information on a very small percentage of the actual copying that takes place, and can therefore issue cheques to non-affiliated owners of copyright for only a very small percentage of copying events.
[135] For example, in this matter, the only information Access currently has on specific copying events is the Volume Study. The Study represents approximately 0.05 per cent of the actual potentially compensable copying that will have occurred during each year of the Tariff period and thus approximately 0.005 per cent for the entire Tariff. Even were this to be supplemented with a similar study, this percentage would only double to approximately 0.01 per cent.
[136] Access can only send cheques, and thus be able to argue for the existence of an agency relationship, in relation to at most 0.005 per cent of copying from works of non-affiliated rights holders. Even if we were to accept the premise that the sending of a cheque by Access in relation to a copying event, and its subsequent cashing by the owner of copyright in the work copied, forms an agency relationship in relation to that particular copying event, it would remain that this would not happen for at least 99.995 per cent of the actual potentially compensable copying of works of non-affiliated rights holders that will occur during the Tariff period.
[137] Any payments that may eventually be made to non-affiliates in relation to copying events captured in the Volume Study are not representative of what will occur in relation to all other past and future copying of works of non-affiliated rights holders. The latter copying cannot be, and will never become – by its very nature of not being measured – the subject of a copying-based payments from Access, and thus never be the subject of the kind of temporary agency relationship contemplated in the K-12 decision.
[138] Since Access cannot license the copying of a work for which it has not itself received authorization from the owner of copyright, the act of copying a work of a non-affiliated rights holder is a potential infringement of copyright. Arguably, this act of potential infringement may be retroactively “legitimized,” by the copyright owner’s cashing a royalty cheque related to that copying. However, if the copyright owner does not receive such a cheque, as will almost always be the case, she can bring proceedings for infringement of copyright. This would be so despite the fact that royalties have notionally been paid in relation to the making of that copy by the licensee to Access.
[139] Access does not provide a list of affiliate copyright owners to licensees. Its web-based “lookup tool” is based on the position adopted by Access’ witnesses: that Access can authorize the copying of all published works, unless the rights holder has explicitly stated that they may not.
[81]
Therefore, in relation to an upcoming act of making a copy of a work, a user cannot know with certainty whether Access’ actually has the authority to license the copying of that work or not and therefore whether she is about to infringe copyright or not.
[140] To count such copying events as compensable would have the effect of requiring the licensee to notionally pay for copying events that will never become ratified by the rights holder, cannot be authorized by Access, and thus remain potential infringements of copyright.
[141] These concerns are heightened by the significant number of copying for which the work in the Volume Study was not in Access’ repertoire: 41 of the 291 copying events were identified by Access as events where the works copied were such that the copyright owner had authorized neither Access, nor a collective society with whom Access has a bilateral agreement, to administer the reproduction right in Canada.
[142] We note that these concerns are not applicable in situations where the rights holder has already authorized Access to act on her behalf. They are only relevant in those cases where Access claims that its authority to license the reproduction of a work derives from an agency relationship formed by the sending and cashing of such payments.
[143] In short, Access’ repertoire for the purposes of this Tariff consists of those works for which the owner of copyright has authorized Access to administer the reproduction right, and those works for which the owner of copyright has authorized another collective society to administer the reproduction right in Canada through agreements with other collective societies.
[144] For the reasons above, we exclude from the royalty calculation events where works were copied for which Access noted their relationship with the owner of copyright as one of “agency.” Of the 287 remaining events, 39 were so designated. This leaves 248 events as potentially compensable.
C.
Other issues with non-affiliate copyright owners
[145] While not raised by the Objectors, the Board notes possible irregularities in Access’s identification of works “in its repertoire.”
[146] For example, several events in the Volume Study for which the work appears to be a US publication, and for which the Copyright Clearance Centre’s (a US-based collective that administers the reproduction right in published works) website states that it cannot provide a business licence (or could not do so in relation to certain activities, such as digital copying), were nevertheless identified by Access as being in its repertoire. While it is possible that a copyright owner may permit collectives outside of the United States to license activities that the owner does not permit a US-based collective to license, it would have been preferable that such discrepancies be addressed and evidence adduced.
[147] In the future, in relation to rights in works for which it has received authorization to administer from another collective, Access may wish to identify from which collective it has received this authorization, and, if possible, address any discrepancies in the scope of the authorization granted to the source collective and the scope of the authorization received by Access.
[148] As another example, while Access provided the Board with the names of the authors and/or publishers of the works in the Volume Study, the copyright owner – the relevant person for the purposes of determining Repertoire – was not identified. Yet, Access receives its authority to license the copying of a work from a person when they are the owner of copyright; merely being a publisher or an author is not sufficient to validly grant Access such an authorization. In the future, Access may also wish to identify the copyright holder – as opposed to the author or publisher – of the works in question.
[149] We raise these issues in the hope that it will encourage the Parties to address them in greater detail in the future. As such, these observations have no bearing in establishing the royalty rate for this Tariff.
IX.
DIGITAL COPIES
[150] Unlike the 2005 Proposed Tariff, the 2010 Proposed Tariff sought to permit the making of digital copies. Section 2 defined “Copy” as a reproduction of a Published Work made by the process of “reproducing by a machine, device or computer that makes a Digital Copy […] and includes the Digital Copy.” “Digital Copy,” in turn was defined as “any electronic file of a Published Work.”
A.
The Deletion Provision
[151] However, the making of digital copies would be restricted by a requirement set-out in paragraph 5(d), the “Deletion Provision”, of the 2010 Proposed Tariff which states that
[w]here the Licensee is no longer covered by a tariff for the making and distribution of Digital Copies, the Licensee shall immediately cease to use Digital Copies of Published Works in the Repertoire, delete from their hard drives, servers or storage area 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.
[152] In a notice of May 6, 2014, the Board asked the Parties to address several issues, including
whether the Board has the jurisdiction to certify a tariff with such a condition;
whether the proposed condition is sufficiently clear in scope and legal effect;
whether such an undertaking by the Objectors would be feasible; and,
what effect non-inclusion of this condition would have on the compensability of Digital Copies. [82]
[153] All Parties, including Access, acknowledged that there were difficulties with the Deletion Provision, as worded. Furthermore, all Parties expressed, either explicitly or implicitly, the view that the provision itself is not directly legally enforceable (i.e., there is no independent remedy for a breach of the provision), nor should it be understood as retroactively causing a copying event to fall outside the scope of the Tariff.
[83]
In short, it appears that the Parties do not believe this provision has any effect, other than to serve as a notice as to how the beneficiaries of the licence are to govern themselves. Indeed, in its response to the Board’s notice, Access proposed to remove virtually all aspects of the Deletion Provision, a proposal that would, in effect, restate that certain acts could not be done without the benefit of a licence.
[84]
[154] The Objectors, both in previous submissions and in their response to the May 6, 2014 notice, submitted that it would not be feasible for them to comply with the Deletion Provision.
[85]
For its part, Access, in its submissions, relied on the statements of the Objectors that adhering to the provision appears not to be feasible.
[86]
However, Access then went on to argue that “such an undertaking by the Objectors would be feasible given that Ontario has agreed to the condition in its licence.”
[87]
[155] Ontario’s acceptance of such a provision in a previous licence does not, on its own, constitute evidence that compliance with the provision is feasible. We do not have evidence to what extent Ontario has complied with, or has taken steps to comply with this condition.
[156] Given that the Objectors do not already have mechanisms in place to track the copying of digital copies, and given that Access relied on this fact itself in its submissions, we conclude that it is more likely than not that the Objectors would not be able to comply meaningfully with the Deletion Provision set out in paragraph 5(d) of the 2010 Proposed Tariff.
[157] While Access argued that the Deletion Provision would apply only during the effective period of the Tariff,
[88]
and implied that it created no obligation after its expiration, we do not accept this reading of the provision. In our view, it would have the effect of creating an obligation to delete digital copies made under the Tariff that could be triggered well after the Tariff’s expiration. In fact, the possibility that the obligation could be triggered would continue indefinitely into the future. This would be a very undesirable effect.
[158] Finally, the provision requires that the licensee ceases to use digital copies once the licensee is no longer covered by an applicable tariff. This raises an additional, important issue. As the Objectors correctly pointed out, once a copy has been lawfully made, be it a digital or a paper copy, under a licence or a tariff, there is no reason why the licensee should eventually be deprived of the use of that copy; mere possession of a digital copy of a protected work after termination of a licence does not infringe any rights of the copyright owner under the Act. Although Access is concerned with the ease of dissemination of digital copies, and while this is a valid concern, the means it proposes to address this concern is simply not appropriate.
[159] For all the above reasons, we do not include the Deletion Provision in the Tariff.
[160] Given our conclusion, it is not necessary for us to address the question of whether the Board has the jurisdiction to include in a tariff a provision which creates an obligation that continues indefinitely into the future, and whose existence is not tied to the duration of the tariff from which it arises.
B.
The effect of non-inclusion of the Deletion Provision
[161] In their responses to the Board’s notice, the Objectors pointed to the fact that Access has stated that the Deletion Provision “reflects the licence that Access Copyright and its affiliates are willing to grant with respect to digital uses and that has been approved by Access Copyright’s Board of Directors,”
[89]
and that it is not authorized to license any digital copying of its affiliates’ works without the corresponding requirement to delete such digital files upon termination of the licence. In the Objectors’ view, non-inclusion of the condition would therefore render the making of digital copies, including those made in the events in the Volume Study, outside of Access’s licence and therefore non-compensable.
[162] In its response, Access stated that it has
obtained permission from its Board of Directors to authorize the licensing of Digital Copies under the Proposed Tariffs without the condition set out in section 5(d). Access Copyright will request express permission from its affiliates to remove this provision from its licences and we reasonably expect that permission to be forthcoming. [90]
[163] Access also submitted that it can issue a licence in relation to the digital copies made in the events in the Volume Study even without the Deletion Provision, on the basis that the condition is unlikely to be triggered and on the basis that it has received approval from its Board of Directors to remove the provision.
[91]
[164] As Access itself stated, “the deletion of digital files following the termination of a licence is a condition of the grant of rights provided by Access Copyright’s affiliates to Access Copyright.”
[92]
We believe this to be a correct interpretation of the effect of Access’ agreement with affiliates, which states that Access may license print-to-digital and digital-to-digital copying, provided that Access “provide[s] for the deletion of all digital files no later than the termination of our licence.”
[93]
[165] 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 of 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.
[166] This finding does not deprive the copyright holder in the works so copied of any rights and remedies she may otherwise have. These events simply represent copying for which this Tariff does not provide a licence, and, as a result, for which Access cannot collect royalties under this Tariff.
C.
Conclusion
[167] Since Access cannot grant a licence for the making of digital copies without the presence of the Deletion Provision, since the Tariff cannot have the effect of providing a licence for uses that Access itself cannot grant, and since we have excluded paragraph 5(d) from the 2010 Proposed Tariff, the making of digital copies is not an act that will be permitted under the Tariff, and is thus not compensable for the purposes of establishing a royalty rate.
[168] Of the 248 events that remain under consideration, 100 events involved only the making of digital copies. These were removed from the events that could be potentially compensable, leaving 148 potentially compensable events.
[169] The events that included both the making of digital copies as well as copies that do not meet this definition were not excluded (e.g., where a photocopy of a work was made and where it was scanned in the same copying event). However, only the non-digital copies captured in these events contributed to the compensable volume if the event was determined to be compensable.
[170] We note that the certification of a tariff for the copying of digital copies is also problematic for other reasons. Ms. Levy testified that while Access had agreements with 31 other collective societies, only 12 of these agreements covered digital copying.
[94]
As such, digital copying raises a number of repertoire-related issues. Given our conclusion above, we do not address them, but simply express the hope that these issues will be given a broader airing the next time the Board considers a tariff filed by Access.
X.
ACTS OUTSIDE OF THE SCOPE OF THE TARIFF
[171] Copying events where the work copied was within Access’ repertoire must be authorized under the Tariff for those events to be compensable. Where the Tariff has limitations whereby a copying event in the Volume Study would not be permitted, that event is not compensable for the purposes of determining the royalty rate for this Tariff. Such acts of copying are not covered by the Tariff, and remain potential acts of copyright infringement.
[172] To count such copying as compensable would have the effect of making the Objectors pay for activities that the Tariff does not authorize. Therefore, those events that would not have been authorized by the Tariff, had it been in place at the time when the copies were made, are not compensable for the purposes of the Tariff.
A.
Amount of work copied greater than permitted by the Tariff
[173] The 2010 Proposed Tariff would permit the making of copies of published works, within certain limitations on the quantity of a given work.
[174] In relation to books, the 2010 Proposed Tariff does not authorize the copying of more than 10 per cent of a work, unless such copying is of an entire chapter from a book, in which case it does not authorize the copying of more than 20 per cent of that book.
[95]
In relation to genres of works not identified in paragraph 3(a), the 2010 Proposed Tariff does not authorize the copying of more than 10 per cent of that work.
[175] We identified four events for which the work copied was a book, and of which more than 20 per cent was copied. Furthermore, we identified one event for which the percentage of the work copied was greater than 10 per cent, and the work was not a work identified in paragraph 3(a) of the 2010 Proposed Tariff as a work for which a copy of greater than 10 per cent could be made. These five events were removed from the events that could be potentially compensable, leaving 143 potentially compensable events.
[176] Just as in relation to the making of digital copies, this finding does not deprive the copyright holder in the works so copied of any rights and remedies she may otherwise have. These events also represent copying for which this Tariff does not provide a licence, and, as a result, for which Access cannot collect royalties under this Tariff.
B.
Where reproduction is not a substantial part of the copied work
[177] Under the Act, “‘copyright,’ in relation to a work, means the sole right to […] reproduce the work or any substantial part thereof.”
[96]
[emphasis added] Therefore, “copyright” does not include the right to reproduce a part of a work, where the part of the work that is reproduced is not a substantial part thereof.
[178] Access submitted that since an analysis of the qualitative aspect is an essential element of determining whether a substantial portion of a work is copied, this determination cannot be made solely on the number of pages copied. Such an approach would not address the qualitative aspect of the portion taken.
[97]
[179] Access pointed to previous jurisprudence on what constitutes a “substantial part” of a work, including U & R Tax Services Ltd. v. H & R Block Canada Inc.
[98]
and Hager v. ECW Press.
[99]
These cases stated that some of the factors to be considered in evaluating a “substantial part” are:
the quality and quantity of the material taken;
the extent to which the defendant’s use adversely affects the plaintiff’s activities and diminishes the value of the plaintiff’s copyright;
whether the material taken is the proper subject-matter of a copyright;
whether the defendant intentionally appropriated the plaintiff’s work to save time and effort; and,
whether the material taken is used in the same or a similar fashion as the plaintiff’s. [100]
[180] According to Access, the fact that employees made “slavish copies” of the works, used them in an identical fashion as the author and publisher, and could have obtained a licence to make the copy, should result in the Board not making any adjustments for non-substantial copying.
[101]
[181] Access also submitted that journal and newspaper articles tend to present their most important information first and that this is likely the portion that was copied by employees. Therefore, copying one page of a journal, magazine or newspaper article will amount to copying a substantial amount of that work.
[102]
As such, it argues, the Board has no basis on which to find any of the copying events as amounting to non-substantial copies.
[103]
[182] The Objectors submitted that some of the copying events captured by the Volume Study do not amount to the copying of a substantial portion of the source work, and that such copying should not be compensable for the purposes of this Tariff. They argued that for a copy to be included in the Board’s estimate, it must be of a substantial part of a work. If an act is done that does not involve at least a substantial part of a work, copyright protection of that work simply never becomes relevant.
[104]
Furthermore, the criteria evaluated by Access, such as what portion of the copied work will be comprised in the copier’s final product, or Access’ licensing practices, are irrelevant to a finding of substantiality.
[105]
[183] The Objectors further submitted that in arguing that parts of works that convey important factual information or ideas are substantial portions of the work, Access ignored the idea/expression dichotomy. An abstract, chart or table that summarizes the ideas presented in a work cannot on its own represent a substantial part of the expression in that work.
[106]
[184] The Government of BC submitted that while the quality of the part taken is an important factor in determining substantiality,
the purpose of the volume study conducted for purposes of these proceedings cannot practically be to precisely determine whether each event would legally constitute infringement. Such would require inordinate time and resources from both the respondents to the survey and the Board to determine minutely accurately all the factors that enter into the issue of substantiality. At the same time it would be unfair to require compensation in the tariff from users for non-compensable, insubstantial copies of work. The purpose of the survey is to provide a fair estimate of the amount of compensable copying being done by provincial or territorial government employees. There needs to be therefore a fair guideline on substantiality applicable generally to determine how many insubstantial copies are typically made in provincial and territorial governments and the rough and ready way to do that is quantitatively, while taking into account the other factors in respect of which data is available. [107]
i.
The current law
[185] As the Supreme Court of Canada stated in Cinar Corporation v. Robinson, “a substantial part of a work is a flexible notion. It is a matter of fact and degree […]. As a general proposition, a substantial part of a work is a part of the work that represents a substantial portion of the author’s skill and judgment expressed therein.”
[108]
[186] We note that the Court in Robinson did not refer to U & R Tax Services or Hager, nor to most of the factors that were enunciated therein. We also note that some of the factors considered in those latter cases appear to evaluate elements that do not go towards establishing the author’s skill and judgment that a particular portion of a work contains. We therefore find that, to the extent previous cases evaluated criteria not related to the amount of skill and judgment expressed in the portion of the work that was copied, they are not good precedent.
[187] In particular, some of the factors, such as whether the defendant intentionally appropriated the plaintiff’s work to save time and effort, should have no bearing on whether the amount copied is substantial or not. To what use a copied portion is put, and for what reason it was copied, cannot ex post facto determine whether the portion copied was protected by copyright. A portion of a work is substantial, or not, as soon as the work is created, whether or not it will be copied, and irrespective of the reasons for which it is copied.
[188] For example, Access argued that the fact that it sells transactional licences for photocopying individual works on a per-page basis demonstrates that individual pages of a work “have value and may not be accessed for free on a theory that one page is only an insubstantial part of a work.”
[109]
We disagree. Access cannot alter the scope of protection granted by the Act by its business practices. Whether Access licenses such uses or not is not relevant in the determination of substantiality. Access’ approach to determining substantiality also risks becoming one where “what is worth copying is prima facie worth protecting,” a position that has been described as unsustainable.
[110]
[189] Considering such factors would transform an evaluation of substantiality into a “fair dealing”- like evaluation. The reproduction of a portion of a work that is not a substantial portion thereof is not an act protected by copyright, ab initio. It is not a legal consequence that is subsequently formed when, for example, the copier does not compete with the owner of copyright in the marketplace.
[190] Given the Supreme Court’s pronouncement in Robinson on how substantiality is to be assessed, we conclude that the test to be applied in the present matter is whether the part of the work that was copied by an employee represents a substantial portion of the author’s skill and judgment expressed in the copied work.
ii.
Quantitative evidence and burden of proof
[191] While the Volume Study presented general information about the number of pages copied (and even this with some limitations), it contains no information about which portions of a work were copied (except in cases where the entirety of the work was copied), nor information about the qualitative aspects of the portion copied. This makes a qualitative determination of whether a particular portion represents a substantial portion of the author’s skill and judgment difficult – if not impossible.
[192] Access argued that without such a qualitative analysis, the Board cannot determine which of the copies made in the Volume Study were not substantial reproductions of the source work for the purposes of establishing a royalty rate.
[193] We disagree. Such a lack of qualitative information is likely to be encountered by finders of fact in any situation where a multitude of works are under consideration.
[194] As the litigation in Robinson
[111]
demonstrated, analyzing the qualitative aspects of even one work can involve numerous expert witnesses with conflicting testimonies, and voluminous supporting evidence. Collecting information on, and evaluating, the qualitative aspects of those works may be unworkable in some situations, such as tariff proceedings.
[195] We agree with the Government of BC that in the context of a royalty-setting exercise, these limitations should not result in the notion of “substantiality” being discarded altogether. As this Tariff is not use-based, to do so would result in a royalty rate which effectively has the licensee pay for even non-compensable copying. The purpose of the Volume Study was to provide a fair estimate of the amount of compensable copying being done by provincial or territorial government employees, and likely “represents the best evidence that can reasonably be expected in a matter that takes into account a plethora of vastly different jobs and government offices, and large numbers of different copying events.”
[112]
[196] While Access argued that the lack of qualitative information is the fault of the Objectors, the Parties both agreed on the manner in which the main source of data (the Volume Study) was collected.
[197] The royalty-setting exercise is not a mini-trial for each of the events in the Volume Study; such events are merely components of a proxy used in arriving at a royalty rate. As an economic regulatory agency, the Board must rely on the evidence before it to establish a fair and equitable royalty. Therefore, we determine the (non-) substantiality of the copying in the Volume Study events on a balance of probabilities, based on the evidence before us. In this case, that evidence speaks to the quantitative aspect of the copying events.
iii.
The work to be considered
[198] Many of the works that were copied in the Volume Study, such as magazine articles, journal articles, or newspaper articles, were part of a larger work, such as magazines, journals or newspapers.
[199] The Consortium submitted that the work to consider in the analysis of substantiality is the newspaper, journal or magazine itself. It argued that
[s]ince the analysis of whether a reproduction of a substantial part has been made involves an economic analysis of whether the copy is replacing the original […], the Board should look at the original in the form in which it is sold. [113]
[200] We disagree. As stated above, the determination of substantiality is based on the evaluation of skill and judgment expressed in the copied portion. Considerations such as whether a copy replaces the original may be relevant in the consideration of the “fairness” of a dealing, but does not go towards establish the substantiality of copying.
[201] Even if considering whether a copy replaces the original could assist in evaluating the skill and judgment expressed therein, a proposition that we do not accept, it remains the case that under the Act, each work is protected individually, and a substantial part of a work in a compilation may be copied even where a substantial part of the compilation is not copied.
[202] While there may be circumstances where it may be difficult to establish what the work to be considered is, such as where a previously larger work is split into smaller portions so that they may be sold separately, there is no evidence in this matter that suggests this was the case for the works copied in the Volume Study.
iv.
Conclusion
[203] We agree with Access that the copying of two pages from a book “may be sufficient to capture a substantial part (even if not all) of the intellectual product contained in the book.”
[114]
However, it is also possible that more significant amounts (quantitatively) of a work may be copied without it constituting a “substantial” portion thereof.
[204] In this matter, without the benefit of a qualitative analysis of each of the copied works, and without even knowing which portions of a work were copied, in our opinion the amounts proposed by the Consortium, being 1 to 2 pages of a work, are reasonable approximations in establishing non-substantiality. However, since 1 to 2 pages of a short work can amount to a great portion of that work, we further limit this approximation by requiring that the copying of 1 to 2 pages not constitute more than 2.5 per cent of the entire work, the percentage equivalent to what the Board had previously considered not to be substantial reproductions in its Satellite Radio Services decision.
[115]
[205] Thus, for the purposes of calculating a royalty rate for this Tariff, we consider that copying events where 2 pages were copied from a work of 80 pages or more, or 1 page was copied from a work of 40 pages or more, represent copying that was not a substantial amount of the work.
[206] We note several limitations resulting from the data of the Volume Study. As noted above in part VII.C, an employee completing a survey in the Volume Study could not enter a value of less than 1 page. According to the thresholds we adopt in this matter, reproductions of works shorter than 40 pages in length cannot be non-substantial reproductions for the purposes of calculating a royalty rate in this matter.
[207] Another limitation arises from the fact that many events did not have reliable evidence on the length of the work being copied. In such cases, it was not possible to evaluate the substantiality of the copying, even on a quantitative basis, and we assumed that the copying was substantial. The use of an average value to substitute for such missing data would not be appropriate in this case. Firstly, it is not clear that the works for which data are available are random selections from the overall data set. Secondly, averages are skewed by very large events (there can be no very small events in the data set, since the number of pages is a non-negative integer). Lastly, an average is a single value, and would therefore make all events that rely on this value appear similar, whereas the actual distribution may be very different (i.e., the use of an average could potentially make all dealings relying on it appear to be substantial or non-substantial). Using a median instead of a mean would only address the second of these issues.
[208] We are aware of these limitations. However, given that concluding that an event represents the substantial copying of a work does not automatically make it compensable (since it may be non-compensable on other grounds), the concern with these limitations is diminished. We consider that the application of these thresholds in this manner serves well in the determination of a royalty rate for this Tariff.
[209] Using the thresholds for substantiality described above, we conclude that 5 copying events represent copying where the amount of the work copied was not a substantial portion thereof. This leaves 138 potentially compensable events.
XI.
ACCESS TO INFORMATION ACT AND PRIVACY ACT EXCEPTIONS
[210] Subsection 32.1(1) of the Act provides that it is not an infringement of copyright for any person
to disclose, pursuant to the Access to Information Act, a record within the meaning of that Act, or to disclose, pursuant to any like Act of the legislature of a province, like material;
to disclose, pursuant to the Privacy Act, personal information within the meaning of that Act, or to disclose, pursuant to any like Act of the legislature of a province, like information;
[211] Out of the remaining potentially compensable copying events, two were indicated by the survey-taker as situations where copies were made for the purpose of disclosing information pursuant to an act akin to the Access to Information Act, or the Privacy Act. In its submissions, Access indicated that it does not consider these events as being compensable.
[116]
We agree.
[212] Two events are therefore removed from compensability for the purposes of the Tariff, leaving 136 potentially compensable events.
XII.
FAIR DEALING
[213] Sections 29, 29.1 and 29.2 of the Act all state that fair dealing for a permitted purpose does not infringe copyright. Copying events from the Volume Study that are potentially compensable may nevertheless be non-compensable if the copying they represent fall within one of the exceptions to infringement.
A.
Burden of proof
[214] Access argued that fair dealing is an affirmative defence that the Objectors must prove
[117]
and that the Objectors bear the onus of satisfying all aspects of the fair-dealing test.
[118]
The Consortium disagrees, stating that “there is no ‘defence,’ and the Objectors are not defendants.”
[119]
[215] While Access is correct that, in a litigation proceeding, the onus is on the person invoking “fair dealing” to satisfy all aspects of the test,
[120]
the proceeding before the Board is not one of copyright infringement: it is a proceeding to certify a proposed tariff. Furthermore, we generally agree with the Consortium that
Copyright Board tariffs are certified on a prospective basis, taking into account future use. A tariff cannot reasonably, or even feasibly, be treated as a claim of copyright infringement. Access Copyright cannot prove future infringement, nor can the Objectors defend against a speculative claim. [121]
[216] Furthermore, the analyses of the copying events in the Volume Study are not “mini-trials” either. While their analysis assists in the determination of the royalty rate for the Tariff, the compensability of these events, as such, are not the primary issue at hand. The Board is not attempting to determine the compensation to be paid for these events, and these events only, as might be the case in an infringement proceeding before a Court. Therefore, the usual “burden” framework present in civil litigation need not be applied in this instance, or not be applied in a strict fashion. This is in alignment with the Copyright Board’s mandate as an economic regulatory agency.
[217] Access also argued that where insufficient evidence has been adduced to evaluate one or more of the CCH factors, the Objectors did not meet their burden to establish fairness.
[122]
[218] While the factors identified in CCH are a “useful analytical framework,”
[123]
they are not requisite elements that have to be met for a dealing to be fair. Indeed, it is possible for a dealing to be fair without any evidence being adduced in regard to one or more of the factors. The Supreme Court of Canada noted in Alberta that
[i]n CCH, the Court concluded that since no evidence had been tendered by the publishers of legal works to show that the market for the works had decreased as a result of the copies made by the Great Library, the detrimental impact had not been demonstrated. Similarly, other than the bald fact of a decline in sales over 20 years, there is no evidence from Access Copyright demonstrating any link between photocopying short excerpts and the decline in textbook sales. [124] [emphasis in the original]
[219] Therefore, it is possible to evaluate the fairness of a dealing without evidence on every factor. Indeed, this is because whether a dealing is “fair” remains a question of fact and “a matter of impression.”
[125]
This impression may be gained through whatever evidence is before the decision-maker. If there is evidence that would tend to show that a dealing is more or less fair, it should be adduced by the party wishing to make the corresponding argument.
[220] Access went further and argued that “the Objectors have not introduced any evidence at all at step 2 of the fairness analysis: they have adduced no evidence about the fairness of the copies they claim are fair dealing.”
[126]
But this is not the case: information about each of the dealings is present in the Volume Study. It provides some information on the purpose of the dealing, the character of the dealing, the amount of the dealing, the nature of the work, and alternatives to the dealing.
[221] The fact that the information for each of the copying events is not as complete as it would likely be in the case of copyright infringement proceeding is not an obstacle to assessing the amount of copying that should be treated as not compensable on the ground of fair dealing for the purposes of this Tariff. In this matter, the Parties agreed on the manner in which the main source of evidence (the Volume Study) was collected. It may be true that “[t]his study represents the best evidence that can reasonably be expected in a matter that takes into account a plethora of vastly different jobs and government offices, and large numbers of different copying events.”
[127]
[222] Therefore, we determine the fairness of the dealings in question on a balance of probabilities, based on the evidence that has been adduced, without the application of a strict “burden” framework. Where evidence in relation to a particular factor has not been adduced, this will generally not weigh towards fairness or unfairness.
B.
Fair-dealing practice
[223] In CCH, the Supreme Court of Canada stated that fair dealing can be made out either by demonstrating that there exists a general practice that is based on an enumerated fair-dealing purpose, and is, in fact, fair, or by demonstrating that a particular copying event (in this instance an event from the Volume Study) was fair dealing.
[128]
[224] Having regard for the totality of evidence, we find that the practices and, to the extent there are any, policies, of the Objectors are not sufficient to constitute a practice in the sense that its existence would be sufficient to demonstrate fair dealing. While other governments had copyright policies for specific ministries or units within those ministries,
[129]
only the Government of BC submitted that it had a government-wide copyright policy. In 2004, responsibility for compliance with the Act was assigned to Deputy Ministers, and notices advising the need for compliance were affixed to photocopiers.
[130]
These notices stated that it is generally contrary to the Act to copy published material without permission. We find it difficult to evaluate these steps taken by the Government of BC as a “practice,” in the sense that CCH contemplated the practice of the Great Library.
[225] As such, we consider the copying events of the Volume Study one-by-one in order to determine whether they represent copying that falls under fair dealing.
C.
Fair dealing – Step 1
[226] The test for fair dealing, as articulated by the Supreme Court in CCH, has two steps. The first determines whether the dealing is for an allowable purpose; the second step assesses whether the dealing is “fair.”
[131]
In the cases of fair dealing for the purposes of criticism, review, or news reporting, the source must be mentioned, as well as the author, if given in the source.
[132]
i.
Purpose (Step 1) - Who is the user? Which purposes should be considered?
[227] In Alberta, the Supreme Court of Canada stated that “the relevant perspective when considering whether the dealing is for an allowable purpose under the first stage of CCH is that of the user.”
[133]
[228] Access argued that since “the persons to be covered by the proposed tariffs, are the mere agents of the Objector governments” and that “all the copying to be licensed is undertaken for purposes of the work performed by the employees on behalf of the Objector governments,” the correct “user” whose perspective should be considered at this stage is that of the government.
[134]
[229] According to Access, in comparing this matter to that in Alberta, the analog to the students (in Alberta) are the governments, while the analog to the teachers, who are “merely an instrument for distributing copies,” are the governments’ employees. In comparing this matter to Society of Composers, Authors and Music Publishers of Canada v. Bell Canada,
[135]
Access argued that the employees are like the music stream providers and the government is like the music consumers. Following this line of reasoning, Access concluded that it is only the employers’ (i.e., the provincial and territorial governments) purposes that are relevant in this case.
[136]
It went on to assert that governments cannot engage in news reporting, criticism, review, or in private study. The only purpose that should be considered in the matter, it argues, is research.
[137]
[230] We disagree. It is not the employer of the person who makes a copy of a work that must be the “user.” In Alberta, even though the teachers were agents of their schools, this did not result in the purpose to be considered to be that of the schools, it was still that of the actual individual who used (in that case, read) the work: the student. Nor is it that since governments are often the beneficiaries of the dealings performed by the employees that they must be the user through whose eyes the purpose will be considered.
[231] In CCH, the Supreme Court of Canada held that the copying and transmission of works carried out by the librarians of the Great Library for lawyers was done for a permitted purpose. This was so even though the librarians did not perform the research. It was the lawyers, sometimes acting as an agent of their client in a particular capacity, who performed the research for the purpose of providing legal advice, while the beneficiaries of that advice may have sought and received it for a myriad of reasons (e.g., court proceedings, business activities).
[232] As the lawyer’s clients were in CCH, so the provincial and territorial governments, and in certain situations, their citizens, are the beneficiaries of the activities that were accomplished as a result of the dealings in question. In CCH, the Supreme Court did not consider the purpose for which a client asked a lawyer to provide legal advice for them; it looked only at the user, the person making use of the copy, to determine whether the first portion of the test was met.
[233] Similarly, we do not consider the purpose for which the Objector governments have their employees carry out their duties, except perhaps to the extent it may be evidence of that employee’s purpose of a particular dealing. Instead, we consider the purpose of each of the users, the person who actually used the work by reading it or otherwise perceiving its contents (usually an employee of one of the Objectors), for which the dealing was carried out. As there is no reason why an employee could not have dealt with a copyrighted work for purposes other than research, we do not exclude from our consideration the possibility that fair dealing occurred in relation to works for purposes other than research.
[234] Lastly, the Volume Study did not indicate the employer-Government’s purpose for which a particular copy was made.
ii.
Interpreting the scope of purposes
[235] In CCH, the Supreme Court of Canada stated that “[t]he fair dealing exception, like other exceptions in the Copyright Act, is a user’s right. In order to maintain the proper balance between the rights of a copyright owner and users’ interests, it must not be interpreted restrictively.”
[138]
It went on to conclude that “‘[r]esearch’ must be given a large and liberal interpretation in order to ensure that users’ rights are not unduly constrained.”
[139]
[236] Access urged the Board to interpret this latter statement as meaning that a “large and liberal interpretation” is to be applied only to the term “research,” but not to the other purposes enumerated in sections 29 and 29.1 of the Act.
[140]
[237] The use of a “large and liberal interpretation” is not restricted to certain aspects of the Act. It is an interpretative principle codified in the Interpretation Act, where section 12 states that “[e]very enactment is deemed remedial, and shall be given such fair, large and liberal construction and interpretation as best ensures the attainment of its objects.”
[141]
[238] In Warman v. Fournier,
[142]
Justice Rennie interpreted the Supreme Court’s statements in CCH as meaning that “fair dealing purposes (in that case, research) ‘must be given a large and liberal interpretation in order to ensure that users’ rights are not unduly constrained.’”
[143]
He then went on to apply a large and liberal interpretation to the purpose of “news reporting.”
[239] The Federal Court of Appeal in Alberta acknowledged that all fair-dealing purposes, including private study, are to be given “a large and liberal interpretation.”
[144]
This finding was not disturbed by the Supreme Court of Canada on appeal, where the Court stated that “the allowable purposes must be given a ‘large and liberal interpretation.’”
[145]
[240] Lastly, there is no apparent reason why “research” must be given a large and liberal interpretation in order to ensure that users’ rights are not unduly constrained, while other purposes, such as news reporting or private study, would not be subject to such a consideration. The Supreme Court stated that fair dealing is a user’s right, not that only fair dealing for the purpose of research is a user’s right. Therefore, all of the purposes enumerated in sections 29-29.2 of the Act must receive a large and liberal interpretation.
iii.
Purpose (Step 1) - Is the dominant purpose a permitted purpose?
[241] Access argued that in evaluating the first step of the fair-dealing test, all purposes of the dealing should be considered. By casting it as a question of evidence, it seeks to introduce a “predominant purpose” analysis into this step by importing the analytical framework used in assessing the fairness of the purpose of the dealing factor in the second step:
Assuming there is a recognized fair dealing purpose(s) applicable to the government, then the Board turns to the second step of the analysis […] [C]ourts should attempt to make an objective assessment of the user/defendant’s real purpose or motive in using the copyrighted work.
[…]
As explained above, the data from the study establishes that the governments and their employees have demonstrably ulterior purposes when they copy copyright protected works: 68% of the volume associated with one of the five enumerated fair dealing purposes relied upon by the Objectors has a non fair-dealing purpose as a predominant purpose […] if the only evidence before the Board of the purpose of the copying is a ticked box in a survey, then that is insufficient to meet step 1 of the fair dealing test, particularly if the other ticked boxes reflect a different predominant purpose.” [146]
[242] In other words, Access argued that the analysis of the “purpose” factor in step two of the fair-dealing test should inform whether the dealing was “truly” being done for a permitted purpose in step one.
[243] We disagree with this approach. Unlike some other exceptions to copyright infringement in the Act (e.g., section 30.61, “reproduce the copy for the sole purpose of obtaining information; section 30.63, “for the sole purpose […] of assessing the vulnerability of the computer”), fair dealing need not be done for a “sole purpose.” Nor does the Act require that fair dealing be “mainly,” “chiefly,” or otherwise “predominantly” done for the enumerated purposes. Instead, “the [Supreme Court] in CCH created a relatively low threshold for the first step so that the analytical heavy-hitting is done in determining whether the dealing was fair.”
[147]
[244] In its K-12 decision, the Board stated that it does not agree with the proposition that “[i]f the predominant purpose is not an allowable one, the exception would not apply, even if the dealing is fair and incidentally for an allowable purpose,”
[148]
explaining that this would render superfluous the analysis of the dealing’s purpose within the discussion on what is fair.
[149]
[245] Indeed, in the CCH and Bell decisions, the fact situations were such that it was possible to identify additional purposes for which the dealing was done (CCH: provision of legal advice, Bell: purchase of music downloads). This did not result in the first step of the fair-dealing test not being met.
[246] In assessing step one of the fair-dealing test, it is not an obstacle that a dealing is done for multiple purposes, as long as it was also done for a permitted purpose. Therefore, even where a dealing is not done predominantly for an enumerated purpose (but is actually done for an enumerated purpose), it will meet the threshold of the first step of the fair-dealing test. The effects of these other purposes will be considered in the “fairness” analysis in step two of the test.
[247] Access further argued that the ultimate purpose of any copying by an employee of the Objector governments can only be to carry-out the administration of government: “Employees and the other persons covered by the Proposed Tariffs […] are public servants and their only purpose is […] to ‘conduct government business and deliver government programs and services.’”
[150]
[248] We disagree with this approach as well. Research, for example, carried out for the further purpose of conducting government business remains research, and such a dealing would pass the first step of the fair-dealing test.
iv.
Education
[249] On November 7, 2012, the Act was amended by the coming into force of certain provisions of the Copyright Modernization Act.
[151]
In particular, section 29 was amended to state that “[f]air dealing for the purpose of […] education […] does not infringe copyright.”
[152]
[250] In our view, it is not necessary to consider the scope of the term “education” for the purposes of this Tariff. First, copying events for which an employee marked “education” as a purpose in relation to making photocopies of a work for multiple persons appear to us to represent, more likely than not, a scenario equivalent to that in Alberta: copies are being made by one person for the private study of another, and thus also meet the “private study” purpose. This is especially true given that education was almost never indicated as the sole purpose, and was often accompanied by other purposes, such as “future reference” or “research.” The private study nature of this activity is not altered by the fact that it was facilitated by a government employee instead of a teacher. Secondly, in situations where an employee indicated “education” as a purpose and made copies for herself, this meets the “private study” purpose, and potentially that of “research” as well.
[251] This approach is in-line with a fair-dealing test where the first step presents a relatively low threshold.
[153]
As discussed in part XII.C.3, above, it is not an obstacle that a dealing is done for multiple purposes, as long as it was also done for a permitted purpose.
[252] Therefore, given the cases considered for fairness, it was not necessary to decide what the scope of “education” in section 29 of the Act is.
v.
Conclusion
[253] The survey asked the employees: “Taking a liberal interpretation of the following purposes, did you copy the [work] for any of the following purposes?” Each of the following purposes could be answered with a “Yes,” “No,” or “Don’t know” response:
- For use in legal proceedings
- To serve the public interest by providing access to information
- In response to a formal access to information request
- For comments
- For criticism
- For review
- For news reporting
- For private study
- For research
- For your own interest
- To file it for future reference, to keep track of it
- To use in a presentation
- To preserve the original
- For others’ interest
- To educate or train others
- To conduct testing
- For administration
- For promotion or marketing
- Other, please specify.
[154]
[254] For most transactions, the employee completing the survey identified more than one purpose. This is not surprising given the form of the question. A person following such instructions will reasonably find overlap between the possible purposes such as: “for research” and “to file it for future reference” and “for your own interest.”
[255] As long as a dealing was performed for at least one of the fair-dealing purposes enumerated in the Act, it passes the threshold first-step of the fair-dealing test.
[256] Furthermore, it is apparent from the answers provided in the Volume Study that employees did not interpret the offered purposes in the legal manner they would be interpreted under the Act. In many instances, we found it more likely than not that the employee either interpreted these purposes as being related to themselves (e.g., education of oneself, which is private study; review or consideration of an article, which is research or private study), or made the copy in preparation of criticism or comment (i.e., the work was not used directly in the criticism or comment, but in the preparation or research for the criticism or comment).
[257] Therefore, the purposes indicated by an employee were used as evidence of the purposes of the dealing, but we considered these answers, along with the other information provided by the employee, to determine whether the first step of the test was met.
D.
Fair dealing – Step 2: Fairness
i.
Goal
a.
Goal - The fairness of various goals
[258] “The first factor identified in CCH is the purpose of the dealing, where an objective assessment is made of the ‘real purpose or motive’ behind using the copyrighted work.”
[155]
[259] Some confusion may have arisen from the fact that the English text of the CCH decision refers to two different parts of the fair-dealing test as the “purpose” of the dealing: the purpose considered in the first step of the test, and the purpose factor considered in the second step of the test. This nomenclature appears to have led Parties to make arguments that are applicable to the first step when discussing the second step, and vice-versa. For this reason, inspired by the phrase “le but de l’utilisation” used in paragraph 54 of the French version of the CCH decision,
[156]
we find it preferable to use the expression “goal of the dealing” when referring to the first factor of the second step in English.
[260] Access argued that the Objectors’ employees perform the vast majority of their research, criticism, private study, etc., in order to “conduct government business and deliver government programs and services.”
[157]
[261] The Government of BC did not contest that this is the job of the Objectors’ employees.
[158]
As it stated,
[i]t is not surprising that in many cases where a fair dealing purpose was indicated (taking research as an example), “for administration” was also checked as another purpose, perhaps the predominant purpose, given that for many government employees “administration” is the predominant purpose for all their activities. That does not diminish the fairness of the dealing for research purposes. [159]
[262] Access, on the contrary, argued that the presence of these other goals tends to make the dealing unfair.
[160]
[263] Access’ view is not supported by the Supreme Court of Canada’s fair-dealing decisions. In CCH, the Supreme Court explicitly contemplated the situation where research was being performed for the benefit of a person other than the user. In that case, research was being performed by lawyers for the benefit of their clients. The goal for which the research was performed did not change the fact that research was performed. In fact, the Supreme Court explicitly stated in CCH that “[r]esearch for the purpose of advising clients, giving opinions, arguing cases, preparing briefs and factums is nonetheless research.”
[161]
And while research done for commercial reasons may be less fair than research done for non-commercial goals,
[162]
the Supreme Court nevertheless held that the research was fair in that case.
[264] The evaluation of this factor involves considering the fairness of the goal for which the permitted (under the first step) activity (e.g., research, private study) took place. In CCH, this involved examining the fairness of research for the goal of providing legal advice in a commercial context. In Bell, it was the fairness of research for the goal of deciding whether to purchase musical works online. The fact that research or another permitted activity are undertaken for a further or additional goal does not, in itself, make this factor tend towards unfairness.
b.
Goal - The public interest
[265] The Consortium submitted that the Board should consider the fact that the dealings in the Volume Study are being copied in the “public interest” as an additional factor in the evaluation of the fairness of a dealing.
[163]
Access argued that no such “public interest” consideration exists.
[164]
[266] In CCH, when considering the “nature of the work” factor, the Supreme Court endorsed the views that “[i]t is generally in the public interest that access to judicial decisions and other legal resources not be unjustifiably restrained,”
[165]
and that “some dealings, even if for an allowable purpose, may be more or less fair than others; research done for commercial purposes may not be as fair as research done for charitable purposes.”
[166]
These statements convey the idea that some dealings may be more in the public interest than others, and that this may be a relevant consideration in the fairness analysis.
[267] Furthermore, the list of factors identified by the Supreme Court in CCH is not an exhaustive list, and fairness is a matter of impression. We agree with Prof. D’Agostino that
[p]arties pleading fair dealing, and courts ultimately deciding those cases, should exercise flexibility when interpreting fair dealing: raise factors germane to the case and assess evidence to support them. Whether there are six factors, or seven factors, or four factors should not be the driving preoccupation. [167] [emphasis added]
[268] Therefore, whether a dealing is in the public interest may be another factor to be considered, to the extent it has not been considered elsewhere, in the evaluation of fairness.
[269] In this matter, we find it unnecessary to evaluate this aspect of the dealing as a separate factor, as it can be evaluated as part of the “goal of the dealing” fairness factor. While there is flexibility in determining which factors are most germane in a given case, evaluating the same consideration through the lens of several factors risks making that consideration appear to have more weight than it otherwise should.
[168]
c.
Goal - Ulterior motives of the governments
[270] In claiming that the presence of other goals makes the goal factor tend to unfairness, Access asserted that the Governments are “hiding” behind the employee’s permitted purposes.
[169]
Given Access’ characterization of the Governments as users and employees as copiers, this would mean that the user cannot hide behind the shield of the copier.
[170]
However, this makes little sense as the goals of the user would be at the forefront of any analysis, and there would be little opportunity to “hide.”
[271] Even once Access’ characterization of the Objectors is rejected, as we have done above, and the survey respondents/employees are viewed as the copiers/users, Access’ statement amounts to the proposition that a payor of a tariff cannot have a separate unfair goal and hide behind the shield of the copier/user.
[272] There is no case-law to directly support either version of this proposition. According to the Supreme Court of Canada, “if [...] the copier hides behind the shield of the user’s allowable purpose in order to engage in a separate purpose that tends to make the dealing unfair, that separate purpose will also be relevant to the fairness analysis.”
[171]
[273] However, since fairness is a matter of impression, the goals of persons other than the copier may tend to make the user’s otherwise fair dealing unfair. For example, it may be possible that in certain circumstances the goals of the payor of a tariff so corrupt the goal of the copier/user, that it cannot be said that the goal for which the copier/user deals tends to be fair.
[274] That being said, and as described above, there is no evidence in this matter to suggest that the Objectors have some separate unfair goal which they are attempting to accomplish behind the veil of their employees’ otherwise fair dealings. The Objectors will pay the royalties on behalf of their employees so that the employees may perform activities, such as research and private study, in the course of carrying out their jobs. The payment of royalties by the Objector governments is a function of legal principles of liability of the Crown for its employees, as well as one of efficiency. The fact that it is a government that is paying the royalties does not create some separation of goals; nor does “[t]he fact that governments employ people to engage in a fair dealing purpose […] disentitle them from the benefit of fair dealing or change the “real purpose” of the dealings.”
[172]
[275] Lastly, even if it were possible to characterize the Objectors’ goals as separate from those of the employees, it is still the case that such a separate goal may tend to make the dealing unfair.
[173]
It follows that there may be separate goals that do not affect the fairness of the use, or perhaps even enhance it. Therefore, even if it were separate, the goal of government administration would not tend to make the Objectors’ employees’ dealings unfair.
d.
Goal of the dealing - Conclusion
[276] Contrary to the approach argued for by Access, we do not consider the presence of more than one goal, even one not enumerated in sections 29-29.2, identified by an employee for a copying event, to automatically tend to make that dealing less fair. Fair dealing may be done for multiple goals. Furthermore, the “number” of goals is merely a function of the design of the Volume Study, and the interpretation that the employees gave thereto. The presence of multiple goals may provide a better picture in what context a copy was made, but does not make a dealing less fair; nor does the presence of multiple goals listed in sections 29-29.2 make a dealing more fair.
[277] The Objector governments employ people in a wide range of positions: cook, enforcement officer, judicial clerk, librarian, lifeguard, liquor store clerk, nurse, physician, programmer, and social worker are just some of the positions that employees held.
[174]
As such, the particular contexts in which employees made copies of works captured by the Volume Study is also extremely broad. Since it would be impractical to become acquainted with the tasks of every job-category held by a person completing Phase II of the Volume Study, we accept that where an employee indicated that the copy they made was entirely or partially job-related, that it was, indeed, made in the furtherance of a valid goal related to their employment.
[278] Where a dealing was related to a person’s government employment, and where the goals identified by the employee provided support to conclude that the objective goal, such as research or private study, was to enable or facilitate the provision of government services, this factor would make the dealing tend to be fair. That is not to say that copying not done for the goal of employment cannot be fair. However, given the nature of the Volume Study, there was often insufficient information to determine the goal of the copying where the employee indicated that the copying was not work-related.
[279] Some employees entered comments in the “Other, please specify” field. Comments such as “This was a required textbook for a business course I am studying” (event # 22), “Continuing Education requirements” (event # 64), or “Ministerial briefing” (event # 279), provide insight into the context in which those transactions occurred. Indeed, some of these comments suggested that the copying was actually done for purposes related to employment, even if the person had indicated otherwise.
[280] In this manner, for a few of copying events for which the goal was indicated as not being related to employment, the evidence was sufficient to suggest a goal that tends to fairness.
ii.
Character of the dealing
[281] In assessing the character of a dealing, courts must examine how the works in question were dealt with. For example,
if multiple copies of works are being widely distributed, this will tend to be unfair. If, however, a single copy of a work is used for a specific legitimate purpose, then it may be easier to conclude that it was a fair dealing. If the copy of the work is destroyed after it is used for its specific intended purpose, this may also favour a finding of fairness. It may be relevant to consider the custom or practice in a particular trade or industry to determine whether or not the character of the dealing is fair. [175]
[282] The Supreme Court of Canada confirmed in Alberta that the quantification of any dissemination may be considered in this factor.
[176]
a.
Character of the dealing - Width of distribution
[283] The Consortium compared the copying captured by the Volume Study to that at issue in Alberta, and argues that since, in the latter case, the Supreme Court did not consider making a copy for each student in a class in that matter to be “wide distribution,” neither should the distribution of copies to a defined number of government employees be considered a “wide distribution.”
[177]
[284] However, in Alberta, the Supreme Court only stated that such a consideration should not be double-counted.
[178]
No explicit finding on the fairness of this factor was made by the Court. Access is correct when it notes that the majority of the Court
simply found that the Board improperly conflated the “character of the dealing” and “amount of the dealing” factors by referring to the total quantity copied when considering the latter rather than the “proportion between the excerpted copy and the entire work” (para. 29). The “character of the dealing” factor was not otherwise considered by the Supreme Court. [179]
[285] The Consortium also asserted that since the Supreme Court has stated that “the purpose of the fair dealing exception […] is to ensure that users are not unduly restricted in their ability to use and disseminate copyrighted works,”
[180]
it follows that “fair dealing is broader than particular uses of copyrighted materials for a specified fair dealing purpose—it also gives users the right to disseminate materials.”
[181]
[286] The above statement is only true to the extent it is understood as saying that fair dealing gives users the right to disseminate materials in order to enable a permitted fair-dealing purpose. In the context of the Great Library’s dealings in CCH, the Supreme Court of Canada explained that a person may disseminate a work in order to enable one of the permitted fair-dealing purposes to occur, even if the disseminator does not, herself, perform the permitted purpose. This does not mean, however, that dissemination outside of one of the permitted purposes can fall within sections 29 or 29.1 of the Act.
[287] That being said, in evaluating the fairness of a particular dealing with a particular work, neither is it correct, as Access proposed, to count the aggregate number of copies disseminated of all works (presumably by a licensee or group of licensees). It argued that
under the character of the dealing factor, the Board is also to make a quantitative assessment based on the aggregate use and the overall quantity of what is disseminated. Using Mr. Gauthier’s estimates (Phase 1), provincial and territorial government employees copy more than 34 million pages of published works in Access Copyright’s repertoire per year. Even using the Whitehead-Wilk estimate, for Phase 1, the “potentially compensable” volume is approximately 15 million copies per year. This aggregate volume of copying by provincial and territorial government employees is patently unfair, and favours Access Copyright. [182]
[288] According to this argument, the fairness of any particular dealing by an employee of a provincial or territorial government is to be considered as a part of all the dealings done by all employees of all provincial and territorial governments. In the context of a tariff, this would mean that the aggregate dealings of all payors of a tariff are the valid measure of the character of every individual dealing.
[289] This approach ascribes the characteristics of the whole to its constituent elements without sufficient justification. The fact that, in total, the copies made by all users that benefit from the tariff ended up in the hands of many people does not automatically mean that each of those dealings was an instance where a “wide” dissemination occurred. Even where, on average, dissemination may be wide, it does not follow that all of the individual dealings should automatically be considered to have a wide dissemination.
[290] We therefore disagree with this “aggregate” approach of determining the character of any particular dealing. While such an approach may have some validity in the case where an objector or defendant attempts to demonstrate that they deal fairly as a matter of practice (as was done by the Great Library in CCH), this approach becomes suspect when the inquiry is into whether a particular dealing was fair or not. In the latter case, only the dissemination of the copied work by the user is the relevant quantity to consider.
[291] Moreover, this approach would mean that dealings that are individually fair could become unfair once they are all done, even by different persons, simply by reason of their total quantity. At some point during the otherwise fair dealings, one would finally be unfair, thereby making all the previous dealings also unfair. In this matter, the dealings of one user should not tend to make the independent dealings of another user less fair. Obliquely, the concern being captured by Access’ approach to evaluating the character of the dealing relates to the effect of the dealing on the work, an aspect that is considered in its own factor.
[292] Therefore, in order to avoid the above results, in circumstances where we consider an individual dealing (as opposed to a practice) for fairness, the dealings with other works, and especially by other users, are not considered in the “character of the dealing” factor.
[293] In addition, we refer to the Supreme Court of Canada’s decision in Bell, where the Court stated that
given the ease and magnitude with which digital works are disseminated over the Internet, focusing on the “aggregate” amount of the dealing in cases involving digital works could well lead to disproportionate findings of unfairness when compared with non-digital works. If, as SOCAN urges, large-scale organized dealings are inherently unfair, most of what online service providers do with musical works would be treated as copyright infringement. This, it seems to me, potentially undermines the goal of technological neutrality, which seeks to have the Copyright Act applied in a way that operates consistently, regardless of the form of media involved, or its technological sophistication. [183]
b.
Character of the dealing - Destruction of the copies
[294] Access pointed to the fact that for a significant amount of the copying in the Volume Study (86.5 per cent, according to its analysis), the survey taker indicated that they kept a copy in their paper or electronic files, rather than destroying them. It argued that this tends to make the dealings unfair.
[184]
[295] Contrary to what Access suggested, CCH does not state that not destroying a copy after it is used favours a finding of unfairness – but only that destroying a copy favours a finding of fairness.
[296] The reason why destruction may favour fairness is hinted at in CCH, where the Supreme Court of Canada concluded that the Great Library’s “policy provides reasonable safeguards that the materials are being used for the purpose of research and private study,”
[185]
as well as in Bell where “[u]sers did not get a permanent copy, and once the preview was heard, the file was automatically deleted from the user’s computer. The fact that each file was automatically deleted meant that copies could not be duplicated or further disseminated by users.”
[186]
In both of those cases, the “character” of the dealing helped ensure that the work was being used for the permitted purpose, and not dealt with otherwise unfairly.
[297] In other words, destruction of a copy may favour fairness where the copy would no longer be necessary to achieve the permitted purpose, and destruction helps ensure that it is not used for other, unfair, purposes. This does not mean that non-destruction of the copy will always favour unfairness. Where destruction does not help ensure that the copy is used for a permitted purpose, or where destruction would undermine the very reason why it was made, non-destruction of a copy may not have any effect on the evaluation of the “character” factor.
[298] Indeed, in many dealings, such as for news reporting or criticism, destroying the copy of the work would run counter to the very purposes permitted in sections 29 to 29.2. Targets of a criticism that dealt fairly with a work could rest easier, knowing that the critical piece (or at least the portion which reproduces the work in question) had to be destroyed after some initial period.
[299] This is especially so where legislation or valid policy prevents a person from destroying the copy, as appears to be the case in relation to many copies captured in the Volume Study. As raised by the Government of BC, “[i]f a published document becomes a government record […] there are legislative and policy requirements that dictate when it can be destroyed, which dictate in favour of the retention of documents.”
[187]
c.
Character of the dealing - Conclusion
[300] Where the character of the dealing helps ensure that the work will be used for a permitted purpose, this would tend to make a dealing fair; where works are unnecessarily kept, or distributed unnecessarily, and where such acts risk that other unfair dealing will occur, this would tend to make a dealing unfair.
[301] For example, by its very nature, news reporting and criticism may result in works being widely disseminated; as another example, providing 40 copies to 40 participants in a course may require making 40 copies of a work. Penalizing such dealings for achieving their goal would go contrary to the enumerated purposes of the fair-dealing provisions.
[302] For some of the events in the Volume Study, the employee appears to have destroyed the copy, which further tends to make the dealings fair. That being said, it will often be reasonable that in the course of research or private study, such copied material is held for prolonged periods. Research can be a process of minutes, days, or years. Discarding information before the process is complete is usually not a viable option.
[303] Where there is a real risk that the copies made under a fair dealing may later be used unfairly, the concern that the copy be destroyed is heightened. In the context of government employees, and in particular in those cases where a single copy was made, this risk does not appear to be realistic.
[304] We further note that the Volume Study questionnaire was typically completed contemporaneously with the making of the copy. It is unclear whether an employee would have indicated what she expects to happen with the copy or copies, or what has actually occurred. In the latter case, it remains possible that the copy was destroyed after it was used for its intended purpose.
[305] We therefore find that, in this case, where a limited number of copies were provided within a government, they are not being sufficiently “widely distributed” as to tend to make the dealing unfair. However, where many copies are distributed outside of the government, this tends to make the character of the dealing less fair.
iii.
Amount used
a.
Amount used - How to measure?
[306] In evaluating this factor, “[b]oth the amount of the dealing and importance of the work allegedly infringed should be considered.”
[188]
“It is an examination of the proportion between the excerpted copy and the entire work, not the overall quantity of what is disseminated.”
[189]
[307] Access correctly noted that the Consortium presented irrelevant information for this factor when it states that the copying is proportionally fair because, using the total number of exposures for Phase II copying events, only “13.31 pages from a journal, 6.95 pages from a magazine, and 1.98 pages from a newspaper [are made] for an average copying event.”
[190]
Since “exposures” are a measure of the total number of pages copied (i.e., the product of original pages copied times the number of copies made), this measure only indirectly provides information about the amount of the work copied.
[191]
Access is also correct that the “work” to be considered is “the journal article, the magazine article and the newspaper article, not the entire journal, magazine, and newspaper issues containing the copied articles.”
[192]
[308] However, the “Average Number of Original Pages Copied” table presented by Access
[193]
is also not relevant in determining whether a particular dealing captured by the Volume Study was fair. Since the fairness of particular events is under consideration, as opposed to the practice of the Objectors as a whole, the amount of the dealing factor should be individually analyzed for each event.
b.
Amount used - Does the context affect the analysis?
[309] In Bell, the Supreme Court of Canada did not disturb the Board’s finding that streaming 30 seconds of a four-minute musical work was a “modest” dealing, “when compared to purchasing the whole work [approximately four minutes] for repeated listening.”
[194]
[310] Access submitted that Bell can be distinguished on the basis that “[e]phemeral ‘previewing’ of a work is not the context of this case.” According to it, the employees in this matter are “simply block-copying works for their direct consumption—as corroborated by the fact that 86.5% of such employees keep a copy in their files.”
[195]
Access further argued that the “amount” analysis in this matter can be distinguished from Alberta (because textbooks were purchased in the case), and CCH (because there was a copying compliance policy).
[311] We agree that the context in which the copies were made in Bell is different from that of the Volume Study. However, in Alberta, the Supreme Court of Canada cautioned against considering the same elements under more than one factor, or “double counting”, and thereby giving it more weight.
[196]
Accepting Access’ implicit argument that the amount used in Bell, 30 seconds, was fair because of the character of the dealing (the copy was destroyed after use), would amount to double counting, letting the evaluation of the character of the dealing determine the fairness of the amount of the dealing.
[312] In CCH, the Supreme Court of Canada stated that
[t]he amount taken may also be more or less fair depending on the purpose. For example, for the purpose of research or private study, it may be essential to copy an entire academic article or an entire judicial decision. However, if a work of literature is copied for the purpose of criticism, it will not likely be fair to include a full copy of the work in the critique. [197]
[313] Therefore, while the “amount of the dealing” analysis may be contextualized by the purpose for which the dealing was done (e.g., was it a necessary amount?), the amount should not become more or less fair because of the fairness or unfairness of the other factors. Under this approach, the Board’s statement, adopted by the Supreme Court of Canada in Bell,
[198]
could be understood as saying that the amount was fair, given the purpose for which the dealing was done – not because it was done in a context where other factors led to fairness.
[314] For example, research for the purpose of determining whether to purchase a book could require a smaller amount of the work to be dealt with (e.g., 10 pages may be enough to identify and get a sufficient sense of a book), than research for the purpose of providing legal advice, where every last word may be critical. This is why the importance of the work to the dealing is also considered with this factor.
c.
Amount used - Should and can the amount used be evaluated qualitatively?
[315] Access stated that “the survey data establishes that 96.7 per cent of the volume identified [as possibly being fair] involves the copying of more than two pages from a book or more than one page from a journal article, magazine or newspaper.”
Access argued that both of these types of activities would tend to make the dealing unfair.
[199]
[316] It is true that the copying of an entire page of a magazine or newspaper could sometimes amount to copying the entire work. However, it is less clear that it “will [often] be the entire work.”
[200]
Dr. Whitehead testified that newspaper articles “might be one page, but in most cases they are split between two or three pages,” except in the case of a pdf which is often one page; statements on which Access itself relied.
[201]
Therefore, the copying of a single page of a print-format newspaper article may not result in the copying of the entire work; in most cases, it would amount to one-third or one-half of the source work. It is also true that the copying of entire works tends to make the dealing unfair.
[317] However, Access’ proposition that copying more than two pages from a book will also tend to make the dealing unfair
[202]
requires greater scrutiny. Based on the “admission” by Dr. Whitehead that “the most important part (qualitatively) of a newspaper article is often the first few paragraphs,” Access claimed that this is also true of magazine articles and chapters of books and that these parts are the portions that were most likely copied in the Volume Study events. It then implicitly assumes that whether the copied portion conveys important information is what is to be measured in an analysis of “amount” of the dealing factor.
[318] Firstly, insufficient evidence was adduced to permit us to conclude that magazine articles and chapters of books are structured in the same way as newspaper articles.
[319] Secondly, as discussed above, the Volume Study does not let us determine which portions of a work were copied. The assertion that the copying in the Volume Study was mostly of the first portions of a magazine article or chapter of a book is not supported by evidence.
[320] Thirdly, it is not readily apparent from the Supreme Court of Canada’s recent fair-dealing case law to what extent the qualitative aspect of the amount of the dealing is even to be considered in assessing the “amount of the dealing” factor. In CCH, Alberta, and Bell, only the quantitative aspects were considered in this factor. Notably, many previews in the Bell case would have likely included the “hook” of a song, arguably the most qualitatively important portion of a song.
[321] It is possible that in Alberta and Bell the “amount of the dealing” factor was analyzed from a purely quantitative aspect due to the fact that those cases considered numerous dealings with a multitude of works and there was no evidence of the qualitative aspects of the portions that were dealt with in those matters. If this is the case, then the Board is in a situation similar to that of the courts (and itself) in Bell and Alberta: namely, it does not have evidence about the qualitative aspects of the amount of the dealing.
[322] Fourthly, even if evaluation of the “amount of the dealing” factor should consider the qualitative aspect of the amount copied, the importance of the information conveyed may not readily equate to the amount of skill and judgment used in expressing the portion that was copied.
[323] In Warman,
[203]
Justice Rennie, possibly in obiter reached the opposite conclusion, stating that the beginning of the newspaper article copied in the case “contained mostly facts and did not contain most of the original commentary by the author” and that the amount of the dealing was therefore “very limited.”
[204]
The portion that was copied in that case may have contained the most important information, but may have contained the least skill and judgment out of the entire news article.
[324] Therefore, even if we were to accept Access’ assertions regarding the structure of magazine articles and book chapters, and the parts of a work that are most likely copied, we would not conclude that this copying would always be an amount such that this factor would tend to make the dealing unfair.
[325] For these reasons, in this matter, we evaluate the “amount” of the dealing based on the quantitative evidence present in the Volume Study.
d.
Amount of the dealing - Conclusion
[326] In CCH, the Great Library provided a service whereby it would fax copies of works to lawyers who requested them. The Great Library’s policy indicated that it would
“typically honour requests for a copy of one case, one article or one statutory reference” and that it would “review requests for a copy of more than five per cent of a secondary source and that, ultimately, such requests may be refused.” [205]
[327] Amounts of up to 5 per cent were ordinarily accepted, while requests for amounts greater than 5 per cent were accepted or rejected on a case-by-case basis. The Supreme Court of Canada was of the view that this suggested that the dealings were fair.
[206]
[328] In Bell, online music services provided streamed previews of musical works to potential customers. In considering the amount of the dealing, the Supreme Court of Canada cited, and did not disturb, the Board’s finding that “a preview of about 30 seconds was a modest dealing ‘when compared to purchasing the whole work [approximately four minutes] for repeated listening.’”
[207]
The Supreme Court, stating that “a preview consists of an extract taken from the work, usually 30 to 90 seconds of a musical track”
,
[208]
went on to find the provision of such previews fair dealing.
[209]
A 30-second preview represented, on average, a 12 per cent portion of the entire musical work.
[329] In its K-12 decision, the Board had concluded that
teachers generally comply with the conditions of the pan-Canadian licence, which sets limits on how much can be excerpted from a work. The licence is certainly more generous than the Great Library’s policy, which generally sets a limit of five per cent of a work, while the pan-Canadian licence allows up to ten per cent. This being said, nothing leads us to conclude that the copies at issue tend to approach the upper limit imposed by the licence. What is more, provided, once again, that these copies are specifically requested by the student, it does not appear to us that this difference is sufficient to render the copies unfair. [210]
[330] In short, copying more than 5 per cent, but less than 10 per cent of a work, did not have the effect of making an otherwise fair dealing unfair.
[331] Based on a qualitative consideration of the portion of the work copied, Justice Rennie in Warman found that the reproduction of a heading and three and a half paragraphs from a newspaper article of eleven paragraphs (being approximately 32 per cent) was a “very limited” amount of dealing.
[211]
He went on to conclude that even though there was arguably an alternative to the dealing, and that the excerpts were widely distributed on the internet, that dealing was fair.
[212]
[332] In Century 21 v. Rogers,
[213]
Justice Punnett, considering the copying of entire real-estate listings and their associated photographs, stated that
the repeated daily access and indexing of such information militates against a defence of fair dealing. This is not a situation of a one-time copy being taken. It is conduct consisting of repeated actions by the defendants. In my view the amount of the dealing exceeds what is fair. [214]
[333] Justice Punnett found that the repeated copying of the entirety of the works made the amount of the dealing factor tend towards unfairness, while leaving the door open to a different conclusion in the case of a one-time copy.
[334] The CCH, Warman, and Century 21 decisions were proceedings for infringement of copyright, while Bell and K-12 were tariff certification proceedings.
[335] While, as a general principle, bright-line rules with specific numbers may not be desirable in the evaluation of fairness, since an evaluation is to be a matter of impression, in the context of analyzing a Volume Study, the use of such a rule or rules may be unavoidable. The impression that we can form is based on the data of the Volume Study, and in order to deal consistently with similar copying events, bright-line rules will almost unavoidably arise. The larger the size of the sample of copying events to evaluate, the more likely this will be true.
[336] We are aware that the application of such rules may sometimes under-include and sometimes over-include events. Were more data available, it may reveal that some events captured by a rule should not have been, while others not so captured should have been. While a rule that errs as little as possible in this manner is desirable, perfection is not possible. For the purposes of setting a royalty rate, a rule that can err on both sides (and as little as possible given the data available), will best establish a fair and equitable royalty rate.
[337] As with any other factor, a finding that the amount of the dealing tends towards fairness or unfairness is not determinative of the evaluation of the dealing. Were it otherwise, the fair-dealing factors would be converted from a “useful analytical framework”
[215]
into a list of requisite elements.
[338] In light of the decisions discussed above, we consider that, for the purposes of establishing a royalty rate in this Tariff, copying within limits similar to those identified in the cases above (being approximately 10 per cent of a book), in the context of research or private study, either makes a dealing tend towards fairness, or, at most, does not make it tend to either fairness or unfairness (i.e., this factor is neutral in such cases and favours neither finding).
[339] Furthermore, given the limitations of the evidence present in the Volume Study, we accept as reasonable that an entire newspaper, journal or magazine article may have to be copied for the purpose of research or private study. While the copying of an entire work may tend to make a dealing less fair, the Supreme Court of Canada noted in CCH that, for some purposes, such as research and private study, it will often be the case that the entirety of the work is necessary to effectively achieve that purpose, something that need not be the case for criticism or news reporting, for example.
[216]
[340] Therefore, for events where an entire article was copied for the purpose of research or private study, while the amount of the dealing factor tends towards unfairness, it does not do so strongly. Where the person copied only one page and additionally indicated that it was interested in less than a page, there is a possibility that she have copied less than a page, but was unable to indicate this in the electronic questionnaire. For such copying events, the weight of this factor is further reduced, making it tend to neither fairness nor unfairness.
iv.
Alternatives to the dealing
[341] Where there are reasonable alternatives to a dealing, that dealing may be less fair. For example, if there is a non-copyrighted equivalent of the work that could have been used instead of the copyrighted work, this should be considered. Furthermore,
[i]t will also be useful for courts to attempt to determine whether the dealing was reasonably necessary to achieve the ultimate purpose. For example, if a criticism would be equally effective if it did not actually reproduce the copyrighted work it was criticizing; this may weigh against a finding of fairness. [217]
[342] However, the alternative must be realistic,
[218]
and must not simply be the “[t]he availability of a licence.”
[219]
[343] Access’ interpretation that the availability of any alternatives always results in the dealing tending towards unfairness is overly simplistic: as the Supreme Court stated in CCH, “[a]lternatives to dealing with the infringed work may affect the determination of fairness.”
[220]
[emphasis added] Therefore, the mere existence of alternatives need not always affect the determination of fairness.
[344] The Consortium claimed that “it is clear that for most government purposes, there is unlikely to be a non-copyright equivalent of a given work.”
[221]
It implicitly argues that the main consideration in evaluating this factor should be whether non-copyrighted material could be used instead of the works that were copied. This approach however, would limit the consideration of this factor to only a single possible alternative. The use of a non-copyrighted work is an example of an alternative that is to be considered, not the only alternative to be considered.
[345] According to Access, the Volume Study “demonstrates that for 91.8% of the volume identified as possibly fair, there were alternatives to the copying;”
[222]
and this tends to make the dealings unfair.
[346] This figure is based on its analysis of Question 54 of the Volume Study, which asked: “If copying of this [work] had been prohibited by law, what would you have done?”
[223]
The genre of the work would be displayed in the questionnaire as the type of document. For example, a person who had entered that they are dealing with a book would have seen “If copying of this book had been prohibited by law, what would you have done?”
[347] The options that employees were offered as possible answers were
I would have requested the purchase of one or more copies of the [document type]
I would have acquired the one-time right to make a copy of the [document type]
I would have referred others to the [document type] without sending them a copy
I would have simply made do without copying the [document type]
I would have copied the [document type] anyway
I would have used part of a different [document type] that was not protected by copyright, such as a Canadian English language government document
Other, please specify [224]
[348] These choices were not exclusive; an employee could select one or more of these responses. As such, some responses are difficult to interpret in a way that is consistent. For example, for 21 copying events, the employee indicated that they would have copied the work anyhow, as well as that they would have made-do without copying the work.
[349] The Volume Study also does not permit evaluating whether the responses simply reflect all possible alternatives, each with some probability of occurring, or whether the responses are to be interpreted as a series of possible alternatives (in unknown sequence) that the employee would have attempted.
a.
Would have copied the work anyway
[350] Only if the employee selected “copied [it] anyway” and no other alternatives, did Access interpret the response to this question to mean that there was no alternative available.
[351] However, there are several issues with this interpretation of the results. Where a person selected “would have copied the [work] anyway” along with several other alternatives, it is reasonable to interpret this as a situation where the person would have attempted other alternatives before copying the work anyhow. Thus, there is still some, unspecified, likelihood that the person would have “copied it anyway.”
[352] Even if we were to accept Access’ assertion that the availability of any other alternative makes this factor tend towards unfairness, Access still understates the amount of persons who would have “copied it anyhow.” A total of 48 responses indicated “copy it anyway” as a possible alternative.
[353] Furthermore, the manner in which the question itself is formulated is problematic for the evaluation of this fairness factor. Instead of attempting to discern which of the proffered alternatives would reasonably achieve the goal for which the dealing was done, the question begins with the premise that copying is not available – or, at least, illegal. Thus, respondents to the question were, according to the question, selecting alternatives that were second-best.
[354] The Supreme Court of Canada has stated that it is useful to consider whether the use of a work was reasonably necessary to achieve a dealing that is equally effective.
[225]
Contrary to the implication by Access,
[226]
willingness to infringe copyright in the course of one’s employment is not the test for whether something is “reasonably necessary.” Such a level of “necessity” appears not to be “reasonable,” but approaches one of strict necessity.
[355] In the context of CCH, where lawyers were requesting copies from the Great Library, possible alternatives that lawyers would have indicated in a similar survey could have been “drive to the Great Library” or “make do without copying.” However, such responses would not have shed light on whether they were reasonable alternatives or whether the dealing would be as effective without a copy being made. It is not possible to equate the unwillingness of a person to commit an act prohibited by law with the availability of a reasonable and practical alternative to the dealing.
[356] Therefore, we do not automatically assess responses which did not indicate a willingness to commit an act prohibited by law as those where the copying was not “reasonably necessary.” However, in those cases where a person did indicate that they would have copied the work anyhow, despite the hypothetical illegality of the act, we conclude that the other hypothetical alternatives proffered, even if selected by the employee, were sufficiently poor alternatives for the copying in question that the person was willing to commit an illegal act. In such cases, we find that it is more likely than not that there are no reasonable alternatives to the dealing.
b.
Purchase one or more copies of the work
[357] Where a copy of a work has already been purchased, it may not be realistic to expect that a copy be purchased for every person who seeks to make a copy thereof. This was explained by the Supreme Court of Canada in Alberta, where schools had “already purchased originals that are kept in the class or library, from which the teachers make copies.” The Supreme Court held that “buying books for each student is not a realistic alternative to teachers copying short excerpts to supplement student textbooks.”
[227]
[358] Similarly, this alternative may not be realistic in those cases where the relevant government already owns copies of the work. While a few employees used the “Other [alternative], please specify” field to mention that their department or institution already owns a copy of the work, information on this issue was not systematically collected by the Volume Study.
[359] Furthermore, there was little evidence as to whether the works in question would be readily and reasonably commercially available in those instances where the employee indicated that they would request the work to be purchased. While it may be possible that some of these works could have been readily purchased for a reasonable price, and in a reasonable format (e.g., similar to the format of the source copy of the dealing, without additional encumbrances, such as technological protection measures), such evidence was not generally available.
[360] Therefore, while we do not exclude the possibility to purchase a copy entirely, we are aware of the limited evidence on this point.
c.
Acquire a one-time right to make a copy of the work
[361] The availability of a licence to copy a work
is not relevant to deciding whether a dealing has been fair […] If a copyright owner were allowed to license people to use its work and then point to a person’s decision not to obtain a licence as proof that his or her dealings were not fair, this would extend the scope of the owner’s monopoly over the use of his or her work in a manner that would not be consistent with the Copyright Act’s balance between owner’s rights and user’s interests. [228]
[362] Thus, the option of acquiring a one-time licence was not counted as a valid alternative.
[363] Given our conclusion on the making of digital copies, we do not need to consider whether this principle applies in situations of digital sales of works where the contract purports to provide a mere licence. In any case, in the context of the Volume Study, it is very unlikely that an employee would have made such a distinction, and would have indicated “the purchase of one or more copies” for such an activity, a response that we do include as a possible alternative.
d.
Made do without copying the work
[364] Given that an employee could select more than one alternative, and that it was possible to select this alternative along with others, there is ambiguity whether it represents an alternative where no copying is done and no further action is taken, or whether it overlaps with other options that do not involve copying the work, such as referring others to the work. Analysis of the responses suggests that employees interpreted this answer in both the former and latter manner.
[365] Even in cases where this was the only alternative selected, two possible interpretations remain. Marking this option could indicate that the other options presented were not realistic, or as effective as making a copy of the work, and an employee was not willing to commit a legally prohibited act; or it could indicate that the copying of the work did not add significantly to the effectiveness of the dealing.
[366] In many cases where a person makes-do without copying the work, none of the goals sought by the Act, the creation and dissemination of works, is supported. Taken to an extreme, if all not-strictly-necessary uses were avoided, neither the users nor copyright owners would be better off. Users would be deprived of any benefit of making a copy of the preferred work, and copyright owners would not receive any royalties, nor the intangible benefits that may be associated with having their works disseminated. The result is deadweight loss
[229]
and should generally be avoided.
[367] However, where the dealing is equally effective without making a copy of the work, not making that copy does not significantly increase deadweight loss. As such, while not using the work is not, by default, a reasonable alternative, it may be so where the use of the work adds little or nothing to the effectiveness of the dealing.
[368] As such, while we do not entirely exclude the possibility of making-do without as a possible alternative, we are aware that it may often not be a reasonable one.
e.
Refer others to the work without sending a copy
[369] Referring others to the work could be done in a variety of ways, such as verbally informing another person about the work, including a citation to the work, or sending a URL to the work. The response does not indicate by which manner such a reference would be done.
[370] Unlike the situation where the work is not used at all, the sending of a link to the work in question is more likely to be a valid alternative. Furthermore, in an on-line context, directing another user to the work may generate revenue for the copyright owner (e.g., from advertisements).
[371] That being said, there may be situations where uncertainty over the impermanence of a work posted on-line may not make this a viable alternative, given that the content of on-line works can be modified or removed at any time and it may be important to retain a copy of the text as it existed at a given point in time.
f.
Use part of a different work that was not protected by copyright
[372] The option to use a work not protected by copyright does not appear to be realistic in most circumstances. Such a work would typically be around 90-110 years old (time between authorship and death + 50 years). Based on the topics of the works in the survey, it is unlikely that many such older works would be relevant to dealings being performed by the employees.
[373] Unfortunately, even the formulation of the question is problematic, as it suggests that “Canadian English language government document[s]” are not protected by copyright. While certain governments may grant broad permission for the use of their documents, it is not true, as a general proposition, that such documents are not protected by copyright.
g.
Alternatives - Conclusion
[374] As mentioned in part XII.D.4, above, it is not appropriate to interpret responses to the question “If copying of this [work] had been prohibited by law, what would you have done?” as representing the availability of reasonable alternatives. The presence of the qualifier that the copying is prohibited by law generally makes the copying unavailable. Faced with such a case, a person may undertake alternatives that are much less efficient, reasonable, or suited for the purpose to be achieved.
[375] Therefore, we could not use the presence of responses to definitely ascertain the availability of reasonable alternatives that would meet the needs of the dealing as well as making a copy of the work. However, the presence of responses to such a question can still be used to inform us how likely it is that a reasonable alternative was available to the person making the copy.
[376] As such, we find that the more “alternatives” that were selected, other than “acquire a one-time licence” and “copy anyhow”, the more likely it is that one or more of the proffered alternatives were actually reasonable, and the more this factor would tend to make the dealing unfair. Where none or only a few were selected, this would tend to make the dealing fair, or not make the dealing tend in one direction or another. Where a person indicated that they would have copied the work despite the prohibition, we interpreted this to mean that the other proffered alternatives were not reasonable in the context of the copying event, and thus this factor would tend to make the dealing fair.
v.
Nature of the work
[377] As the Supreme Court of Canada stated in CCH,
[t]he nature of the work in question should also be considered by courts assessing whether a dealing is fair. Although certainly not determinative, if a work has not been published, the dealing may be more fair in that its reproduction with acknowledgement could lead to a wider public dissemination of the work—one of the goals of copyright law. If, however, the work in question was confidential, this may tip the scales towards finding that the dealing was unfair. [230]
[378] Access argued that the inverse of this statement is also true. In fact, it goes even further: it does not merely claim that dealing with a published work “may” make a dealing less fair, but that dealing with published works tends to make the dealing unfair.
[231]
[379] This argument ignores the comparison that the Supreme Court was making in that case: namely, between those works that should be publically disseminated, and those for which there is a valid interest in restraining such dissemination.
[380] The Consortium, on the other hand, eschewed the published/non-published/private distinction, and instead characterized the nature of all the works in the Volume Study as those that are “analyzed and used by government employees in the public interest,”
[232]
arguing that this tends to make the dealings fair. It compared this situation to that in CCH where the Supreme Court held that “the nature of the works in question—judicial decisions and other works essential to legal research—suggests that the Law Society’s dealings were fair.”
[233]
The Court agreed that “[i]t is generally in the public interest that access to judicial decisions and other legal resources not be unjustifiably restrained.”
[234]
[381] The Consortium’s argument amounts to the circular claim that because the works were used by government employees during the work, it shows that they must have been of a kind that get used by government employees. It would also appear to double-count the purpose of the dealing in assessing the nature of the work, by claiming that since they are typically used “for the public good,”
[235]
their nature tends to fairness – an approach that, as we have discussed above, has been criticized by the Supreme Court of Canada.
[236]
[382] The vast majority of the copied works were newspaper articles; or academic, scientific, or professional journal articles. Dissemination of works is one of the Act’s purposes, and the dissemination of the kind of works captured in the Volume Study will almost always be in the public interest. However, in our fairness analysis, the benefits of such dissemination are already captured in our consideration of the fairness of the “purpose” of the dealing.
[383] Generally, the natures of the works in this matter do not tend to make the dealing more or less fair. On the one hand, they are published works, and are not of a nature where further dissemination without the dealing is unlikely. On the other hand, they are not private writings where such dissemination could be undesirable.
[384] In one case (event # 180), the book from which copies were made appears to be a collection of crossword puzzles. While not a strict rule, the copying of works that are, by their very nature, consumed and discarded upon use can tend to make a dealing less fair, and in this case, it does.
vi.
Effect of the dealing on the work
[385] The effect of the dealing on the work is another factor that can be considered when attempting to determine whether a dealing is fair. If the reproduced work is likely to compete with the market of the original work, this may suggest that the dealing is not fair. However, “[a]lthough the effect of the dealing on the market of the copyright owner is an important factor, it is neither the only factor nor the most important factor that a court must consider in deciding if the dealing is fair.”
[237]
[386] Access argued that
[t]he evidence available from the 2011 Study strongly suggests that the copying of works by provincial and territorial government employees competes with the market for the original works. As an example, every time a provincial or territorial government employee copies a newspaper article, he does not buy either the subscription to the paper or the one time right to licence it. This reduces licensing and subscription revenues for the newspaper publisher. [238]
[…]
Here […] there is no paucity of evidence on the likely effect on the market. Over the past decade, the market has been generating opportunities for organizations and individuals to easily obtain licences on their own to copyrighted material—licensing mechanisms that would be undermined if the Objectors can exempt all their copying under fair dealing. The 2011 Study questionnaire contains the respondents’ own answers on how they would act if they could not simply make copies, and many indicate alternatives such as purchasing a one-time copy or referring others to the works. The Objectors’ dealings circumvent the market licensing schemes in place today, and, thus, the Board is justified in drawing an inference that the effect on the dealing in the marketplace is negative. [239]
[387] Access further argued that, “the 2011 Study establishes that for 91.8% of the volume, there were alternatives to copying the work.” In effect, Access argued that since the analysis of “the alternatives to the dealing” would, in their view, tend towards unfairness, the analysis of the “effect on the market” must also tend towards unfairness.
[388] Access’ approach relies on the premise that every dealing with a copyrighted work will be one where there was an opportunity for the copyright owner of the work to sell a one-time right to license that use. If this factor were evaluated on these grounds, every dealing would “compete with the market for the original work.” It is likely for this reason that the Supreme Court of Canada in CCH stated that the availability of a licence to copy a work is not relevant to deciding whether a dealing is fair.
[240]
[389] Access also attempted to harness other factors to support the inference that these dealings must have a negative effect on the works. Indeed, factors such as amount of the dealing, character of the dealing, and alternatives to the dealing may be intertwined with the effect of the dealing.
[390] However, attempting to determine the likely effect of copying based on evidence already considered in other factors, such as the character of the dealing, the amount that is copied, or alternatives to the dealing, would likely result in the kind of “double counting” that was criticized by the Supreme Court of Canada in Alberta.
[241]
As discussed previously, the effect of such an approach would be to give a particular consideration more weight than it otherwise should.
[391] The Consortium, on the other hand, argued that at issue is whether the market
for publishers’ works had decreased as a result of copying by provincial and territorial governments, and in particular, as a result of the copies in issue here. The evidence before the Board has failed to show that the market for publishers’ works had decreased as a result of copying by provincial and territorial governments. [242]
[392] While it is true that Access did not adduce evidence to directly establish that the copies in issue had this effect, it is not automatically the case that this means that “this factor therefore tends towards fairness.” Instead, as in previous factors, where there is insufficient evidence to examine a particular factor we conclude that the analysis of this factor favours neither a finding of fairness nor unfairness.
[393] Given that there is no direct evidence that would permit us to ascertain with any certainty the effect of the dealing captured by the Volume Study on the works that were reproduced, and given that relying on aspects that have already been considered under other factors could have the effect of erasing proportionality from the fairness analysis, we find that this factor neither tends to make the dealings in the Volume Study more fair nor less fair.
E.
Fair dealing – Findings
[394] As stated above, the fair-dealing factors are not criteria or elements – they are factors. That is, they are useful considerations in a fair-dealing analysis, not conditions that must be met for a dealing to be fair. It is also important to note that since fair dealing involves the weighting of factors, the presence of a single factor that would tend to make a dealing unfair does not automatically make that dealing unfair.
[395] While these factors assist in the determination of fairness, fair dealing is a matter of impression. In matters such as the one before us, that impression has to be made on the basis of evidence that may not be as detailed as in situations where only one or several dealings are at issue, as could be the case in an infringement proceeding. Certain evidence may be partly or entirely lacking. As such, the impression that is to be formed will be based on the evidence that is available.
[396] Due to the nature of the Volume Study, there were fundamental difficulties in interpreting the data it generated. For example:
Where more than one copy of a work was made, were all copies made for all the purposes noted in the recorded event, or were some copies made for some noted purposes, and other copies for other purposes?
Where a work was emailed, was one of the recipients (or the only recipient) the employee making the copy? Many office scanners send the resulting digital file by email to the employee.
Where a person indicates a purpose such as “research,” and where that employee emailed a copy of the work to another person, was the email sent for the research purpose of the recipient?
[397] In considering the data from the Volume Study, we were aware of these and similar limitations, and generally favoured simpler interpretations over more complex ones (e.g., all copies were made for the same purposes).
[398] We analyzed each of the remaining 136 events for fair dealing individually. Of these, 22 had no evidence of a purpose for which fair dealing is permitted. These were excluded from consideration for fair dealing, and are thus compensable for the purpose of this Tariff.
[399] Sixty-seven events had very similar characteristics and this “category” of events can therefore be approximately described. These events were indicated as being for research (including future reference) or private study (including education), or for the purposes of criticism, review, comment, or news reporting, but without any incorporation into another work – and therefore interpreted as being for research or private study (likely in preparation for criticism, review, etc.).
[400] The amount copied was either of an entire article, or less than 10 per cent of a book. Only a single copy of the work was made in these events. The single copy was not made by posting the work on an intranet, the internet, or sent by e-mail or fax. The purpose was indicated to be work-related, or that no copy of the work was kept – thus making a wider unfair distribution very unlikely. Furthermore, any distribution of the copy was within the same government where it was made. Lastly, the employee indicated only a few “alternatives” that they would consider if the dealing were otherwise illegal.
[401] We find these 67 events to represent fair dealings for the purpose of establishing a royalty rate for the Tariff.
[402] The remaining 47 events are more difficult to describe as a generally uniform group. Of these, the Objectors accepted 5 events as being compensable.
[243]
We therefore consider those events as compensable, leaving 42 for consideration.
[403] Applying the legal conclusions, principles and approaches we have identified above, and congruent with the approach already taken in relation to the 67 events above, we conclude that of the 42 remaining events, 30 events represent fair dealings for the purposes of this Tariff. Hence, the total number of events found to be fair dealing is 97.
[404] We believe it is useful to provide some examples in order to demonstrate the manner in which these 42 copying events were analyzed.
[405] For example, in event # 8, a person indicated that she made 3 copies of an entire 8-page journal article by e-mailing, saving, and printing it, retaining 2 copies for herself; provided one copy to someone within the same Branch; made the copies mainly for her own interest (along with private study, review [without incorporation of the material in a review], and future reference); that it was in part work related; did not distribute the copy outside of the government; that she would have (if copying were illegal) attempted to obtain a one-time right, referred to it without making a copy, made without copying, and copied it anyhow.
[406] In this case, the combination of the purposes let us conclude that the dealing was for work-related research or private study. While the employee copied the entire article, this can be necessary for such activities. The dissemination in this case was narrow. Finally, while the employee indicated several potential alternatives, she also indicated that she would have copied the article anyhow. This leads us to conclude that the other alternatives were likely not good, reasonable alternatives. Furthermore, the availability of a licence is not a relevant alternative. Neither the nature of the work (an article from a published journal), nor the effect of the dealing (unknown) tend to make the dealing more or less fair. In this case, the analysis of the fair-dealing factors leads us to conclude that the copying event represents fair-dealing copying.
[407] As another example, in event # 9, a person indicated that she printed and saved 2 pages of a magazine article of unknown length, retaining both copies for herself; did not know whether she distributed a copy to another person; that she did so for private study, education, to serve the public interest, own interest, future reference, and others’ interest; that it was not work related; that she did not distribute the copy outside of the government; and that she would have (if copying were illegal) referred to it without making a copy, made-do without copying, and used a different source.
[408] In this case, the combination of purposes considered in context with the other responses (e.g., “to serve the public interest,” despite no copies being distributed outside of the government and not being work related) makes it difficult to ascertain the actual goal of the dealing. The breadth of the dissemination, while likely narrow, is not known. While the amount of the dealing is not known exactly, it would likely be the entire article, as this was the case for the vast majority of events in which articles were copied in the Volume Study. Furthermore, the number of potential alternatives was sufficient to let us conclude that there likely was a reasonable alternative available. Neither the nature of the work (an article from a published magazine), nor the effect of the dealing (unknown) tend to make the dealing more or less fair. In this case, the analysis of the fair-dealing factors leads us to conclude that the copying event does not represent fair-dealing copying.
[409] Finally, it is important to remember that the finding of fairness in relation to a particular representative event is not the same as a finding that the event that was captured by the 2011 Volume Study was fair dealing. Rather, it is a finding that the event represents a certain volume of copying that occurred during the Tariff period that we expect to be fair. While this distinction may appear nuanced, it is important.
XIII.
NON-STATUTORY DEFENCES
[410] The Objectors submitted that there are defences to copyright infringement beyond those set out in the Act. Access responded that since “copyright law is a creature of statute, only Parliament has the prerogative to identify exceptions to copyright protection.”
[244]
[411] The Act states that “[n]o person is entitled to copyright otherwise than under and in accordance with this Act or any other Act of Parliament.”
Notwithstanding this restriction on the manner in which copyright-like rights may be created,
[245]
it is possible that there are defences not set-out in the Act, such as equitable remedies, that may be available.
[412] Common-law defences remain applicable in other areas of law that have been the subject of codification. For example: despite the existence of certain enumerated defences in the Ontario Libel and Slander Act,
[246]
and similar legislation, common-law defences remain available in that context. Similarly, codification in the Criminal Code
[247]
of indictable offences and of certain defences to those offences, along with the abolition of common-law offences, did not thereby remove all common-law defences not so codified.
A.
Public interest defence
[413] Citing two decisions from the United Kingdom, Beloff v. Pressdram Ltd.
[248]
and Hubbard v. Vosper,
[249]
the Government of BC submitted that there is some authority for a “‘ public interest defence’ to infringement where individual rights, including copyright, may be overridden by the public interest, where the matter is of a serious nature and in the national interest.”
[250]
However, it conceded that “[s]uch a defence would only apply in exceptional circumstances.”
[251]
[414] If such a “public interest” defence does exist, the limited evidence from the Volume Study does not lead us to conclude that any of the events were of such a nature as to benefit from such a defence.
B.
Use of works in legal proceedings
[415] The Consortium submitted that fair dealing may be done for the purpose of legal proceedings. In support of this proposition, it pointed to the decision in CCH, where the Supreme Court of Canada stated that the Great Library’s photocopy service policy, which was being evaluated for fairness,
does not allow all legal works to be copied regardless of the purpose to which they will be put. Requests for copies will be honoured only if the user intends to use the works for the purpose of research, private study, criticism, review or use in legal proceedings. This further supports a finding that the dealings were fair. [252] [emphasis added]
[416] However, the Consortium’s understanding of this passage may stem from the ambiguity created by using the term “purpose” in relation to both the first and second step of the fair-dealing test, as discussed above in part XII.D.1. It is likely that the Supreme Court, in considering the second step of the fair-dealing test, was of the opinion that the goal of using a work in a legal proceeding tended to make the dealing fair.
[417] It is possible that a fair-dealing defence may be successfully raised in relation to the use of copyrighted material in legal proceedings. For example, where parties to a proceeding require material for research (either for their own or for that of the decision maker), it may be fair for a party to make copies to permit such research.
[418] However, to the extent that the Consortium is arguing that there is a separate, common-law defence to infringement for the purpose of legal proceedings, the Consortium has not adduced sufficient evidence of its existence.
C.
Findings
[419] Given our conclusion regarding the non-applicability in this matter of non-statutory defences, this leaves 36 events as compensable for the purpose of establishing a royalty rate for this Tariff.
[420] The summary of our findings on the compensability of the Volume Study copying events appear in the Annex, Table 2.
XIV.
ECONOMIC ANALYSIS
A.
Two different approaches to determining royalty rates
i.
Introduction of the two approaches
[421] In the present case, Access and the Objectors propose two very different approaches to determining the royalty rate for the Tariff.
[422] Access proposes an approach called “fair market value” (FMV). As explained by Access’ expert Mr. Heys, FMV is “[t]he highest price available in an open and unrestricted market between informed and prudent parties, acting at arm’s length and under no compulsion to act, expressed in terms of cash.”
[253]
This is the standard definition of FMV; as such, we take no issue with it.
[423] As a set of possible proxies, Access proposed certain transactions between either itself or Copibec, and various provincial governments. Mr. Heys, applying the FMV approach, selected the transaction that, in his view, most closely met the definition of FMV and proposed it as the proxy to use in this matter.
[424] The Objectors proposed an approach called “volume times value” (VTV) in which the volume of copying is measured using a survey or census. Value is measured in cents per page, and is typically related to the retail price of the copied item. For each type of copied item (referred to in this case as “genre”), the value and volume are multiplied and the products are then summed across all genres.
[425] Both Access and the Objectors have commented on the methodology used by the other party. Despite the fact that the Parties agreed to conduct a study “to provide estimates of the amount of copying of published documents,”
[254]
Access argued that there is insufficiently good data to use VTV and that, in any event, FMV should always be preferred to VTV.
[255]
By contrast, the Objectors maintained that while FMV could be used in some circumstances where VTV was available, Access used FMV incorrectly.
[256]
ii.
An issue with using FMV
[426] In a competitive market, there are many buyers and many sellers for a particular good or service. Each of these buyers and sellers has an individual reservation price.
[257]
However, because of market clearing,
[258]
a single price usually emerges. This single price is the equilibrium price. The equilibrium price is the highest reservation price that any of the transacting buyers is willing to pay and simultaneously the lowest reservation price that any of the transacting sellers is willing to accept. This is also the FMV price.
[427] In a bilateral market, there is only one buyer and one seller for the good or service. Once again, each of these two has a reservation price. On the assumption that the buyer’s reservation price (the highest price he is willing to pay) is greater than the seller’s reservation price (the lowest price he is willing to accept), a transaction will occur. The transactional price is somewhere in the interval between the seller’s reservation price and the buyer’s reservation price.
[259]
Every price in that interval is an equilibrium price (including the transactional price). The FMV price, however, is the buyer’s reservation price.
[428] In a market with only one buyer and one seller, the FMV price is unlikely to be the transactional price. Furthermore, since the FMV price is the highest acceptable price, it is unlikely to be a fair and equitable price.
[429] We have concerns about one other aspect of the data put forward by Access. The licence data are few in number; as a result, they are somewhat “grainy” and less reliable for setting a tariff.
[430] For all the above reasons, we are not inclined to use FMV in the context of data from a small number of bilateral transactions.
[260]
However, notwithstanding these concerns, we will analyze how Access used the concept of FMV in practice.
iii.
FMV as used by Access
[431] As a practical matter, one typically determines the FMV of a good by finding and evaluating a set of market transactions (or proxies) for similar goods. To the extent that these market transactions deal with sufficiently similar goods, and meet the FMV conditions (as expressed in the definition of FMV), their price can be said to be the FMV price.
[432] In the abstract, the quality of the FMV calculation depends on the quality of the proxies. If no good proxies can be found, the FMV calculation will necessarily be poor.
[433] For the 2005-2009 portion of the Tariff, Access considered four possible proxies.
[261]
According to Access, the rates of the Saskatchewan and Alberta licences generated a proxy of $2.72 per FTE. The rates of the 1998 Ontario Licence generated a proxy of $3.03 per FTE. The 2004 Canada Licence rate generated a proxy of $8.40 per FTE.
[262]
Finally, the 2003 Quebec Licence with Copibec generated a proxy of $10.50.
[434] Access argued that neither the Saskatchewan and Alberta licences nor the Ontario licence could serve as an FMV proxy because the parties had limited information
[263]
and Access felt a compulsion to act. For the same reasons, Access further argued that the 2004 Canada Licence could not serve as an FMV proxy.
[264]
Having thus eliminated three of the four proxies it proposed, Access used as the FMV the proxy of $10.50 provided for by the remaining licence, i.e. the 2003 Quebec Licence.
[435] Access’ arguments for the 2010-2014 portion of the Tariff are similar. For this period, Access used its proposed FMV price of $10.50 for the 2005-2009 period and adjusted it for inflation.
[265]
As the Objectors pointed out, this is tantamount to using the 2003 Quebec Licence to set the prices for both Tariff periods.
[266]
[436] If we intended to use the FMV approach, we would find that Access was too quick to dismiss the three licences for the 2005-2009 Tariff period. First, the definition of FMV requires that the parties be informed and prudent. Mr. Heys stated that FMV does not require perfect information, but it requires more than the limited information the parties had when they negotiated the three licences in question. In particular, Mr. Heys is of the view that it requires information that might come from a copying survey.
[267]
We agree with Mr. Heys that perfect information is not required, but we disagree that a copying survey is required.
[437] The fact that no copying survey has taken place for the Alberta, Saskatchewan, and Ontario licences means that there is an asymmetric information problem, since the governments may know how much copying takes place but Access does not. However, this problem was not so severe that the transaction did not occur. It is likely the case therefore that the parties could still be classified as informed and prudent.
[438] Second, we find Access’ usage of compulsion incorrect.
[268]
Compulsion means that there is no reasonable alternative. Access was not compelled to transact with the governments of these three provinces; if it found the royalties too low, it could have simply not provided a licence to them. To the extent that these governments nevertheless copied works in Access’ repertoire, and did not obtain a licence for that copying from the rights holders, these rights holders could have pursued litigation against suspected infringers. It may very well be the case that Access’ best option was to negotiate with these three provinces, but so long as there are reasonable alternatives, there is no compulsion.
[439] We also find it inconsistent that Access used the FMV approach for the 2005-2009 period of the Tariff and then effectively abandoned it for the 2010-2014 period, by applying an inflation factor to the value obtained for 2005-2009. If FMV is the appropriate approach for setting these tariffs, and since proxies were available for both Tariff periods, it would have been normal to use FMV for both Tariff periods. It is regrettable that this fact was not clear in Access’ written evidence, but rather emerged as a result of vigorous cross-examination of Mr. Heys.
[269]
This is a key part of Access’ methodology; it should have been more explicit. In effect, all the prices for both Tariff periods are linked to a single agreement, signed between Quebec and Copibec in 2003.
[440] We now examine the 2003 Quebec Licence.
[270]
We know very little of the context of the licence. The licence itself is more than 10 years old and, although it was renewed subsequently, it reflects the negotiating period of 2003. It also reflects the bargaining power of the Government of Quebec and Copibec, about which we also have no evidence.
[441] An examination of the text of that licence reveals several clauses that were not present in the 2010 Proposed Tariff. First, there is an indemnity clause.
[271]
Access generally tried to value the indemnity clause it offered in its licences at 6.5 per cent of its benchmark licences or less.
[272]
But this is not an indication of the value of the Copibec indemnity clause. Even if the size of the repertoire of Access and Copibec is about the same, this does not necessarily imply that the value of an indemnity clause is the same for licences issued by both collectives.
[442] Dr. Dujsic, an expert called by the Consortium, explained that the indemnity clause is a contingent liability
[273]
and that the probability of a catastrophic event (in this instance, a lawsuit by a non-affiliated rights holder against an Access licensee) occurring needs to be assessed to value this clause. This however, cannot be done by looking at past uses of the clause alone.
[274]
[443] We agree with Dr. Dujsic. We would however add other factors that could affect the value of the indemnity clause, such as the propensity of non-Copibec-affiliates to sue, the jurisdictions in which those suits are launched, the tendency of courts in those jurisdictions to find in favour of rights holders and the amounts awarded by courts that find in favour of rights holders, as part of such a calculation of contingent liability.
[444] In addition, the parties agreed to split the cost of a volume study, but cap the cost to the Government of Quebec at $30,000.
[275]
The cost of a study for the Government of Quebec would certainly be much higher. For instance, the cost of the 2005-2009 K-12 Study was about $3,000,000.
[276]
Thus, it may have been worth it to Quebec to pay a higher per-FTE rate to avoid a very high study cost.
[445] Thus, for all the reasons above, we reject the use of FMV as applied by Access in the present matter.
[446] We turn now to some specific considerations relating to the VTV methodology.
iv.
VTV: Did Phase II fail?
[447] Counsel for the Objectors, in his opening statement made two points relating to VTV. First, it is the approach approved by the Board in the K-12 decision
[277]
. Second, almost 10,000 people participated in the survey that underlies the VTV calculation; discarding their efforts requires some justification.
[278]
We find both of these arguments unconvincing. Ultimately, the utility of VTV is an empirical question.
[448] In the abstract, the quality of the VTV calculation depends on the quality of the survey/census as well as the measurement of the prices. But since the prices of copied items are usually measured fairly easily (most of these items trade regularly in a retail market), the quality of the VTV calculation typically comes down to that of the instrument used to measure volume. This leads us to a consideration of the key question: did Phase II fail? In the preliminary discussions before conducting the Study, one of the concerns of Access was that Phase II would fail.
[279]
[449] As discussed above in part VI, Phase I was the recall phase of the Volume Study and Phase II was the logging phase of the Volume Study. Access characterized Phase II failure as the occurrence of some or all of the following criteria:
[280]
Low rate of acceptance of participation in Phase II;
Low rate of initiation of Phase II;
Low rate of completion of Phase II;
A number of events reported in Phase II that would be substantially lower than the number of events reported for the most recent 30 days in Phase I;
A much lower number of events reported in week 2 of Phase II compared to week 1;
Reporting time demonstrating that the reporting was not contemporaneous with the activity; and,
Large proportions of “don’t know” or “no answer” selections.
[450] We accept the characterization of Access for Phase II failure, without accepting either Access’ claim that Phase II was likely to fail (pre-survey) or that Phase II had failed (post-survey).
[281]
[451] In order to evaluate these issues, on December 4, 2012, the Board asked the Parties to provide the following information:
Of those invited to complete Phase II, 62 per cent (Group A) did so and 38 per cent (Group B) did not. For each Group A and B, please provide summary statistics (mean, standard deviation and number of respondents) for responses to the following two questions:
Q3_A. How many times, if any, would you say that you photocopied from a book, magazine, journal article or newspaper or asked someone else to do so on your behalf between 31 and 60 days ago?
Q4_A. How many times, if any, would you say that you photocopied from a book, magazine, journal article or newspaper or asked someone else to do so on your behalf in the past 30 days?
[452] Following the receipt of this information and analysis by Board staff, the Board sent a further notice to the Parties on March 14, 2013, stating further:
The purpose of [these questions posed on December 4] was to determine whether Phase II was biased and, if so, if the bias was correctable. Based on the answers supplied by the parties, Board staff examined the possibility of two biases – participation selection bias and completion selection bias.
Board staff examined these possible biases using a t-test. Based on this analysis, the Board’s tentative conclusion is that (a) there is no evidence of participation selection bias, either for photocopying, or for all methods of copying considered in combination; and (b) that while there is empirical evidence of completion selection bias (copiers who complete Phase II copied 5.75 per cent less than those who did not do so, on a weighted-average basis), it is possible to bring a correction to the Phase II data when using it to account for that evidence.
Attached to this Notice are the t-test analysis and the relevant Excel files. Parties may comment on the t-test, its relevance and reliability as well as on the Board’s tentative conclusions in this respect. [282]
[453] The Consortium accepted the tentative conclusions of the Board: that there is no evidence of participation bias, that there is evidence of completion bias and that the completion bias can be corrected by an adjustment of 5.75 per cent.
[283]
[454] Access did not dispute the calculations by Board staff but disputed whether the results of the t-test
[284]
are appropriate to determine whether Phase II has failed. In this regard, Access made several comments the Board should take into account:
There was low participation in Phase II;
Phase II can only err by underreporting;
Very few events were reported in Phase II;
Phase II has a very large margin of statistical sampling error; and,
It is not possible to correct the completion bias by augmenting the number of copies by 5.75 per cent. [285]
[455] We note that comments 1, 3 and 4 raise essentially the same point: the sample size is small. However, the small sample size is not surprising. Phase II was a double-recruited voluntary survey. Despite the fact that the survey was championed by each of the provincial governments, it was a voluntary survey and these are known to have low participation rates. If the participation rate is low, there will be a low number of events reported. If few events are reported, there is necessarily a large margin of error. Furthermore, as argued by counsel for the Objectors, the margins of error of the two phases are not directly comparable, because Access applied truncation methodologies
[286]
to Phase I but not to Phase II.
[287]
Furthermore, as noted by Dr. Wilk, imputation necessarily reduces the margin of error because it adds in data at the mean value.
[288]
[456] Moreover, whether or not the sample is small is not, on its own, relevant. What is relevant is whether the small sample is representative, since the sampling design was a simple random sample with two attractive design features. First, the initial recruitment sample (Phase I) was extremely large relative to the universe (more than one in six government employees were contacted). The larger the recruitment sample, the better chance the survey sample has at being statistically representative. Second, because detailed observations were kept relating to the respondents’ decisions to cease participating in the survey of respondents, observations can be reweighted to achieve a theoretical stratification.
[289]
That is, it is possible to account for the fact that the raw survey data are not representative of the size of the provincial workforces.
[457] Comment number 2 is simply wrong. In fact, it is possible for Phase II to err in both directions. Respondents can over report (by reporting the same transaction multiple times) or underreport (by not reporting a transaction). However, even if comment number 2 were true, it is irrelevant because it is unrelated to the question of whether Phase II actually failed. It is a design question for Phase II. If it were true that Phase II can only err in one direction, there is no need to collect data from Phase II to verify this. Thus, whether we view comment number 2 as an abstract comment or focus on the allegation by Access that the Consortium never showed evidence of Phase II erring in both directions, we would not come to the conclusion that Phase II failed in either case.
[458] Finally, the question of whether it is possible to correct the completion bias is ultimately one of judgement. We could identify the bias and not correct it, on the grounds that the bias is not quantifiable. Or, if it is quantifiable, we could correct the bias. In our view, the ability to quantify the bias, combined with the fact that the Objectors support the quantification of the bias is reason enough to correct it.
v.
VTV: Should we use the data from Phase I or Phase II?
[459] Prior to conducting the survey, the Parties made many arguments relating to recall-based surveys like Phase I and logging-based surveys like Phase II.
[460] As the Government of British Columbia details, recall surveys may be subject to telescoping (stretching the time frame of the valid period), or respondents may use a process of inference to get to their response.
[290]
This process of inference may have any or all of the following five components. The respondent
May not have taken in the information at the time;
May not be willing to make the effort to retrieve the information;
May recall instead generic information about the type of event, not the specific event itself;
May recall only partial information; and,
May recall erroneous information about the event. [291]
[461] As part of a larger filing on November 1, 2010 (prior to the conduct of the survey), the Consortium also cited substantial literature relating to recall-based surveys.
[292]
The Consortium noted that significant or dramatic events are easier to recall. They are more likely to have created memory cues and there is a greater likelihood that the respondent paid attention to the event at the time. These factors combine to make it more likely that the respondent remembers a large copying event than a small one. The Consortium also drew the Board’s attention to research that indicated that the accuracy of recall decreased with the length of interval between the event and the survey.
[293]
It also highlighted research that recall surveys often produce overestimates or underestimates, depending on the subject being surveyed.
[462] In Access’ filing on the same date, it noted that government employees may be able to refer to their records to mitigate recall issues.
[294]
Access disputed the Consortium’s claim that the last-copied-event, as documented in Phase I, is more likely to be a large one than a small one. In support of its claim, it mentioned that there is a large body of literature on memory effects and it filed a bibliography of such literature.
[295]
However, Access did not discuss this literature and its conclusions in its submission.
[463] The Parties replied to one another on November 15, 2010. In the Consortium’s reply, it noted the following.
[296]
First, Access has merely repeated its earlier claims without providing supporting evidence. Second, Access’ suggestion that respondents will search their records is new and unsupported; there is no mention of using one’s records in the invitation letter to respondents. Finally, many of the citations given by Access are obscure, irrelevant, or support the position of the Consortium with respect to recall.
[464] In its reply, Access noted (and gave citations for) a number of problems with logging-based surveys.
[297]
First, there is a risk of underreporting as respondents lose interest in the survey. Second, respondents tend to make their responses conform to norms of social desirability.
[298]
Third, behaviour may change as a result of the survey itself. Fourth, rare events are underrepresented by logging-based surveys. Finally, logging-based surveys can only be biased in one direction, namely underreporting.
[465] In its evidence filed after the Study had been completed, Access argued that:
[w]hen all of the above factors are taken into account, it is our professional opinion that the data produced by Phase I of this study are of value to inform the Copyright Board of Canada of the extent and nature of copying performed in the provincial and territorial governments involved. However, given the frailties of the Phase II data, we do not value them in the same manner. For these reasons, the analyses carried out in the next chapter are based on the Phase I data. [299]
[466] The Consortium drew the Board’s attention to the percentage of missing values
[300]
in Phase I and Phase II. For the variable “number of pages copied,” this was uniformly higher for Phase I than for Phase II.
[301]
As Dr. Whitehead explained: “[t]hat's not to say that the Phase 1 data are terrible, but what it says is that the Phase 2 data are of better quality.”
[302]
The Consortium summarized its conclusion in the report by Drs. Whitehead and Wilk: “[f]inally, we consider that the Phase 2 data are superior in type and quality to the Phase 1 data and deserve to be relied upon.”
[303]
[467] We have two legitimate sources of evidence and we must choose to rely on one or the other. The Consortium capably outlined the theoretical problems with Phase I in its letter of November 1, 2010; Access equally capably outlined the theoretical problems with Phase II in its reply of November 15, 2010.
[468] After the survey was conducted, the situation became even more complicated. Access stated that it would not base a tariff on the survey at all, but that if it did, it would use data from Phase I.
[304]
The Consortium stated that Phase II data were superior, but that Phase I data could also be used in some way.
[305]
[469] Ultimately, we find the logic of Exhibit Consortium-18 most compelling. If we use the data from Phase II, we get a better picture of the volume of copying by reason that there are fewer missing data.
[306]
The variable the Consortium highlighted, pages copied, is one of the key variables for determining volume.
[470] Based on all the above, we find it most appropriate to use data from Phase II. For these reasons, we asked the Parties
[307]
to do certain further calculations regarding Phase II data, as described in parts VI.D and VI.E, above.
B.
Volume considerations
i.
Whose calculations are more adequate
[471] On July 21, 2014, the Board sent a notice to the Parties, indicating that, in its preliminary view, 26 of the 291 events were compensable.
[308]
The Board asked the Parties to perform certain calculations and to explain those calculations, as well as supply their computer code.
[472] We have reviewed the submissions of Access and the Consortium and find the calculations of the Consortium more adequate.
[473] First, we prefer the annualization procedure used by the Consortium. Access annualized the data by multiplying the volume by 365 and dividing by 14.
[309]
This is tantamount to assuming that government employees copy every single day of the year. The Consortium annualized the data by multiplying by 49 and dividing by 2.
[310]
The Consortium’s explanation is that there would be an average of three weeks not worked per year, accounting for vacation time and statutory holidays. Admittedly, this approximation is not perfect, since different employees take different amounts of vacation time and statutory holidays differ by province. However, we find this calculation more convincing than that of Access.
[474] Second, Access failed to adjust for Phase II completion bias. The Consortium boosted its estimate of volume copied by 5.75 per cent to account for this bias.
[311]
This is consistent with the Notice of the Board of March 14, 2013, which indicated that it was possible to account for this bias quantitatively. In its response to the Board’s notice, Access indicated that there were further biases associated with Phase II data but
[312]
did not adjust its calculations for these other biases either.
[475] We would be remiss if we did not remark on the issue of quality of the two filings by the Parties in respect of the July 21 notice. Access failed to document its weighting procedures adequately. Access’ filing consisted of a single page of introduction, a single page of results, and terse, relatively uncommented SAS code. The Consortium filed a detailed explanation of its weighting procedures, as well as an Excel spreadsheet detailing its calculations. Furthermore, the filing by the Consortium allowed us to see the marginal contributions of each transaction to the total value of the Tariff, which we found very useful. We did not select the calculations of the Consortium because they are more explicit; that being said, having selected the calculations of the Consortium, we are pleased that their form is easy to use.
ii.
Compensable events
[476] We have reviewed the Volume Study copying events and find the 39 events listed in Table 3 of the Annex to be compensable. A summary of Table 3 is shown in Table 4. It is noteworthy that the weighted annualized pages corresponding to magazines are two orders of magnitude higher in 2010-2014 than in 2005-2009. This can be attributed exclusively to one event, number 285.
[477] Event 285 is a copying event from the Government of Nunavut. The document copied was entitled “Fuelling the young athlete.” While Access classified this document as being from the “other” genre, we have reclassified it as a magazine, since it was closest to this genre among the four genres under study. In this event, a two-page document was printed 300 times. Since printing was not a compensable method of copying in 2005-2009, a zero appears in the column entitled “# of pages 2005-2009” but 600 appears in the column entitled “# of pages 2010-2014.” Applying the Nunavut weight gives the amount in weighted pages, shown in the two rightmost columns.
[478] We can also express these amounts on a per-FTE basis. There were 121,022 FTEs in the provinces and territories that participated in the Volume Study; all Parties accepted this figure. Table 5 shows the numbers in Table 4 on a per-FTE basis. According to these calculations, a full-time government employee did an average of less than two (1.92) compensable copies of works in the repertoire of Access per year during the 2005-2009 period. This average increased to almost 22 (21.61) copies per year for the 2010-2014 period.
C.
The question of prices
i.
General
[479] The approach taken by Access in its statement of case did not address the question of prices at all, since Access used the FMV approach, not the VTV approach. In his reply report, Mr. Heys criticized some of the calculations made by the Objectors that form the basis of the VTV approach. Those criticisms form the core of Access’ position on the question of prices.
[480] One of the most interesting comments made by Mr. Heys concerned the non-linearity of the pricing function. The value of copied pages depends on the nature of these pages, how many pages are being copied, and how many copies of each page are being made.
[313]
This observation may be valid. The non-linearity may be upward sloping, in the sense that copying additional pages adds additional marginal value. Or it may be downward sloping, in the sense of decreasing marginal utility. We regret that Mr. Heys did not develop this idea further in his reply report. Perhaps the next time the government tariffs are examined, one party or the other will submit a proposal with this sort of non-linear pricing. Until that time, however, we adopt a price that remains constant with the number of pages copied from a work.
[481] The following sections set out our approach to prices in this Tariff. Summaries of our conclusions are presented in Tables 7 and 8 of the Annex.
ii.
Books
[482] On behalf of the Consortium, Nordicity collected prices on 19 books found to be compensable in the Whitehead-Wilk analysis. Nordicity described its methodology as follows:
To derive an average value per page for these titles, we searched for retail pricing information for each title on Amazon.ca. In cases in which we could not find pricing information on Amazon.ca, we also searched Indigo.ca and Amazon.com. In some cases, we also obtained retail pricing information directly from the publisher’s or author’s web site. […] For each title, we used the lowest available new-copy price to determine the per-page value. […] Where prices were quoted in United States dollars (e.g., Amazon.com), we converted amounts to Canadian dollars at an exchange rate of par. [314]
[483] Nordicity noted that the average book price (for the 18 titles for which pricing was available) was 13.46 cents per page for the period 2010-2014. Nordicity deflated this price by 8.7 per cent to account for general inflation during the Tariff period, obtaining a price of 12.29 cents per page. Finally, Nordicity multiplied both of these figures by 48.4 per cent to account for the creative contribution
[315]
of the books, obtaining 5.95 cents per page for the 2005-2009 portion of the Tariff and 6.51 cents per page for the 2010-2014 portion of the Tariff.
[484] On behalf of Access, NERA disputed the methodology used by Nordicity. NERA noted that Nordicity’s calculations were very sensitive to the assumption that the lowest new-copy price should be used. NERA redid the calculations using the list price. It obtained a price of 24.31 cents per page for the period 2010-2014, which translated into a value per exposure of 15.30 cents.
[316]
[485] The fundamental question raised here is which prices should be used: list prices or lowest (Amazon) prices? The Objectors note, quite correctly, that list prices are not transaction prices. Most transactions do not take place at list prices.
[317]
But this argument does not convince us to use Amazon prices. For one thing, Amazon prices are not necessarily transaction prices, either. To purchase a book at Amazon, a buyer pays the Amazon price plus a shipping fee.
[318]
In effect, customers buying books in a retail store pay the shipping fee as well; it is embedded in the retail transaction price.
[486] The ideal situation would be to use transactional prices, removing all costs not related to creative contribution. The difficulty with this approach is that we do not have information about book retailing costs in Canada. By contrast, the figures used in the K-12 decision for creative contribution were based on list prices and the Board’s understanding of the Canadian book-publishing industry. After considering all of the options, as we did in the K-12 decision, we adopt list prices as the starting point for our calculations, as proposed by Access. We thus use 24.31 cents per page as the starting point for calculating the value of a compensable exposure of books.
[487] We agree with the Consortium that there should be an adjustment for inflation and also agree with the Parties that the price per page should be subject to a 30 per cent selection premium. To account for the creative contribution, the Parties have been using the figures the Board used in its K-12 decision. We also use them.
[488] The final price per page for the 2010-2014 portion of the Tariff is set at 15.30 cents, as calculated by Access. This price is to be deflated by 8.7 per cent to 14.07 cents per page for the 2005-2009 period. See Table 6 in the Annex.
iii.
Magazines
[489] Nordicity began its analysis of magazine prices with the assertion that there is no reason to believe that the profile of magazines copied in provincial and territorial governments differs substantially from the profile of magazines copied in K-12 institutions. Accordingly, the starting point for the price for magazines is the price used in the 2005-2009 K-12 decision, namely, 2.70 cents per page. This price is increased by 9.8 per cent to account for inflation, taking the price for the 2010-2014 period to 2.96 cents per page. Finally, multiplying these prices by 26.9 per cent to account for creative contribution yields 0.73 cents and 0.80 cents for the two Tariff periods, respectively.
[319]
[490] Access did not challenge this pricing calculation.
[491] According to data filed by Access, magazine copying amounted to 11 per cent of copying in the Volume Study.
[320]
In the K-12 case, magazine copying amounted to 6.5 per cent, or a little more than half as much.
[321]
This calls into question the assertion that the two copying profiles are the same.
[322]
[492] When we examine the calculations done in the K-12 decision, it appears that most of the compensable copying from magazines captured by the survey used in that tariff was done from five magazines published by Rogers Media.
[323]
There were only five magazines compensably copied in the survey used in the present matter. None of them are published by Rogers.
[493] We do not accept the Consortium’s assertion that the profile of copying in the K-12 Study and the Volume Study are similar. However, Access did not offer an alternative. We therefore reluctantly accept the starting point of 2.70 cents per page offered by the Consortium.
[494] We are uncomfortable using the magazines and periodicals subgroup of the Consumer Price Index (CPI) to adjust the rates. In a decision relating to multiple SOCAN tariffs in 2004, the Board considered the question of whether to use the overall CPI or some subcomponents of the CPI for indexing SOCAN tariffs. It stated: “However, the Board believes it is preferable to use the overall CPI rather than certain subcomponents, as SOCAN proposes. Because the subcomponents proposed by SOCAN are based on a small sampling, they could be subject to substantial variations, both upward and downward.”
[324]
[495] The Board confirmed its confidence in the overall CPI in a decision relating to CBC Radio in 2011. There the Board stated that “[t]he CPI remains the best measure of inflation to preserve purchasing power, since it reflects the prices of the basket of goods and services purchased by Canadians on average.”
[325]
[496] We note that we are dealing with a seemingly different situation here than was the case in either in SOCAN Multiple Tariffs (2004) or CBC Radio (2011). In those cases, the Board dealt with indexation of the tariff itself. Here we are dealing with indexing of prices which are themselves parameters of the model which determines the tariff. However, because the VTV model is a linear model, the choice of indexation for its parameters is equivalent to the choice of indexation for the tariff itself.
[497] We concur with the decisions of the Board of 2004 and 2011: the appropriate measure of inflation is that created from the all-items-CPI. This is the approach the Consortium takes for books; in our view, the point is more general, and is applicable to the other genres captured in the Volume Study as well.
[498] By using the all-items-CPI, we inflate the2.70 cents for the 2005-2009 period of the Tariff to 2.93 cents for the 2010-2014 period. To account for the creative contribution portion of the value of the works, we then multiply these figures by 26.9 per cent to obtain 0.73 and 0.79 cents, respectively. Finally, we adjust by a further 30 per cent to account for the selection premium. This yields prices of 0.94 and 1.03 cents. These are the prices we use in this decision.
iv.
Newspapers
[499] The Consortium claims that the profile of newspapers copied in the 2005-2009 K-12 decision is similar to the profile of those copied in the present decision. The starting point is thus the 2.80 cents from the K-12 decision. This is increased by the newspapers-CPI to reach a rate of 3.45 cents. In this case, creative contribution is 34.6 per cent. This implies adjusted rates of 0.97 cents for the 2005-2009 period and 1.19 cents for the 2010-2014 period.
[326]
[500] Once again, Access did not present an alternative valuation scheme for newspapers.
[501] We suspect that these rates may be out of date or otherwise incorrect. However, in the absence of any other evidence, we accept the proposal of the Consortium, but adjust using the all-items-CPI. We also add a 30 per cent selection premium. The final prices are 1.26 and 1.37 cents, respectively.
v.
Journals
[502] The Consortium began by noting the relationship between journals and coursepacks, namely, that coursepacks are generally made up of journal articles and (portions of) books. As a result, it would be possible to infer journal pricing by looking at coursepacks. Its starting point was the 10.0 cents per page
[327]
that was charged for coursepacks to students at institutions that were members of the Association of Universities and Colleges of Canada (AUCC).
[328]
The Consortium then took note of the price of textbooks in the 2005-2009 K-12 proceeding, 9.2 cents per page, and argued that this is a reasonable proxy for the price of books in coursepacks.
[503] By using these two figures, 10.0 cents per page and 9.2 cents per page, it is possible to infer a price for journal articles in coursepacks, so long as one makes an assumption about the relative frequency of journals and books in coursepacks. The Consortium assumed that the relative frequency of journals and books in coursepacks was equal, that is, that each makes up 50 per cent of the coursepack. Using this assumption, the implicit price of journal articles in coursepacks is 10.8 cents per page.
[329]
Finally, the Consortium applied the CPI subcomponent magazines and periodicals to obtain the value for the 2010-2014 portion of the Tariff, namely, 11.85 cents.
[330]
[504] Access disputed several of the Consortium’s assumptions. In particular, it noted that its transactional licence prices range between 30 and 60 cents per page, depending on the use intended for the copy. As such, Access argues the per exposure value for journals should be no lower than 30 cents per page.
[331]
[505] We agree with Access that the Consortium’s starting assumptions are questionable. We do not see why coursepacks need consist of similar types of materials as copied in the Volume Study. Furthermore, we do not see why one need make the assumption that 50 per cent of material in coursepacks is from journals. With most journals available online at many universities, it is possible that few journal articles are available in coursepacks. In fact, it is possible that coursepacks consist mostly of materials that are not available online, such as excerpts from books.
[506] We find Access’ suggestion of a value of 30 cents per page reasonable and use it as the per exposure value. As a result, there is no derivation of this price in Tables 7 and 8. We use the price of 30 cents for the 2005-2009 period of the Tariff. Adjusting for inflation, we obtain the price for the 2010-2014 period, namely 32.61 cents.
D.
Calculating the royalty rates
[507] The calculations for the final royalty rates are shown in Table 8 for the period 2005-2009 and in Table 9 for the period 2010-2014. In both cases, and for each genre, the value of all compensable exposures is obtained by multiplying the value of a compensable exposure and the volume of compensable exposures. The values for each genre are then added to obtain the total value of compensable exposures. Once this total value is divided by the number of FTE employees, the final rate in pennies per FTE is obtained. As such, we obtain royalty rates of 11.56 cents for 2005-2009 and 49.71 cents for 2010-2014.
[508] These rates are lower than what Access was seeking ($10.50 for the first Tariff period, $11.70 for the second Tariff period) and higher than what the Consortium was offering ($0.01 for the first Tariff period, $0.07 for the second Tariff period). Several explanations are necessary.
[509] First, the Volume Study indicates that two sorts of copying are most common in government: printing from electronic files and making digital copies (scanning, emailing, or posting). As proposed by Access, the Tariff covers printing from electronic files only for the 2010-2014 period. Furthermore, for reasons set out earlier in this decision and that relate to the scope of Access’ mandate, the Tariff does not cover digital copies for all of its duration.
[510] This Tariff therefore only sets a royalty for the making of paper copies by employees of provincial and territorial governments; the making of digital copies remains an act for which this Tariff does not provide a licence. We estimate that, had Access been in a position to authorize the making of digital copies without requiring their deletion, the resulting per-FTE annual rate would have been significantly higher, possibly as much as $2.50 per FTE.
[511] Second, a large portion of copying in the 291 copying events from the Volume Study did not represent compensable copying because we found that the copying was fair dealing.
[332]
[512] Third, as noted by counsel for the Objectors, about 60 per cent of all employees engaged in no copying at all, let alone compensable copying.
[333]
This means that fewer transactions were identified as potentially compensable (the 291) than would have been had all employees been copiers. This is particularly true in relation to the making of paper copies.
[513] The disposition of all 291 copying events is displayed in the Annex, Table 2.
E.
Total amount of royalties generated from the Tariff
[514] As can be seen from Tables 8 and 9 of the Annex, the amounts of royalties that are likely to be generated by the Tariff we are certifying are $14,000 per year for the 2005-2009 period and $60,000 per year for 2010-2014, based on the number of FTEs we have used in Tables 9 and 10. The calculations to arrive at these amounts excluded the governments of Ontario and of the Northwest Territories. The total amounts of royalties generated by the Tariff would have been higher had they been included.
[515] The Board finds that these amounts are fair and equitable given the very low frequency of the making of compensable paper copies among government employees as shown in the evidence of this file.
F.
Other economic questions
i.
A ten per cent discount?
[516] In its letter to the Board relating to administrative provisions, the Consortium proposed that a 10 per cent discount apply to the retroactive portion of the tariff payments.
[334]
It justified this by reference to the K-12 decision, in which a 10 per cent discount was applied.
[517] In that decision, the Board took note of the fact that the amount it was certifying was more than double the amount of royalties payable under the licence the tariff was meant to replace. The decision noted that in the long run, the education system was perfectly capable of dealing with this increase in royalties. But because changing budget allocation takes time, a short term discount of 10 per cent was fair.
[335]
[518] In its reply, Access made several points.
[336]
First, the Objectors are not start-up companies, but rather “well-funded sovereign entities.” Second, the amounts payable under the tariff are unlikely to be noticeable on the balance sheets of the Objectors, unlike in the K-12 case, where many school boards were prohibited from running deficits. Third, the typical risk-sharing rationale does not apply here: there is no evidence of new technologies that the Objectors have adopted, or of the Objectors having taken on substantial risks. Fourth, since most of the Objectors have never paid Access for their copying, a discount would effectively reward their “infringing behaviour.”
[519] Access’ arguments are for the most part irrelevant. The issue is whether the royalties generated by this Tariff are important enough to warrant a discount to account for short time rigidities in budget allocation. The answer is clearly no. The Tariff we certify is orders of magnitude smaller than the royalties paid under the licences of the various provinces to Access, not orders of magnitude larger, as was the case in K-12. Hence, we do not certify a discount.
ii.
Interest on retroactive payments
[520] In its letter to the Board, the Consortium stated that
[i]t is inappropriate to charge more than eight years worth of interest on an amount that is retroactively determined to be owed under a tariff. This imposes a significant burden, and should not be approved for the same reasons that the Consortium submits that there should be a minimum ten per cent discount on the retroactive portion of the tariff. [337]
[521] Access disagreed with the Consortium’s submission.
[338]
First, it noted that the Objectors have effectively had a payment holiday since 2005. Second, it compared the present Tariff with the 2005-2009 K-12 tariff,
[339]
which had multiplying interest factors. Finally, it remarked that the amount owed will be trivial compared to the annual budgets of the Objectors.
[522] In CBC Radio (2011), the Board wrote: “[t]he practice of using interest factors should be generalized.”
[340]
The term “generalized” implies that there need to be special circumstances for interest factors not to be applied. There are no such special circumstances here.
[341]
The general rule set out in CBC Radio applies here. That said, the interest factors have been recalculated to reflect the date of certification of the Tariff. The updated interest factors appear in the “Transitional Provisions” section of the Tariff.
XV.
ADMINISTRATIVE PROVISIONS
[523] The Parties agreed that a single tariff could be certified for the 2005-2014 period and that the wording of the administrative provisions in the 2005 Proposed Tariff should be discarded where it differs from the 2010 Proposed Tariff wording of the administrative provisions.
[342]
[524] That being said, there are some disagreements between the Parties in relation to certain provisions of the 2010 Proposed Tariff. These are addressed below.
A.
Section 2 – Definitions – “Copy” – Printing from an electronic file
[525] One of the methods of copying that was captured in the Volume Study was the printing of a work from an electronic file. However, the wording of the 2010 Proposed Tariff is ambiguous as to whether it permitted such printing. Each of the Parties had a slightly different view of how this activity was captured by the Proposed Tariff.
[526] According to the Consortium, while “it is not readily apparent where the act of ‘printing from an electronic file’ might fit within the definition of “Copy” from the plain wording of the tariff,” it concludes that perhaps it could fit within “a broad and liberal interpretation of the term ‘reprography’” in paragraph 2(a).
[343]
The Government of BC believes that such a copy may fit either into paragraph 2(a) (reprography), or paragraph 2(d) (“reproducing by a machine, device or computer that makes a Digital Copy”).
[344]
[527] In response to the Objectors’ submissions, Access proposed that the definition be clarified that a “Copy” includes a copy made by printing.
[345]
i.
Whether printing from an electronic file is a form of reprography
[528] Usual definitions of the term “reprography,” in the sense of a process,
[346]
do not include printing from an electronic file. For example, the term has been defined as
the science and practice of copying documents by photography, xerography, etc. [347] [emphasis added]
[529] Indeed, the provision itself reinforces this idea: “reprography, which includes facsimile reproduction by photocopying and xerograph.”
[348]
Photocopying and xerograph are both processes which take a physical graphical representation of a work, and then expose it to light, thereby changing the charge on an exposure surface. This is not the case in the situation where a copy is printed from an electronic file.
[530] Access also stated that “‘ Printing from an electronic file’ was not a use covered by the 2005-2009 tariff: that use was first licensed by Access Copyright under the 2010-2014 tariff.”
[349]
[emphasis omitted] This means that if a use were permitted under the 2005 Proposed Tariff, then it cannot be a use that entails the printing from an electronic file.
[531] A comparison of permitted uses in the 2005 Proposed Tariff and the 2010 Proposed Tariff shows that the uses permitted under paragraph (a) are virtually identical. The broadening created in the 2010 Proposed Tariff to include non-visually perceptible reproductions would not affect the inclusion or exclusion of printing from digital copies.
| 2005 Proposed Tariff
|
2010 Proposed Tariff
|
|---|---|
| “Copy” means a visually perceptible reproduction of a Published Work made by any of the following processes:
(a) reproducing by reprographic process, which includes facsimile reproduction by photocopying and xerography;
[…]
|
“Copy” means a reproduction of a Published Work made by any of the following processes:
(a) reprography, which includes facsimile reproduction by photocopying and xerography;
[…]
|
[532] Therefore, in our view, printing from electronic files is not currently captured as “reproduction” in paragraph 2(a) of the 2010 Proposed Tariff.
ii.
Whether printing from an electronic file is “reproducing by a machine that makes a Digital Copy”
[533] Both the Government of BC and Access suggested that printing from an electronic file may be captured in paragraph 2(d), “reproducing by a machine, device or computer that makes a Digital Copy.” Presumably, they mean that this definition can capture any reproduction made by any machine that happens to make a Digital Copy during an act of reproduction process, as neither believes the main result of printing to be a Digital Copy.
[534] “Digital Copy” is defined as “any electronic file of a Published Work.” While Dr. Murphy testified about what copies are made during the process of e-mailing, or the placements of a work on an intranet,
[350]
he did not testify as to whether a printer makes an electronic file of a work when it makes a paper reproduction of a work.
[535] Even if it were the case, such a definition remains opaque. The fact that a printer could make a Digital Copy while printing a paper copy appears to be a circuitous way of conveying the idea that printing from an electronic file may be permitted. Indeed, Dr. Murphy’s remarks suggest that a digital copy of a work is made even during photocopying,
[351]
yet the wording of the 2010 Proposed Tariff is much clearer that such an activity is permitted.
iii.
Conclusion
[536] Since there is ambiguity as to whether the 2010 Proposed Tariff, as worded, sufficiently conveys the idea that printing from an electronic file is permitted, we add another paragraph to section 2, which explicitly states that a “Copy” includes a “reproduction of a published work made by […] (i) printing from an electronic file…”
[537] Furthermore, since Access did not seek in its 2005 Proposed Tariff to include printing from an electronic file, this activity was only included in the 2010 Proposed Tariff. As such, printing is only compensable for the 2010-2014 portion of the Tariff.
[538] Therefore, the provision will read as follows:
“Copy” means a reproduction of a published work made by any of the following processes:
[…]
(i) printing from an electronic file, in the case of reproductions made on or after January 1, 2010
B.
Section 2 – Definitions – “Copy”
[539] Given our conclusions in part IX regarding the compensability of digital copies in this Tariff, we exclude the “digital copy” from the definition of “Copy”, as follows:
“Copy” means a reproduction of a published work made by any of the following processes:
(a) reprography, which includes facsimile reproduction by photocopying and xerography;
[…]
but excludes the digital copy.
C.
Section 2 – Definitions – “FTE”
[540] Section 2 of the 2010 Proposed Tariff states that
“FTE” means a full-time employee or individual working for the Licensee under contract or a part-time employee or individual working for the Licensee under contract whose combined ordinary working hours are counted in proportion to a full-time employee’s ordinary working hours. [352]
[541] The Objectors submitted various arguments as to why it would be inappropriate to include persons such as individuals under contract within the definition of “FTE.” They argue that the governments of the Consortium do not define “FTE” all in the same way. Furthermore, it is inappropriate to include contracts and other such persons as employees.
[353]
Lastly, the definition adopted should be consistent with the definition used for the Volume Study.
[354]
[542] Access, on the other hand, argued that “it would be incongruous if the provincial and territorial governments did not pay for copying performed by its agents, which includes those performing work for the governments under contract.”
[355]
[543] It further argued that
[t]he intent here is to ensure that individuals engaged in copying for government, who act under the control of the government – whether employed or individuals under contract – should be caught by the tariff; to remove the reference would render a (increasing) significant number of persons who copy for provincial or territorial government purpose outside the scope of the tariff. [356]
[544] The report by Circum on the Volume Study indicates that
[t]he population under study is provincial or territorial employees covered by the proposed tariffs. The operational definition of this population differs among jurisdictions; for example, some include health workers while others exclude them. Jurisdictions were responsible for defining the population according to their circumstances and to deliver lists of individuals corresponding to their definition. [357]
[545] This, along with the statement made by the Objectors, convinces us that the various jurisdictions do, in fact, have a different understanding of who is an “employee” of each of the Objectors.
[546] Departing from this understanding of employee in the definition of FTE would risk making the conclusions drawn from the Volume Study less representative. Since the royalty rate for the Tariff is derived from the sample collected in the Volume Study, the Tariff should apply as closely as possible to those persons that belonged to the population under study. Therefore, the definition should only include employees, and the concept of employee should be in accordance with the meaning given to that term in each of the jurisdictions.
[547] For this reason, we exclude from the definition of FTE the notion of persons under contract, and revise the definition as follows:
“employee” means an employee of one of the licensees, as determined in accordance with the applicable enactments, policies, and bookkeeping practices of that licensee.
“FTE” means a full-time employee or a part-time employee whose combined ordinary working hours are counted in proportion to a full-time employee’s ordinary working hours.
[548] A similar structure was used in the K-12 tariff, where the term “full-time equivalent student” was defined as
a full-time student or the equivalent of one student qualifying as a full-time student of an educational institution as determined in accordance with the policies of a ministry related to the funding of educational institutions under its jurisdiction. [358]
[549] To the extent an individual under contract is considered by the government of a particular province or territory to be an employee, they were represented in the Volume Study, and will continue to be captured by the revised definition. If such a person is not considered to be an employee, then this person is not represented in the Volume Study, and is correctly excluded from the definition.
D.
Section 2 – Definitions – “Published Work”
[550] The Government of BC submitted that the definition of “Published Work” should conform to that in the K-12 tariff, which is restricted to works that have been issued to the public in print form.
[359]
The Consortium takes no issue with the definition in the 2010 Proposed Tariff.
[551] Given the manner in which the Volume Study was conducted, namely that all published works were considered, whether issued to the public in print form or otherwise, we see no reason to limit the application of the licence in this manner.
E.
Section 3 – Application
i.
Purposes
[552] The 2010 Proposed Tariff states that the making and distribution of copies may be made “for the non-profit purposes of conducting business within the mandate of the Licensee and for purposes of the delivery of government programs and services.”
[360]
[emphasis added]
[553] Both the Consortium and the Government of BC objected to the inclusion of the “non-profit” limitation.
[361]
[554] We find that the Volume Study did not generally distinguish between copies made for-profit purposes and those made for non-profit purposes. Including such a limitation would make the results from the Volume Study less representative.
[555] Furthermore, the language of section 3 of the 2010 Proposed Tariff suggests that while conducting business within the mandate of a Licensee must be for non-profit purposes, the delivery of programs and services has no such restriction. Where the delivery of programs and services ends and conducting business within one’s mandate begins would be difficult to ascertain.
[556] Therefore, the phrase “non-profit” is removed from the wording of section 3 of the Tariff.
[557] The Consortium also argued that the phrase “delivery of government programs” is insufficient to capture the extent of government activities.
[362]
[558] In our view, the expression “conducting business within the mandate of the Licensee” is very broad, and is sufficient to capture “establishing, delivering and carrying out government programs, services and activities”– as sought by the Consortium. However, in order to avoid any disputes as to the scope of the mandate of any particular licensee, we amend the provision as follows:
Subject to section 4, an FTE shall be permitted to make and distribute copies of published works in the repertoire, for the purpose of conducting the business of the licensee, including for purposes of delivery of government programs and services by means of activities such as, but not limited to, professional, research, archival, communication and administrative activities of the licensee, as follows:
[…]
ii.
Substantial reproductions
[559] The Government of BC argued that sections 3 and 4 of the 2010 Proposed Tariff should be modified to specifically acknowledge that insubstantial reproductions and activities that are exempted from infringement under the Act and are also permitted. It points to previous licences that some governments had with Access that include such a provision to show that such a provision is useful.
[363]
[560] Access responded that acknowledging uses that do not require a licence because they are exempted under the Act would be redundant and unnecessary. In its view, the Tariff should not have such a statement.
[364]
[561] We agree with Access. Unless it is necessary to clarify otherwise ambiguous provisions of the Tariff, it is not necessary to restate copyright law.
iii.
Distribution
[562] Paragraph 3(d) of the 2010 Proposed Tariff provides that a licensee may
(d) subject to paragraph 3(a), distribute:
(i) Copies to FTEs;
(ii) Copies, except for Digital Copies, to persons other than to FTEs; and
(iii) Digital Copies to other Access Copyright licensees that have a licence covering the reproduction and distribution of Digital Copies.
[563] While we do not have evidence as to whether Access has the authority to authorize the distribution of Copies, the Objectors did not oppose the inclusion of this provision. Given that no value is specifically being attributed to this authorization, we include this provision in the Tariff.
[564] However, given our conclusion on digital copies in part IX, above, we remove references to the distribution of Digital Copies in paragraph 3(d), so that it reads as follows:
(d) subject to paragraph 3(a), distribute:
(i) copies to FTEs; and,
(ii) copies to persons other than to FTEs.
F.
Paragraph 4(h) – Moral rights
[565] Paragraph 4(h) of the 2010 Proposed Tariff states that “Copies shall not be made or used in a manner that would infringe the moral rights of any author.”
[566] While the Government of BC submitted that Access cannot enforce the moral rights of authors,
[365]
neither they nor the Consortium explicitly objected to the provision in paragraph 4(h) of the Proposed Tariff. We therefore include this provision in the Tariff.
G.
Digital copies
[567] Paragraph 5(d) of the 2010 Proposed Tariff states that
[w]here the Licensee is no longer covered by a tariff for the making and distribution of Digital Copies, the Licensee shall immediately cease to use Digital Copies of Published Works in the Repertoire, delete from their hard drives, servers or storage area 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.
[568] Given our conclusion on digital copies in part IX, above, we do not include section 5 from the 2010 Proposed Tariff in the Tariff.
H.
Section 5 – Attribution
[569] Section 6 of the 2010 Proposed Tariff states that
The Licensee shall notify all persons entitled to make Copies under this tariff that, where reasonable under the circumstances, Copies made and/or distributed shall include, on at least one page:
(a) a credit to the author (including writer, artist, illustrator and photographer) and to the source; and,
(b) a notice stating, “Copied under licence from Access Copyright. Further reproduction or distribution is prohibited, except as otherwise permitted by law.”
[570] The Consortium argued that “this notice provision cannot be reasonably enforced or monitored across the numerous government offices.”
[366]
[571] The Government of BC submitted that the requirements under section 6 of the 2010 Proposed Tariff “are unreasonable, unenforceable and unduly onerous given the nature of BC’s copying activities,” and that this section should be removed.
[367]
It further argues that “[i]n any case, given that BC has roughly 27,000 employees, a reasonable effort to provide notice is already provided under section 7 below, which should be sufficient.”
[368]
[572] Access argued that this provision is aimed at protecting the moral rights of the authors and is a reasonable condition of a licence. Access noted that this provision is included in the 2010 Ontario Licence.
[369]
[573] We disagree that this provision acts to protect moral rights. The moral rights of authors are protected whether or not this provision is present in the Tariff, as such rights are not diminished by the granting of a licence for making and distributing copies of the author’s work.
[574] However, it appears to us that the Consortium and the Government of BC misunderstand the effect of this requirement. It is not an obligation on each employee to actually provide a credit and notice, but merely an obligation on the licensee to inform their employees that they are to do so when reasonable.
[575] Since we conclude that a failure of an employee to include such a credit or notice does not result in a copy so made to fall outside of the scope of the Tariff, we do not see this requirement as being overly cumbersome. A licensee can comply with this provision by sending a single e-mail to each of its employees.
[576] However, for the same reason that Access objected to restating the law in relation to substantial copying, we view it as unnecessary to include the phrase “Further reproduction or distribution is prohibited, except as otherwise permitted by law” – a statement that does little to inform a potential user of what she/he may or may not do.
[577] The provision is therefore modified to read as follows:
The licensee shall notify all persons under its authority who are entitled to make copies under this Tariff that, where reasonable under the circumstances, copies made and/or distributed shall include, on at least one page:
(a) a credit to the author (including writer, artist, illustrator and photographer) and to the source; and,
(b) a notice stating: “Copied under licence from Access Copyright.”
I.
Section 6 – Notification of the terms and conditions of copying
[578] Section 7 of the 2010 Proposed Tariff states that
Access Copyright shall provide for free, and each Licensee shall affix, within the immediate vicinity of each machine or device used for making Copies in a place and manner that is readily visible to and legible by persons using such machine or device, a notice in the form set out by the Copyright Board.
[579] Access later proposed the following version of this provision:
Access Copyright shall provide for free a notice in the form attached as Schedule “A”. Each Licensee shall make reasonable and good faith efforts to affix the notice within the immediate vicinity of each photocopier used for making Copies in a place and manner that is readily visible to and legible by persons using such photocopier. Each Licensee shall also prominently post a hyperlink to the notice on the home page of each computer network used to make and/or distribute Copies of Published Works in the Repertoire. [370]
[580] The Consortium argued, and we agree, that it is unclear what is meant by “home page of each computer network.” A computer network is simply the connection of two or more computers to share information. Many computer networks do not have anything like a “home page.”
[371]
[581] The Government of BC agreed with the wording in the 2010 Proposed Tariff, with the change of “machine or device” to “photocopier.”
[372]
[582] We note that the phrase “machine or device used for making Copies” can include desktop computers, laptop, mobile phones, cameras, and a myriad of other devices. Such a provision would be very broad and make compliance difficult.
[583] We further note that attempting to craft a notice with a full and accurate description of what may or may not be copied by a user of a photocopy machine would result in a notice that is unworkably long. However, a notice can inform the user of the existence and scope of the Tariff in a reasonable length. We therefore set as the notice to reproduce, with modifications as necessary, those portions of the Tariff that are most applicable to the user of a photocopy machine. We leave it to Access to decide whether they wish to provide such a notice to the licensees or not. The wording of Section 7 of the Tariff is set as follows:
Access Copyright may provide, for free, a notice in the form set out in the Appendix. Each licensee shall make reasonable and good faith efforts to affix the notice within the immediate vicinity of each photocopier used for making copies in such a way that the notice is readily visible to, and legible by, persons using the photocopier.
J.
Section 10 – Surveying
[584] The 2010 Proposed Tariff states that
[w]hen requested by Access Copyright, but not more than once a year, the Licensee shall co-operate with Access Copyright in implementing a method of data collection reasonably required to assist Access Copyright in distributing royalties paid by the Licensee under this tariff. [373]
[585] In its submissions, Access argued that
[s]urveying is not for Access Copyright’s benefit, per se. Rather, surveying is essential so that Access Copyright can distribute the royalties collected under this tariff to its affiliates. This provision is similar to the surveying methodology currently employed under the 2004 Canada Licence. [374]
[586] It also proposed alternative wording, as follows:
As requested by Access Copyright, but not more than once each Year, the Licensee shall cooperate with Access Copyright in implementing a method of data collection that measures a sample of the Licensee’s libraries’ holdings and circulation to assist Access Copyright in distributing the royalties paid by the Licensee under this licence. Access Copyright will work with the Licensee to ensure compliance with legislative and policy requirements concerning privacy, security and non-disclosure. [375]
[587] While the Government of BC’s main concern relates to the cost to the licensees of implementing the surveys,
[376]
the Consortium argued that
it is inappropriate for any data gathered as part of a bibliographic survey to be admissible as evidence in a legal proceeding. Normal evidentiary disclosure requirements would not require the Consortium members to allow a survey in their offices or libraries. [377]
[588] Recognizing the Consortium’s concern regarding disclosure requirements, and recognizing that the effective period of the Tariff will have occurred in the past, we therefore limit the provision as follows:
As requested by Access Copyright, but not more than once, the licensee shall cooperate with Access Copyright in measuring a sample of the licensee’s libraries’ holdings and circulation. Access Copyright will work with the licensee to ensure compliance with legislative and policy requirements concerning privacy, security and non-disclosure. Access Copyright will only collect this information for the purpose of distributing royalties. Any request by Access Copyright must be made no later than one year after the publication of this tariff.
K.
Section 11 – Records and audits
[589] Section 12 of the 2010 Proposed Tariff states that
The Licensee shall keep and preserve, for a period of six (6) years, records from which the royalties due to Access Copyright under this tariff can be readily ascertained.
Access Copyright, or its representative, may audit these records on seven (7) days’ written notice to the Licensee and during normal business hours.
If an audit discloses that royalties due pursuant to section 8 have been understated by more than ten per cent (10%), the Licensee shall pay the reasonable costs of the audit within thirty (30) days of the demand for such payment.
In the event that an audit reveals an overpayment, the Licensee may reduce the amount due on the next royalty payment by the amount of such overpayment.
[590] The Consortium and the Government of BC raised concerns regarding the frequency of such audits, the speed with which they would be able to respond to a request for an audit, as well any obligations they may have regarding the safeguarding of personal and other sensitive information.
[378]
[591] In responses to some of these concerns, Access later proposed the following:
The licensee shall keep and preserve, for a period of six years, records from which the royalties due to Access Copyright under this tariff can be readily ascertained.
No more than once per year, Access Copyright, or its representative, may audit these records on fifteen (15) days’ written notice to the licensee and during normal business hours.
Access Copyright shall, upon receipt, supply a copy of the report of the audit to the licensee which was the subject of the audit.
If an audit discloses that royalties due pursuant to section 7 have been understated by more than ten per cent (10%), the licensee shall pay the reasonable costs of the audit within thirty (30) days of the demand for such payment.
In the event that an audit reveals an overpayment, the licensee may reduce the amount due on the next royalty payment by the amount of such overpayment. [379]
[592] We accept this submitted wording. While the Objectors submitted that it may not be possible to comply with certain laws, regulations or policies as well as this audit provision, they did not specify in which manner such problems would arise. They could have done so, by, for example, stating which law, regulation, or policy would prevent them from meeting this audit provision.
XVI.
TRANSITIONAL PROVISIONS
[593] Given that the Tariff is certified after its effective period, we include transitional provisions. For the reasons set out in previous decisions,
[380]
we also set interest multiplying factors applicable to yearly payments.
Gilles McDougall
Secretary General
ANNEX
Table 1 - DETERMINING THE NUMBER OF POTENTIALLY COMPENSABLE EXPOSURES
|
|
Number of Events
|
|---|---|
| Number of copying events in Volume Study
|
1,466
|
| Of which, copies made for self
|
768
|
| Of which, published documents
|
483
|
| Of which, Access claimed in repertoire
|
311
|
| Of which, Access initially claimed compensability
|
291
|
Table 2 - DISPOSITION OF ALL 291 EVENTS
| Category
|
Events
|
Events Captured
|
Events Remaining
|
|---|---|---|---|
| Initial number of qualifying copying events
|
-
|
291
|
291
|
| Government-owned copyright
|
108, 214, 267, 281
|
4
|
287
|
| Non-affiliated copyright owner
|
2, 4, 13, 15, 41, 47, 50, 51, 53, 79, 97, 98, 99, 105, 111, 120, 124, 127, 132, 134, 135, 144, 145, 148, 156, 202, 207, 215, 231, 233, 235, 241, 243, 247, 259, 260, 262, 277, 288
|
39
|
248
|
| Only digital copying
|
3, 5, 7, 11, 12, 17, 20, 21, 23, 24, 26, 27, 29, 30, 32, 33, 34, 35, 37, 39, 40, 43, 48, 52, 54, 59, 62, 66, 67, 69, 78, 80, 84, 85, 87, 90, 96, 103, 104, 107, 114, 117, 122, 125, 128, 130, 131, 133, 140, 146, 147, 150, 157, 158, 159, 161, 162, 163, 164, 167, 169, 172, 176, 179, 185, 188, 189, 191, 194, 198, 203, 204, 208, 210, 212, 216, 217, 222, 224, 225, 226, 227, 232, 237, 239, 242, 245, 252, 257, 261, 265, 266, 270, 272, 274, 275, 279, 289, 290, 291
|
100
|
148
|
| Outside scope of Tariff
|
77, 119, 149, 180, 234
|
5
|
143
|
| Non-substantial portion copied
|
166, 197, 213, 271, 278
|
5
|
138
|
| ATIP or similar legislation
|
70, 141
|
2
|
136
|
| Compensable (without fair-dealing purpose)
|
6, 14, 18, 44, 49, 55, 56, 60, 83, 89, 94, 118, 129, 153, 154, 155, 206, 219, 230, 236, 246, 254
|
22
|
114
|
| Fair dealing (research and private study category)
|
1, 10, 16, 19, 28, 36, 42, 45, 63, 68, 72, 73, 74, 81, 82, 86, 88, 91, 92, 93, 100, 101, 102, 106, 109, 112, 113, 116, 126, 136, 137, 139, 142, 143, 151, 160, 165, 173, 174, 175, 177, 178, 181, 182, 183, 184, 190, 193, 195, 199, 200, 205, 211, 218, 220, 223, 238, 240, 251, 253, 264, 268, 269, 273, 276, 280, 287
|
67
|
47
|
| Compensable (Objectors accept)
|
58, 65, 95, 258, 284
|
5
|
42
|
| Fair dealing (one-by-one evaluation)
|
8, 22, 25, 38, 57, 61, 64, 71, 76, 115, 121, 123, 152, 168, 170, 171, 186, 192, 196, 201, 209, 221, 228, 248, 249, 250, 255, 256, 282, 283
|
30
|
12
|
| Compensable (one-by-one evaluation)
|
9, 31, 46, 75, 110, 138, 187, 229, 244, 263, 285, 286
|
12
|
0
|
| Total Events
|
|
291
|
|
| Total Compensable Events
|
|
39
|
|
Table 3 - COMPENSABLE EVENTS
| Event #
|
Genre
|
# of pages 2005-2009
|
# of pages 2010-2014
|
Weight
|
Weighted pages 2005-2009
|
Weighted pages 2010-2014
|
|---|---|---|---|---|---|---|
| 60
|
Book
|
20
|
20
|
1,809.48
|
36,189.68
|
36,189.68
|
| 110
|
Book
|
25
|
25
|
2,029.83
|
50,745.84
|
50,745.84
|
| 31
|
Journal
|
0
|
15
|
1,699.81
|
0.00
|
25,497.09
|
| 229
|
Journal
|
0
|
14
|
1,699.81
|
0.00
|
23,797.29
|
| 244
|
Journal
|
0
|
9
|
1,699.81
|
0.00
|
15,298.25
|
| 9
|
Magazine
|
0
|
4
|
1,699.81
|
0.00
|
6,799.22
|
| 58
|
Magazine
|
5
|
5
|
1,574.77
|
7,873.84
|
7,873.84
|
| 95
|
Magazine
|
0
|
9
|
2,029.83
|
0.00
|
18,268.50
|
| 187
|
Magazine
|
1
|
1
|
2,456.13
|
2,456.13
|
2,456.13
|
| 285
|
Magazine
|
0
|
600
|
3,778.68
|
0.00
|
2,267,205.95
|
| 286
|
Magazine
|
5
|
5
|
2,573.56
|
12,867.79
|
12,867.79
|
| 6
|
Newspaper
|
5
|
5
|
1,699.81
|
8,499.03
|
8,499.03
|
| 14
|
Newspaper
|
5
|
5
|
1,699.81
|
8,499.03
|
8,499.03
|
| 18
|
Newspaper
|
5
|
5
|
2,456.13
|
12,280.67
|
12,280.67
|
| 44
|
Newspaper
|
5
|
5
|
1,699.81
|
8,499.03
|
8,499.03
|
| 46
|
Newspaper
|
0
|
1
|
2,456.13
|
0.00
|
2,456.13
|
| 49
|
Newspaper
|
5
|
5
|
1,699.81
|
8,499.03
|
8,499.03
|
| 55
|
Newspaper
|
1
|
1
|
1,809.48
|
1,809.48
|
3,618.97
|
| 56
|
Newspaper
|
0
|
1
|
1,809.48
|
0.00
|
1,809.48
|
| 65
|
Newspaper
|
0
|
2
|
1,699.81
|
0.00
|
3,399.61
|
| 75
|
Newspaper
|
0
|
5
|
1,699.81
|
0.00
|
8,499.03
|
| 83
|
Newspaper
|
1
|
1
|
2,456.13
|
2,456.13
|
2,456.13
|
| 89
|
Newspaper
|
5
|
5
|
1,699.81
|
8,499.03
|
8,499.03
|
| 94
|
Newspaper
|
1
|
1
|
2,456.13
|
2,456.13
|
2,456.13
|
| 118
|
Newspaper
|
1
|
1
|
2,456.13
|
2,456.13
|
2,456.13
|
| 129
|
Newspaper
|
1
|
1
|
1,574.77
|
1,574.77
|
1,574.77
|
| 138
|
Newspaper
|
0
|
1
|
1,809.48
|
0.00
|
1,809.48
|
| 153
|
Newspaper
|
5
|
5
|
1,699.81
|
8,499.03
|
8,499.03
|
| 154
|
Newspaper
|
0
|
1
|
2,573.56
|
0.00
|
2,573.56
|
| 155
|
Newspaper
|
1
|
1
|
1,699.81
|
1,699.81
|
1,699.81
|
| 206
|
Newspaper
|
1
|
2
|
1,809.48
|
1,809.48
|
3,618.97
|
| 219
|
Newspaper
|
0
|
1
|
1,624.61
|
0.00
|
1,624.61
|
| 230
|
Newspaper
|
5
|
5
|
1,699.81
|
8,499.03
|
8,499.03
|
| 236
|
Newspaper
|
5
|
5
|
1,699.81
|
8,499.03
|
8,499.03
|
| 246
|
Newspaper
|
5
|
5
|
1,699.81
|
8,499.03
|
8,499.03
|
| 254
|
Newspaper
|
1
|
1
|
1,809.48
|
1,809.48
|
1,809.48
|
| 258
|
Newspaper
|
4
|
4
|
1,809.48
|
7,237.94
|
7,237.94
|
| 263
|
Newspaper
|
2
|
2
|
1,574.77
|
3,149.54
|
3,149.54
|
| 284
|
Newspaper
|
4
|
4
|
1,809.48
|
7,237.94
|
7,237.94
|
| TOTAL
|
|
|
|
|
232,602.07
|
2,615,259.26
|
Table 4 - VOLUME OF COMPENSABLE EXPOSURES, IN PAGES
| Genres
|
Number of Events
|
Weighted Annualized Pages, 2005-2009
|
Weighted Annualized Pages, 2010-2014
|
|---|---|---|---|
| Book
|
2
|
86,935.52
|
86,935.52
|
| Magazine
|
6
|
23,197.76
|
2,315,471.44
|
| Newspaper
|
28
|
122,468.78
|
148,259.66
|
| Journal
|
3
|
0.00
|
64,592.63
|
| Total Volume of Compensable Exposures
|
39
|
232,602.07
|
2,615,259.26
|
Table 5 - COMPENSABLE EXPOSURES PER FTE, IN PAGES
| Genres
|
Number
|
Annualized Exposures per FTE, 2005-2009
|
Annualized Exposures per FTE, 2010-2014
|
|---|---|---|---|
| Book
|
2
|
0.72
|
0.72
|
| Magazine
|
6
|
0.19
|
19.13
|
| Newspaper
|
28
|
1.01
|
1.23
|
| Journal
|
3
|
0.00
|
0.53
|
| Total
|
39
|
1.92
|
21.61
|
Table 6 - VALUE OF A COMPENSABLE EXPOSURE, IN PENNIES, 2005-2009
|
|
Books
|
Magazines
|
Newspapers
|
Journals
|
|---|---|---|---|---|
| Average Retail Price per Page
|
22.36
|
2.70
|
2.80
|
n/a
|
| Minus portion of costs not corresponding to creative contribution
|
11.54
|
1.97
|
1.83
|
n/a
|
| Value of Creative Contribution
|
10.82
|
0.73
|
0.97
|
n/a
|
| Plus value added through selection of segments of works (30 per cent)
|
3.25
|
0.22
|
0.29
|
n/a
|
| Value of a compensable exposure
|
14.07
|
0.94
|
1.26
|
30.00
|
Table 7 - VALUE OF A COMPENSABLE EXPOSURE, IN PENNIES, 2010-2014
|
|
Books
|
Magazines
|
Newspapers
|
Journals
|
|---|---|---|---|---|
| Average Retail Price per Page
|
24.31
|
2.93
|
3.04
|
n/a
|
| Minus portion of costs not corresponding to creative contribution
|
12.54
|
2.15
|
1.99
|
n/a
|
| Value of Creative Contribution
|
11.77
|
0.79
|
1.05
|
n/a
|
| Plus value added through selection of segments of works (30 per cent)
|
3.53
|
0.24
|
0.32
|
n/a
|
| Value of a compensable exposure
|
15.30
|
1.03
|
1.37
|
32.61
|
Table 8 - FINAL TARIFF RATE CALCULATION, 2005-2009
|
|
Books
|
Magazines
|
Newspapers
|
Journals
|
Total
|
|---|---|---|---|---|---|
| Value of a Compensable Exposure (pennies)
|
14.07
|
0.94
|
1.26
|
30
|
|
| Volume of Compensable Exposures
|
86,935.52
|
23,197.76
|
122,468.78
|
0.00
|
|
| Value of Compensable Exposures (dollars)
|
12,233.24
|
219.03
|
1,542.42
|
0.00
|
13,994.69
|
| Full-Time Equivalent Employees
|
|
|
|
|
121,022
|
| Final Rate (pennies)
|
|
|
|
|
11.56
|
Table 9 - FINAL TARIFF RATE CALCULATION, 2010-2014
|
|
Books
|
Magazines
|
Newspapers
|
Journals
|
Total
|
|---|---|---|---|---|---|
| Value of a Compensable Exposure (pennies)
|
15.30
|
1.03
|
1.37
|
32.61
|
|
| Volume of Compensable Exposures
|
86,935.52
|
2,315,471.44
|
148,259.66
|
64,592.63
|
|
| Value of Compensable Exposures (dollars)
|
13,297.53
|
23,764.48
|
2,029.69
|
21,063.66
|
60,155.36
|
| Full-Time Equivalent Employees
|
|
|
|
|
121,022
|
| Final Rate (pennies)
|
|
|
|
|
49.71
|
[1]
Copyright Act, R.S.C. 1985, c. C-42.
[2]
Statement of Proposed Royalties to Be Collected for the Reproduction of Literary Works in Canada by Provincial and Territorial Governments, for the Years 2005 to 2009, Canada Gazette, April 24, 2004.
[3]
Statement of Proposed Royalties to Be Collected for the Reproduction of Literary Works by Provincial and Territorial Governments (2010-2014), Canada Gazette, May 9, 2009.
[4]
Access Copyright - Provincial and Territorial Governments Tariffs 2005-2014 (5 January 2012) Copyright Board decision.
[5]
Manitoba v. Canadian Copyright Licensing Agency (Access Copyright), 2013 FCA 91.
[6]
See Exhibit AC-2, Witness Statement of Maureen Cavan and Roanie Levy at paras. 65 and following.
[7]
Access was then Cancopy.
[8]
Copibec was then l’Union des écrivaines et écrivains québécois.
[9]
Exhibit AC-2KK.
[10]
Exhibit AC-2LL.
[11]
Exhibit AC-2MM.
[12]
Exhibit AC-2NN.
[13]
Exhibit AC-2L.
[14]
Exhibit AC-2OO.
[15]
CCH Canadian Ltd. v. Law Society of Upper Canada, 2004 SCC 13, [2004] 1 S.C.R. 339. [CCH]
[16]
Exhibit AC-2 at para. 25.
[17]
Exhibit AC-2N.
[18]
This was done by way of an addendum to the 1995 Ontario Licence. See Exhibit AC- 2O. The term of the licence remained the same, namely March 31, 1998.
[19]
Exhibit AC-2Q.
[20]
Exhibit AC-2K.
[21]
Exhibit AC-2T.
[22]
Exhibit AC-2U.
[23]
Exhibit AC-2Z.
[24]
Exhibit AC-2GG.
[25]
Exhibit AC-2-RR.
[26]
Exhibit AC-2-RR.
[27]
Exhibit AC-2-PP.
[28]
Exhibit AC-2-PP.
[29]
Exhibit AC-2QQ.
[30]
Exhibit AC-2M.
[31]
$87,890 for the period of May 3, 2011 to March 31, 2012; $97,800 for the period of April 1, 2012 to March 31, 2013 and $99,755 for the period of April 1, 2013 to March 31, 2014.
[32]
Exhibit AC-1 at paras. 42 and 43.
[33]
Exhibit AC-5 at para. 86.
[34]
Transcripts, Vol. 9, at p. 1861 and following.
[35]
Exhibit AC-5, Table 4.
[36]
Transcripts, Vol. 9, at p. 1863 and following.
[37]
Access Copyright (Educational Institutions) 2005-2009 (June 26, 2009) Copyright Board decision. [K-12]
[38]
Exhibit Consortium-5 at paras. 5 and 6.
[39]
Exhibit AC-4 (revised).
[40]
However, other submissions by Access or its witnesses relied only on the 2003 Quebec Licence. See para. 439 of the decision.
[41]
Exhibit AC-5 at paras. 10 and 11.
[42]
Exhibit Consortium-3 at p. 26.
[43]
Exhibit Consortium-5 at paras. 5 and 6.
[44]
Exhibit AC-4 (revised), Appendix B.
[45]
Exhibit AC-4 (revised) at p. 34.
[46]
Exhibit AC-R-4 (revised) at p. 46.
[47]
Exhibit Consortium-3 at p. 22.
[48]
Ibid.
[49]
Letter of Access to the Board, May 30, 2013, Appendix A.
[50]
E-mail of the Consortium to the Board, May 31, 2013.
[51]
Letter of Access to the Board, August 28, 2014, Appendix A, at p. 2.
[52]
Response of the Consortium to the Board, August 28, 2014, at p. 20.
[53]
Letter of Access to the Board, May 30, 2013; E-mail of the Consortium to the Board, May 31, 2013.
[54]
Consortium, Objectors’ Compensability Reanalysis (E-mail, June 17, 2013); Access, Compensability Reanalysis: Legal Arguments (E-mail, July 22, 2013).
[55]
Transcripts, Vol. 6, at p. 1289.
[56]
Exhibit AC-3 at para. 18.
[57]
Exhibit AC-2 at para. 22 (referencing Access’ online Look Up Tool at: http://www.accesscopyright.ca/look-up-tool/).
[58]
Cinar Corporation v. Robinson, 2013 SCC 73, [2013] 3 S.C.R. 1168 at para. 26. [Robinson]
[59]
CCH, supra note 15 at para. 35.
[60]
Publication in this case not including the making-available of a work on-line.
[61]
Transcripts, Vol. 1, at p. 47; Transcripts, Vol. 2, at p. 318.
[62]
Exhibit AC-2 at para. 21.
[63]
Exhibit AC-3 at para. 10.
[64]
2010 Proposed Tariff, supra note 3, section 2, “Repertoire.”
[65]
Exhibit BC-9 at para. 44; Exhibit Consortium-32 at para. 190.
[66]
Ibid.
[67]
Exhibit AC-23 at para. 102.
[68]
Copyright Act, supra note 1, s. 2, “collective society.”
[69]
Exhibit Consortium-32 at para. 189.
[70]
Exhibit BC-11 at para. 49.
[71]
K-12, supra note 37.
[72]
Ibid. at para. 133.
[73]
Ibid. at para. 133.
[74]
Letter of Access to the Board, April 29, 2013 at p. 6.
[75]
Ibid. at p. 2.
[76]
Ibid. at p. 3.
[77]
Exhibit AC-3 at paras. 25-29.
[78]
Ibid. at paras. 8-9.
[79]
K-12, supra note 37 at para. 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.”) [emphasis in original]
[80]
Transcripts, Vol. 1, at p. 46; Transcripts Vol. 2, at p. 296.
[81]
Transcripts, Vol. 1, at pp. 48-49.
[82]
Access Copyright - Provincial and Territorial Governments Tariff 2005-2014, Order of the Copyright Board (6 May 2014).
[83]
Letter of the Consortium to the Board, June 6, 2014 at pp. 5-6. [Consortium, Re. Digital Copying] Letter of Access to the Board, June 6, 2014 at p. 4. [Access, Re. Digital Copying] Response of the Government of BC, June 6, 2014 at pp 4-5. [BC, Re. Digital Copying]
[84]
Access, Re. Digital Copying supra note 83 at p. 4 (“Where the Licensee ceases to pay the tariff, the Licensee shall immediately cease to make Digital Copies of Published Works in the Repertoire available, and shall make reasonable efforts to delete or take down Digital Copies from any computer or computer network that makes Digital Copies of Published Works in the Repertoire available, to its FTEs on any Internet via Secure Authentication.”)
[85]
Consortium, Re. Digital Copying, supra note 83 at pp. 6-7; BC Re. Digital Copying, supra note 83 at 5-6.
[86]
Access, Re. Digital Copying, supra note 83 at p. 2.
[87]
Ibid. at p. 5.
[88]
Ibid. at p. 4.
[89]
Consortium, Re. Digital Copying, supra note 83 at pp. 6-7 citing Exhibit AC-26, at 6. BC, Re. Digital Copying, supra note 83 at pp. 6-7
[90]
Access, Re. Digital Copying, supra note 83 at p. 5.
[91]
Ibid. at pp. 5-6.
[92]
Exhibit AC-28 at p. 9.
[93]
Exhibit AC-2E, Standard Affiliation Agreement, Appendix A, s. 3A(i)(ii) and s. 3B(i)(ii).
[94]
Transcripts, Vol. 1, at p. 42.
[95]
2010 Proposed Tariff, supra note 3, ss. 3(a), 3(a)(vi).
[96]
Copyright Act, supra note 1, s. 3.
[97]
Exhibit AC-23 at para. 70.
[98]
U & R Tax Services Ltd. v. H & R Block Canada Inc. (1985), 62 C.P.R. (3d) 257; 97 F.T.R. 259 (FCTD). [U & R Tax Services]
[99]
Hager v. ECW Press Ltd., 1998 CanLII 9115 (FC), [1999] 2 FCR 287, (1998) 85 CPR (3d) 289. [Hager]
[100]
Hager, supra note 99 at para. 35; U & R Tax Services, supra note 98 at p. 268.
[101]
Exhibit AC-23 at para. 72.
[102]
Ibid. at para. 70.
[103]
Ibid. at paras. 66-67.
[104]
Exhibit Consortium-35 at para. 36.
[105]
Ibid. at paras. 51-54.
[106]
Ibid. at paras. 47-49.
[107]
Exhibit BC-11 at p. 12.
[108]
Robinson, supra note 58 at para. 26.
[109]
Exhibit AC-23 at para. 82.
[110]
Vaver, David, Copyright Law (Toronto: Irwin Law Inc., 2000), at p. 146.
[111]
Robinson c. Films Cinar inc., 2009 QCCS 3793; France Animation, s.a. c. Robinson, 2011 QCCA 1361; supra note 58.
[112]
Exhibit Consortium-35 at para. 7.
[113]
Ibid. at para. 45.
[114]
Exhibit AC-23 at para. 74.
[115]
Statement of Royalties to Be Collected by SOCAN, NRCC and CSI in Respect of Multi-Channel Subscription Satellite Radio Services, (6 May 2009, Corrected Version) Copyright Board decision at para. 98. In the decision, the Board held that the reproduction of 4 to 6 seconds of a musical work, of approximately 4 minutes in length on average, did not constitute a substantial reproduction of that work.
[116]
Access submissions, July 22, 2013 at para. 49, events 70 and 141.
[117]
Exhibit AC-23 at para. 5.
[118]
Ibid. at para. 11.
[119]
Exhibit Consortium-35 at para. 4.
[120]
Alberta (Education) v. Canadian Copyright Licensing Agency (Access Copyright), 2012 SCC 37, [2012] 2 S.C.R. 345 at para. 12. [Alberta]
[121]
Exhibit Consortium-35 at para. 11.
[122]
Exhibit AC-23 at para. 11.
[123]
CCH, supra note 15 at para. 53.
[124]
Alberta, supra note 120 at para. 35.
[125]
Alberta, supra note 120 at para. 37; CCH, supra note 15 at para. 52.
[126]
Exhibit AC-23 at para. 16.
[127]
Exhibit Consortium-35 at para. 7.
[128]
CCH, supra note 15 at para. 63.
[129]
Exhibit AC-R-2.
[130]
Exhibit BC-2.
[131]
CCH, supra note 15.
[132]
Copyright Act, supra note 1, ss. 29.1, 29.2.
[133]
Alberta, supra note 120 at para. 22.
[134]
Exhibit AC-23 at para. 22.
[135]
Society of Composers, Authors, and Music Publishers of Canada v. Bell Canada, 2012 SCC 36, [2012] 2 S.C.R. 326. [Bell]
[136]
Exhibit AC-25 at para. 24.
[137]
Exhibit AC-23 at para. 22.
[138]
CCH, supra note 15 at para. 48.
[139]
Ibid. at para. 51.
[140]
Exhibit AC-25 at paras. 8 to 10.
[141]
Interpretation Act, R.S.C. 1985, c. I-21, s. 12.
[142]
Warman v. Fournier, 2012 FC 803, http://www.canlii.org/en/ca/fct/doc/2012/2012fc803/2012fc803.html. [Warman]
[143]
Ibid. at para. 31.
[144]
Alberta (Education) v. Access Copyright, 2010 FCA 198.
[145]
Alberta, supra note 120 at para. 19.
[146]
Exhibit AC-23 at paras. 53-54.
[147]
Bell, supra note 135 at para. 27.
[148]
K-12, supra note 37 at para. 88.
[149]
Ibid.
[150]
Exhibit AC-23 at para. 21.
[151]
Order Fixing Various Dates as the Dates on which Certain Provisions of the Act Come into Force, SI/2012-85 (7 November 2012), http://canadagazette.gc.ca/rp-pr/p2/2012/2012-11-07/html/si-tr85-eng.html.
[152]
Copyright Act, supra note 1, s. 29.
[153]
Bell, supra note 135 at para. 27.
[154]
Exhibit AC-4 (revised) at 153-154.
[155]
Bell, supra note 135 at para. 33.
[156]
CCH, supra note 15 at para. 54.
[157]
Exhibit AC-23 at para. 28.
[158]
Exhibit BC-11 at para. 14.
[159]
Ibid. at para. 15.
[160]
Exhibit AC-23 at para. 29.
[161]
CCH, supra note 15 at para. 51.
[162]
Ibid. at para. 54.
[163]
Exhibit Consortium-32 at paras. 69-77.
[164]
Exhibit AC-35 at para. 25.
[165]
CCH, supra note 15 at para. 71.
[166]
Ibid. at para. 54.
[167]
Giuseppina D’Agostino, “The Arithmetic of Fair Dealing at the Supreme Court of Canada,” in Michael Geist, ed., The Copyright Pentalogy: How the Supreme Court of Canada Shook the Foundations of Canadian Copyright Law (University of Ottawa Press, 2013) at p. 197.
[168]
Alberta, supra note 120 at para. 30.
[169]
Exhibit AC-23 at para. 28.
[170]
Ibid.
[171]
Alberta, supra note 120 at paras. 21-22.
[172]
Exhibit Consortium-32 at para. 21.
[173]
Alberta, supra note 120 at para. 22.
[174]
Exhibit AC-13; Exhibit AC-14; Exhibit AC-15.
[175]
CCH, supra note 15 at para. 55.
[176]
Alberta, supra note 120 at para. 3.
[177]
Exhibit Consortium-32 at paras. 28-29.
[178]
Alberta, supra note 120 at paras. 29-30.
[179]
Exhibit AC-25 at para. 27.
[180]
Exhibit Consortium-32 at para. 36 citing CCH, supra note 15 at para. 63.
[181]
Ibid. at para. 36.
[182]
Exhibit AC-23 at para. 33 (footnotes omitted).
[183]
Bell, supra note 135 at para. 43.
[184]
Exhibit AC-23 at para. 61.
[185]
CCH, supra note 15 at para. 66.
[186]
Bell, supra note 135 at para. 38.
[187]
Exhibit BC-11 at para. 17.
[188]
CCH, supra note 15 at para. 56.
[189]
Alberta, supra note 120 at para. 29.
[190]
Exhibit AC-25 at para. 33, citing Exhibit Consortium-32 at para. 42.
[191]
Ibid. at para. 33.
[192]
Ibid. at para. 34.
[193]
Ibid. at para. 35.
[194]
Bell, supra note 135 at para. 39, citing the decision of the Board SOCAN Statement of Royalties, Internet - Online Music Services, 1996-2006 (Tariff 22.A) (Re), (2007) 61 CPR (4th ed.) 353 at para. 113 (brackets in original).
[195]
Exhibit AC-25 at para. 36.
[196]
Alberta, supra note 120 at para. 30.
[197]
CCH, supra note 15 at para. 56.
[198]
Bell, supra note 135 at para. 39.
[199]
Exhibit AC-23 at para. 31.
[200]
Ibid.
[201]
Exhibit AC-25 at para. 34, footnote 21.
[202]
Exhibit AC-23 at para. 31.
[203]
Warman, supra note 142.
[204]
Ibid. at para. 33.
[205]
CCH, supra note 15 at para. 68.
[206]
Ibid.
[207]
Bell, supra note 135.
[208]
Ibid. at para. 4.
[209]
Ibid. at para. 49.
[210]
K-12, supra note 37 at para. 103.
[211]
Warman, supra note 142 at para. 33.
[212]
Ibid. at para. 34.
[213]
Century 21 Canada Limited Partnership v. Rogers Communications Inc., 2011 BCSC 1196. [Century 21]
[214]
Ibid. at para. 268 (The mention of “repeated actions” in considering the amount of the dealing factor appears to have been influenced by the Supreme Court’s statement in CCH, supra note 15 at para. 68 that “dealings might not be fair if a specific patron of the Great Library submitted numerous requests for multiple reported judicial decisions from the same reported series over a short period of time.”)
[215]
CCH, supra note 15 at para. 53.
[216]
Ibid. at para. 56. (“For example, for the purpose of research or private study, it may be essential to copy an entire academic article or an entire judicial decision. However, if a work of literature is copied for the purpose of criticism, it will not likely be fair to include a full copy of the work in the critique.”)
[217]
Ibid. at para. 57.
[218]
Alberta, supra note 120 at paras. 31-32.
[219]
CCH, supra note 15 at para. 70.
[220]
Ibid. at para. 57.
[221]
Exhibit Consortium-32 at para. 46.
[222]
Exhibit AC-23 at para. 35.
[223]
Exhibit AC-4 (revised) at p. 157.
[224]
Ibid.
[225]
CCH, supra note 15 at para. 57.
[226]
Exhibit AC-23, footnote 43.
[227]
Alberta, supra note 120 at para. 32.
[228]
CCH, supra note 15 at para. 70.
[229]
In short, a loss of economic efficiency resulting from situations such as artificial scarcity.
[230]
CCH, supra note 15 at para. 58.
[231]
Exhibit AC-23 at para. 34.
[232]
Exhibit Consortium-32 at para. 63.
[233]
CCH, supra note 15 at para. 71.
[234]
Ibid., citing Linden, J.A. from the decision below CCH Canadian Ltd. v. Law Society of Upper Canada, 2002 FCA 187, [2002] 4 FCR 213 at para. 159.
[235]
Exhibit Consortium-32 at para. 63.
[236]
Alberta, supra note 120 at para. 30.
[237]
CCH, supra note 15 at para. 59.
[238]
Exhibit AC-23 at para. 36.
[239]
Exhibit AC-25 at para. 48.
[240]
CCH, supra note 15 at para. 70.
[241]
Alberta, supra note 120 at paras. 29-30, Abella J. and at para. 50, Rothstein J., dissenting.
[242]
Exhibit Consortium-32 at para. 67.
[243]
The order in which we categorize the events matters. For example, some of the 67 events we found to be fair dealing are events which the Objectors submitted were compensable. However, had we accepted the Objectors’ submissions on those events, we would have treated some of those fair- dealing events inconsistently. We therefore did not automatically accept the Objectors’ submissions on those events.
[244]
Exhibit AC-23 at para. 2.
[245]
See e.g., Reference re Broadcasting Regulatory Policy CRTC 2010-167 and Broadcasting Order CRTC 2010-168, 2012 SCC 68, [2012] 3 SCR 489 at para. 80.
[246]
Libel and Slander Act, RSO 1990, c L.12, ss. 3-4.
[247]
Criminal Code, R.S.C., 1985, c. C-46.
[248]
Beloff v. Pressdram Ltd., [1973] All E.R. 241 (Ch.D.).
[249]
Hubbard v. Vosper, [1972] 2 Q.B. 84.
[250]
Exhibit BC-9 at para. 42.
[251]
Ibid.
[252]
CCH, supra note 15 at para. 71.
[253]
Exhibit AC-5 (revised) at para. 8.
[254]
Exhibit AC-4 at p. 50.
[255]
Transcripts, Vol. 1, at p. 10.
[256]
Exhibit Consortium-4 at para. 55.
[257]
The reservation price is the point beyond which a negotiator is ready to walk away from a negotiated agreement.
[258]
Market clearing is the process by which supply equals demand at the market price.
[259]
Under the assumption of Nash bargaining, (a type of bargaining between two equally informed, equally matched parties) the transactional price is the midpoint of the price interval.
[260]
This is not unusual. Dr. Dujsic suggested that most analyses of market comparables end in rejecting the market comparable approach. (See Transcripts, Vol. 6, at p. 1384). This is because there are usually too many differences between the “comparable” market and the target market.
[261]
For parsimony, we present these prices on the assumption that a single price will prevail, as opposed to having a starting price for 2005 and adjustments for inflation. Access used both assumptions. We also present these prices on the assumption that the value of the indemnity clause relates to the percentage of repertoire covered by cheques to non-affiliates. Once again, Access used both assumptions.
[262]
While the 2004 Canada Licence provided for a lump-sum payment, the implied per- FTE rate is equal to this lump sum, divided by the number of FTEs.
[263]
In particular, the limited information situation arose because there had not been a copying survey in those provinces.
[264]
Although Access claimed that the Canada licence was subject to limited information, there had been a study, namely the Goss Gilroy Study mentioned above.
[265]
Transcripts, Vol. 4, at p. 698.
[266]
Transcripts, Vol. 4, at p. 756.
[267]
Exhibit AC-5 (revised) at para. 88.
[268]
Transcripts, Vol. 3, at p. 647.
[269]
Transcripts, Vol. 4, at p. 756.
[270]
Exhibit AC-2RR.
[271]
Ibid. at clause 10.2.
[272]
Exhibit AC-5 at para. 83.
[273]
A contingent liability is a potential obligation that may be incurred depending on the outcome of a future event.
[274]
This is similar to arguments made by Messrs. Dujsic and Smith, see Transcripts, Vol. 6, at p. 1396.
[275]
Exhibit AC-2RR at clause 8.4.
[276]
Transcripts, Vol. 4, at p. 748.
[277]
Transcripts, Vol. 4, at p. 863.
[278]
Transcripts, Vol. 4, at p. 871.
[279]
Letter of Access to the Board, August 9, 2010 at pp. 4-5.
[280]
Letter of Access to the Board, November 1, 2010 at p. 8.
[281]
Exhibit AC-R-4 at section 2.12.
[282]
Notice of the Board, March 14, 2013.
[283]
Letter of the Consortium to the Board, April 12, 2013.
[284]
A t-test is a statistical test of whether some calculated statistic is equal to zero. In this case, the t-test was set up to test whether the participation bias was equal to zero and whether the completion bias was equal to zero.
[285]
Letter of Access to the Board, April 12, 2013.
[286]
Truncation is a method of cleaning the data that eliminates values above or below a given threshold.
[287]
Transcripts, Vol. 3, at p. 581.
[288]
Transcripts, Vol. 5, at p. 1074.
[289]
That is, a stratification as if it had been part of the sample design.
[290]
Letter of British Columbia to the Board, August 18, 2010, at p. 2.
[291]
Ibid.
[292]
Letter of the Consortium to the Board, November 1st, 2010, at p. 6.
[293]
Ibid. at p. 5.
[294]
Letter of Access to the Board, November 1st, 2010, at p. 2.
[295]
Ibid. at Appendix A.
[296]
Letter of the Consortium to the Board, November 15, 2010 at pp. 5-7.
[297]
Ibid. at pp. 2-4.
[298]
Social desirability bias is the systematic underreporting of undesirable attitudes or behaviour and the systematic over reporting of desirable ones.
[299]
Exhibit AC-4 (revised) at p. 28.
[300]
These values are considered “missing” because the respondent did not enter the values when completing the Study.
[301]
Exhibit Consortium-18.
[302]
Transcripts, Vol. 5, at p. 1045.
[303]
Exhibit Consortium-3 at p. 72.
[304]
Exhibit AC-R-4 (revised) at p. 21.
[305]
Transcripts, Vol. 5, at p. 1073.
[306]
This fact was conceded by Mr. Gauthier (see Transcripts, Vol. 3, at p. 418).
[307]
Order of the Board, July 21, 2014.
[308]
Upon further review of the Parties’ submissions, the Board found 39 compensable transactions.
[309]
Exhibit AC-27A at p. 16.
[310]
Exhibit Consortium-36 at p. 17.
[311]
Ibid. at p. 12.
[312]
Letter of Access to the Board, April 12, 2013, at p. 3.
[313]
Transcripts, Vol. 4, at p. 719.
[314]
Exhibit Consortium-5, at paras. 26-27.
[315]
Creative contribution was defined obliquely in the K-12 decision as follows: “We find that only elements related to the making of a protected work and its “master” must be considered, and all other elements related to subsequent stages in the making should be excluded, such as printing, distribution, marketing and administration costs.” K-12, supra note 37 at para. 162.
[316]
Exhibit AC-R-5 at para. 54.
[317]
Transcripts, Vol. 7, at p. 1579.
[318]
The amount of the shipping fee is not always obvious, in part because the right to receive free shipping for items delivered within a certain number of days depends on several factors, including the value of the total order, the relationship between the seller and the buyer, and various promotions.
[319]
Exhibit Consortium-5 at paras. 38-39.
[320]
Exhibit AC-R-4 at p. 47.
[321]
K-12, supra note 37, Table 4.
[322]
To be sure, if both proportions are measured with a margin of error, there is a positive probability that the two frequencies of magazine copying are equal.
[323]
Exhibit AC-8D at p. 7 (of K-12, supra note 7).
[324]
SOCAN Multiple Tariffs (2004) at p. 18.
[325]
SOCAN-Re:Sound CBC Radio Tariff, 2006-2011, Copyright Board decision, 8 July 2011. [CBC Radio]
[326]
Exhibit Consortium-5 at paras. 40-43.
[327]
As noted by Mr. Chodorowicz, the marginal cost of copying one page, $0.10, was the relevant cost from an economic perspective. (See Transcripts, Vol. 7, at p. 1579). The fixed cost of $3.38 per FTE does not affect copying behaviour.
[328]
Exhibit Consortium-5 at para. 47.
[329]
Ibid. at paras. 44-48.
[330]
Ibid. at para. 51.
[331]
Exhibit AC-R-5 at paras. 55-56.
[332]
In fact, as mentioned by Mr. McGrath, one of the reasons for the governments to do the Study was to show that they engaged in far less compensable copying than believed by Access. (See Transcripts, Vol. 4, at p. 894).
[333]
Transcripts, Vol. 4, at p. 872.
[334]
Letter of the Consortium to the Board, March 28, 2013, at p. 7.
[335]
K-12, supra note 37 at para. 188.
[336]
Letter of Access to the Board, April 26, 2013, at pp. 13-14.
[337]
Letter of the Consortium to the Board, March 28, 2013, at p. 11.
[338]
Letter of Access to the Board, April 26, 2013, at p. 20.
[339]
K-12, supra note 37.
[340]
CBC Radio, supra note 325 at para. 131.
[341]
We do not view the fact that the 2005 Proposed Tariff was not considered for four years after its filing to be a special circumstance.
[342]
Exhibit AC-26 at p. 1; Exhibit Consortium- 37 at p. 1; Exhibit BC-1; Exhibit BC-2.
[343]
Consortium, Re. Digital Copying, supra note 83 at pp. 10-11.
[344]
BC, Re. Digital Copying, supra note 83 at para. 17.
[345]
Letter of Access to the Board, June 13, 2014, at p. 5.
[346]
2010 Proposed Tariff, supra note 3, s. 2, “Copy” (“means a reproduction of a Published Work made by any of the following processes”).
[347]
The Canadian Oxford Dictionary, s.v. “reprography.”
[348]
2010 Proposed Tariff, supra note 3, s. 2, “Copy.”
[349]
Access, Re. Digital Copying supra note 83 at p. 5.
[350]
Exhibit AC-19; Exhibit AC-R-6.
[351]
Transcripts, Vol. 8, at pp. 1727-1728.
[352]
2010 Proposed Tariff, supra note 3, s. 2, “FTE.”
[353]
Exhibit Consortium-37 at p. 2; Exhibit BC- 12 at pp. 1-2.
[354]
Exhibit BC-12 at pp. 1-2.
[355]
Exhibit AC-26 at p. 3.
[356]
Exhibit AC-28 at p. 4.
[357]
Exhibit AC-4 (revised) at p. 6.
[358]
Statement of Royalties to Be Collected by Access Copyright for the Reprographic Reproduction, in Canada, of Works in its Repertoire - Educational Institutions (2005- 2009), s. 2, “full-time equivalent student”.
[359]
Exhibit BC-12 at para. 5.
[360]
2010 Proposed Tariff, supra note 3, s. 3.
[361]
Exhibit Consortium-37 at pp. 3-4; Exhibit BC-12 at para. 11.
[362]
Exhibit Consortium-37 at p. 3.
[363]
Exhibit BC-12 at paras. 12-14.
[364]
Exhibit AC-26 at p. 5.
[365]
Exhibit BC-12 at para. 17.
[366]
Exhibit Consortium-37 at p. 5.
[367]
Exhibit BC-1 at para. 28.
[368]
Exhibit BC-12 at para. 18.
[369]
Exhibit AC-26 at p. 7.
[370]
Ibid. at p. 8.
[371]
Exhibit Consortium-37 at p. 6.
[372]
Exhibit BC-12 at para. 19.
[373]
2010 Proposed Tariff, supra note 3, s. 11.
[374]
Exhibit AC-26 at p. 11.
[375]
Ibid.
[376]
Exhibit BC-12 at paras. 26-27.
[377]
Exhibit Consortium-37 at p. 8.
[378]
Ibid. at pp. 9-10.
[379]
Exhibit AC-28 at p. 18.
[380]
See for instance CBC Radio, supra note 325 at paras. 131-133.