Copyright Board Canada

Commission du droit d’auteur Canada

 

Date	2021-05-28
Citation	Pay Audio Tariff (2007–2016), 2021 CB 5
Members	The Honorable Robert A. Blair Mr. Claude Majeau Mr. J. Nelson Landry
Proposed Tariffs Considered	SOCAN Pay Audio Services Tariff, 2010, 2011, 2012, 2013, 2014, 2015 SOCAN Tariff 26 – Pay Audio Services Tariff, 2016 SOCAN Tariff 22.2 – Audio Webcasts, 2007, 2008 SOCAN Tariff 22.B – Audio Webcasts, 2009 SOCAN Tariff 22.B – Audio Webcasts, 2010, 2011, 2012, 2013 SOCAN Tariff 22.B – Commercial Radio, Satellite Radio and Pay Audio, 2014, 2015, 2016 NRCC Pay Audio Services Tariff, 2007-2011 Re:Sound Pay Audio Services Tariff, 2012-2013, 2014-2016 Re:Sound Tariff 8.A – Simulcast and Webcasting, 2009-2012 Re:Sound Tariff 8 – Simulcasting, Non-Interactive Webcasting and Semi-Interactive Webcasting, 2013-2015 Re:Sound Tariff 2.B – Pay Audio Services Simulcasts, 2016 Re:Sound Tariff 8 – Non-Interactive and Semi-Interactive Webcasting, 2016

Approval of Proposed Tariffs

As

Re:Sound and SOCAN – Stingray Pay Audio and Ancillary Services Tariff (2007 –2016)

Reasons for decision

I. SUMMARY

[1] These are the reasons for our approval of various proposed tariffs filed by the Society of Composers, Authors and Musical Publishers of Canada (SOCAN) and the Re:Sound Music Licensing Company (Re:Sound). We approve these tariffs as a single tariff entitled Re:Sound and SOCAN – Stingray Pay Audio and Ancillary Services Tariff (2007–2016) (the “Tariff”).

[2] The Tariff covers the following activities: the performance to the public of musical works and sound recordings done for the purpose of broadcasting pay audio programming, the simulcasting of that programming on the Internet (“simulcasts”), and the provision of a semi-interactive service with content similar to that of pay audio programming to customers of broadcast distribution undertakings (“near-simulcasts”). The Tariff is not intended to cover all webcasting services provided by, or through, Stingray. Activities not covered by this Tariff may be covered by other tariffs, including SOCAN – Tariff 22.A (Internet – Online Music Services), 2007-2010; [1] CSI, SOCAN, SODRAC – Tariff for Online Music Services, 2010-2013; [2] and Re:Sound – Tariff 8 (Non-Interactive and Semi-Interactive Webcasts), 2009-2012. [3]

[3] Pay audio programming is a set of music channels distributed by cable or satellite broadcast distribution undertakings (BDUs), provided to them as a service by a third party. [4] There is currently only one provider of a pay audio service in Canada: the Stingray Digital Group Inc. (Stingray). [5]

[4] Stingray authorizes various BDUs to distribute its pay audio programming to their subscribers. The last approved tariff covering this activity was the 2007–2009 Pay Audio Tariff. [6]

[5] Stingray also provides certain ancillary services related to pay audio to BDUs. In this proceeding, we are concerned with simulcasts and near-simulcasts based on pay audio programming. Stingray provides other, non–pay-audio related services, to BDUs; these are not under consideration in this proceeding.

[6] All these services—those related to pay audio as well as those that are not—are sold together by Stingray as a bundle to BDUs and provided by BDUs to their subscribers as bundled services. Each BDU pays Stingray a monthly fee, computed as a negotiated rate per subscriber. This monthly fee constitutes Stingray’s revenues from the bundle of services. The Parties in this proceeding referred to this monthly fee as the “affiliation payment” ; we do so in these reasons as well.

A. The Issues

[7] The Board has used the following formula for determining royalties in many previous tariffs:

Rate Base ($) x Gross Royalty Rate (%) x Repertoire-Use Adjustment (%) = Royalties ($)

[8] The proposed tariffs for pay audio are also based on this formula, albeit with the product of the Gross Royalty Rate and the Repertoire-Use Adjustment expressed as a single Royalty Rate. The Parties’ submissions all implicitly or explicitly adopt this formula for pay audio as well as for simulcast. [7] We use this formula as the framework for determining all the royalty rates under consideration.

[9] Each element of this formula was in dispute in this proceeding. After providing an overview of the Parties in this proceeding, we deal with the elements of this formula in the following manner:

• The Rate Base (the revenue to which the royalty rates apply) pertaining to the proposed tariffs for pay audio;
• The Gross Royalty Rate applicable to the proposed tariffs for pay audio;
• The Gross Royalty Rate and Rate Base applicable to the Ancillary Proposed Tariffs; and
• The Repertoire-Use Adjustment applicable to all proposed tariffs under consideration.

B. Summary of Decision

[10] Based on the reasons below, we approve the royalty rates in Table 1.

Table 1: Summary of Royalty Rates

Table 1: Summary of Royalty Rates

Activity/Year	2007	2008	2009	2010	2011	2012	2013	2014	2015	2016
SOCAN
Re:Sound
Pay Audio	N/A			12.45%	12.04%	11.63%	11.21%	10.79%	10.38%	9.96%
Simulcasts	nil			12.45%	12.04%	11.63%	11.21%	10.79%	10.38%	9.96%
Near-Simulcasts	5.3%	5.3%	5.3%	5.3%	5.3%	5.3%	5.3%	5.3%	5.3%	5.3%
Pay Audio	N/A			5.85%	5.66%	5.46%	5.27%	5.85%	6.88%	6.60%
Simulcasts	N/A		nil	5.85%	5.66%	5.46%	5.27%	5.85%	6.88%	6.60%
Near-Simulcasts	N/A		N/A				2.49%	2.87%	3.51%	3.51%

[11] In respect of simulcasts, we conclude that royalties payable for pay audio, whether in the Tariff we approve today, or under the 2007–2009 Pay Audio Tariff, already include royalties for simulcasts. We therefore set no additional simulcast royalties here.

[12] We also establish the portion of affiliation payments that should be used in calculating the royalties for pay audio under the Tariff. The exact percentage will vary among BDUs and from year to year, depending on the exact composition of services they receive from Stingray. However, in respect of near-simulcasts, we do not fix such a proportion in the Tariff.

C. Estimated Total Royalties

[13] We have attempted to estimate the total royalties that will be payable pursuant to the Tariff. However, this estimate comes with many limitations. For example:

• We do not know which BDUs offered which services in which years, other than the major BDUs and those that were members of the Canadian Cable Systems Alliance;
• We do not know which BDUs are entitled to the 50% discount f or being a small cable transmission system; and
• We do not know the amount of revenues properly attributable to near-simulcasts, especially as distinguished from other semi-interactive webcast services (e.g., semi-interactive webcasts accessed through direct subscriptions to Stingray Mobile).

[14] With these caveats, we nevertheless find it useful to present our estimates; they are in Table 2.

Table 2: Estimated Total Royalties for 2010 to 2016 ($ millions)

Table 2: Estimated Total Royalties for 2010 to 2016 ($ millions)

Year	Pay Audio and Simulcast Royalties	Near-Simulcast Royalties	Total Royalties
2010	3.3	0.0	3.3
2011	3.0	0.0	3.0
2012	3.2	0.0	3.2
2013	3.4	0.0	3.4
2014	2.5	0.2	2.7
2015	2.6	0.2	2.8
2016	2.5	0.3	2.7
Total	20.5	0.7	21.2

II. OVERVIEW

A. THE PARTIES

[15] Two collective societies, SOCAN and Re:Sound, jointly the “ Collectives,” each proposed tariffs related to the pay audio service and certain related services.

[16] SOCAN is a collective society that administers copyrights in musical works. Re:Sound is a collective society that administers the remuneration right in sound recordings.

[17] While there were many objections filed in response to these proposed tariffs (see Appendix for a list of these objectors), only Stingray and the following group of BDUs participated in the oral hearings: Bell Canada, the Canadian Cable Systems Alliance, Cogeco Cable, Quebecor Media Inc., Rogers Communications Partnership, Shaw Communications Inc., Telus Communications Co., and Videotron G.P. We refer to them as the “Objectors.”

[18] Headquartered in Montreal, Stingray is a multinational corporation that operates in 111 countries. [8] It broadcasts music and video content on a number of platforms including digital TV, satellite TV, Internet TV, Internet radio, mobile devices and game consoles.

[19] We refer to the two Collectives, and the Objectors, jointly as the “Parties.”

B. THE PROPOSED TARIFFS UNDER CONSIDERATION

[20] As per Notices CB-CDA 2016-002 and CB-CDA 2018-205, we are considering the following proposed tariffs, or portions thereof (jointly, the “Proposed Tariffs”) in this proceeding.

For SOCAN:

• Pay Audio Services Tariff, 2010, 2011, 2012, 2013, 2014, 2015;
• SOCAN Tariff 26 – Pay Audio Services Tariff, 2016;
• SOCAN Tariff 22.2 – Audio Webcasts, 2007, 2008 [simulcast of pay audio and near-simulcast based on pay audio programming only];
• SOCAN Tariff 22.B – Audio Webcasts, 2009 [simulcast of pay audio and near-simulcast based on pay audio programming only];
• SOCAN Tariff 22.B – Audio Webcasts, 2010, 2011, 2012, 2013 [simulcast of pay audio and near-simulcast based on pay audio programming only]; and
• SOCAN Tariff 22.B – Commercial Radio, Satellite Radio and Pay Audio, 2014, 2015, 2016 [simulcast of pay audio and near-simulcast based on pay audio programming only].

For Re:Sound:

• NRCC Pay Audio Services Tariff, 2007-2011 [2010-2011 only];
• Re:Sound Pay Audio Services Tariff, 2012-2013 and 2014-2016;
• Re:Sound Tariff 8.A – Simulcast and Webcasting, 2009-2012 [simulcast of pay audio only];
• Re:Sound Tariff 8 – Simulcasting, Non-Interactive Webcasting and Semi-Interactive Webcasting, 2013-2015 [simulcast of pay audio and near-simulcast based on pay audio programming only];
• Re:Sound Tariff 2.B – Pay Audio Services Simulcasts, 2016; and
• Re:Sound Tariff 8 – Non-Interactive and Semi-Interactive Webcasting, 2016 [only near-simulcast based on pay audio programming].

[21] Put succinctly, we are considering proposed tariffs for SOCAN and Re:Sound, from 2010 to 2016 relating to the pay audio service (the “Proposed Pay Audio Tariffs”), and any unapproved tariffs, or portions of tariffs, from 2007 to 2016, for the simulcast or near-simulcast services (the “Proposed Ancillary Tariffs”).

C. PROCEDURAL HISTORY

[22] This proceeding had two phases. The first phase dealt primarily with the issues of the Gross Royalty Rates and the Rate Base for pay audio and simulcasts. An oral hearing on these issues took place in May 2017, followed by further questions from the Board to the Parties. In October 2018, near-simulcasts were formally moved into this proceeding.

[23] The second phase dealt solely with the issue of Repertoire-Use Adjustment, which accounts for the proportion of music used that was in the Collectives’ repertoire and eligible for royalties. After the completion of a study on this issue, the Parties filed settlements regarding the Repertoire-Use Adjustment by October 2019.

D. POSITION OF THE PARTIES

i. Objectors’ Position

[24] The Objectors state that since 2010, Stingray has been selling a Suite of Services to BDUs, which included pay audio and other channels. Stingray has not been disaggregating its revenues into those associated with pay audio and those associated with the other services (although it does measure costs on a disaggregated basis). Rather it, and the BDUs, have been paying the royalties set in the 2007–2009 Pay Audio Tariff on the entire amount of this revenue—effectively an overpayment. [9] The Board should clarify that the rate base does not capture revenues from activities not covered under the proposed tariffs. [10]

[25] The Objectors argue that the Board should not approve the same royalty rate for pay audio as in the 2007–2009 Pay Audio Tariff. When the Board first approved a tariff for pay audio in 2002, there was no other competing service that supplied commercial-free music channels, each devoted to a specific genre of music. Now, there are hundreds of streaming services online, in addition to satellite radio, which streams into people’s homes and car radios. It is inconsistent that these streaming services be subject to a significantly lower rate than pay audio, and that principles of equity and of technological neutrality support setting a similar rate for pay audio as for similar services. [11]

ii. Collectives’ Position

[26] The Collectives argue that the royalty rate in the Proposed Pay Audio Tariffs should apply to revenue from the entire Suite of Services. The revenue allocation from Stingray is arbitrary and not used by Stingray in its internal accounting systems. The allocation is a transparent way to avoid paying royalties and to penalize the Collectives for Stingray’s business model.

[27] The Collectives state that there is no reason to change the rate from that fixed in the 2007– 2009 Pay Audio Tariff, since the basic pay audio service has not changed. Basic pay audio has been augmented with several additional features, including the ability to simulcast pay audio away from the television set. Pay audio uses more music than it did in 2002.

E. WITNESSES

[28] A total of ten witnesses provided either oral or written testimony, or both. While the Board does not formally qualify witnesses as “expert,” five witnesses provided testimony in their capacity as experts in a particular domain, and five as fact witnesses. We list them below in the order in which they appeared before us.

i. Jeff Vidler

[29] Jeff Vidler is a research consultant. He was commissioned by the Objectors to “provide expert commentary informed by consumer research on where the pay audio service, Stingray Music, is situated among today’s expanding range of music services.” [12]

[30] Mr. Vidler reviews changes in the music listening landscape from 1997, when pay audio services were first introduced in Canada, to 2016. He also presents findings from a consumer study conducted in October 2016 to obtain an end-user perspective on the Stingray Pay Audio Service. Mr. Vidler concludes that listener benefits of Stingray Music Pay Audio are similar to those of broadcast radio, non-interactive, and semi-interactive services, as they are “closely associated with delivering a lean-back background music experience.” [13]

ii. Geoff Wright

[31] Geoff Wright has previously been an independent consultant for Stingray, and Director of Corporate Strategy at Bell Canada. He was commissioned by Stingray and the BDUs to “provide an expert assessment of the value of the Stingray Music Pay Audio Service and the Stingray Music Web Player, as considered by the BDUs.” [14] Mr. Wright is an independent consultant for Stingray. [15]

[32] Mr. Wright concludes that there were growing competitive substitutes including streaming music services, during the period covered by the Proposed Pay Audio Tariffs.

iii. Dr. Leslie Marx

[33] Dr. Leslie Marx is a professor at the Fuqua School of Business at Duke Un iversity, and a partner at Bates White, LLC, a professional-services firm that performs economic and statistical analysis in a variety of industries and forums.

[34] Dr. Marx provides expert testimony on the methods to price the right to communicate works and sound recordings in the context of the pay audio service and simulcasts. In her report, she concludes that “a benchmark analysis is an appropriate methodology for determining reasonable royalty rates in this matter.” [16] She is also of the opinion that limitations of data required to calibrate a theoretical model make the use of such theoretical models inappropriate in this proceeding.

iv. Jean-Pierre Trahan

[35] Mr. Jean-Pierre Trahan is the Chief Financial Officer at Stingray. He oversees all finance and accounting matters at Stingray, and supervises directors and staff responsible for finance, financial information and compliance, and controllers. [17]

[36] Mr. Trahan testified about

• Stingray’s overall business, including the domestic residential pay audio service ;
• the rationale employed by Stingray for undertaking the revenue allocation exercise; and
• the profitability of the domestic residential pay audio service.

v. Mathieu Péloquin

[37] Mathieu Péloquin is the Senior Vice-President, Marketing and Communications at Stingray. He oversees all communications, marketing, research and content matters, including music programming.

[38] Mr. Péloquin describes the Stingray Music Pay Audio Service and, how it is marketed and sold.

vi. Andrew Michelin

[39] Andrew Michelin is a partner at Richter L.L.P., an accounting and business advisory firm.

[40] Mr. Michelin was commissioned by the Objectors to provided expert testimony on

• whether revenue allocation among products and services sold as a bundle is an acceptable practice;
• whether Stingray’s decision to allocate revenue it receives for the bundle of services among each of those services is reasonable; and
• the manner in which Stingray could implement a revenue allocation among each of its services.

vii. Dr. Michael J. Murphy

[41] Dr. Michael J. Murphy is Full Professor at Ryerson University’s RTA School of Media and a professional engineer. Dr. Murphy testified as an expert in digital technology applications in media and broadcasting.

[42] Dr. Murphy was commissioned by the Collectives to

• describe the technology in use in 2016 to deliver Digital Pay Audio Services in Canada; and
• compare the similarities and differences between the end-user experience and use of music on Digital Pay Audio Services and other music services available to Canadian consumers.

viii. Elsie Mbuoben

[43] Elsie Mbuoben is the Finance Manager in the Finance Department of Re:Sound. Her role includes collecting and processing royalty payments and reporting from BDUs pursuant to the 2007–2009 Pay Audio Tariff.

[44] Ms. Mbuoben testimony addressed

• the Stingray revenue allocation and its effect on the royalties paid;
• the effect of the Objectors’ proposed rate reduction on the royalties payable;
• Stingray’s non-payment of webcasting royalties;
• demands by certain BDUs for a retroactive royalty adjustment and credit;
• historical royalty payments;
• changes in music use by pay audio services since 2002; and
• proposed changes in the administrative provisions.

ix. Jamie Leacock

[45] Jamie Leacock is the Manager of Media in the Licensing Department at SOCAN. He is responsible for the licensing of accounts and collection of revenue from media and new media services.

[46] Mr. Leacock’s testimony addressed SOCAN’s Pay Audio Tariff 26 and SOCAN’s communication with Stingray and BDUs both prior to, and following, Stingray allocating its revenues amongst the various components in its Suite of Services.

x. Dr. Marcel Boyer

[47] Dr. Marcel Boyer is Emeritus Professor of Economics at the University of Montréal, whose published research is on various topics, including efficient organizations, innovation and competition, public policy, competition policy, cartel fining, environmental liability, and copyrights.

[48] Dr. Boyer was commissioned by the Collectives to

• review the Board’s prior decisions relating to the pay audio service tariffs;
• discuss the relevant economic theory and concepts underlying the determination of music royalty rates;
• provide an application of the above in this proceeding; and
• provide an opinion on expert reports of Dr. Marx and Mr. Wright.

xi. Alan T. Mak

[49] Alan T. Mak is a chartered accountant and chartered business valuator. He is the Principal at Rosen & Associated Limited, whose services include forensic accounting, and business valuation.

[50] Mr. Mak was commissioned by the Collectives to provide expert opinion on:

• The financial performance of Stingray, including observed revenue and profit trends, and the company’s profitability and cash flows; and
• The reasonableness, from accounting and financial reporting perspectives, of the Objectors’ allocation of revenue to the Stingray Music service and other components of the bundle of services provided by Stingray to BDUs.

III. ISSUE 1: THE RATE BASE APPLICABLE TO PAY AUDIO (“REVENUE ALLOCATION”)

[51] The Rate Base is the sum of money to which a royalty rate applies. In many tariffs, the Rate Base constitutes the revenues attributed to the activity that is the target of the tariff. The mechanics of this attribution is usually not fixed in the tariff itself.

[52] For the reasons that follow, we conclude that only those portions of affiliation payments that are attributable to the pay audio service constitute the Rate Base in the royalty formula, and that this is so for the entire period of the Proposed Pay Audio Tariffs.

[53] Furthermore, we conclude that this allocation of affiliation payments should be fixed in the Tariff, be based on the allocation originally developed by Stingray for its client BDUs, and be adjusted based on the particular combination of services a particular BDU receives during a given year.

[54] After providing context for the dispute regarding the allocation of affiliation payments, we consider the following questions:

• What portion of affiliation payments should be included in calculating royalties?
• To which years should this apply;
• Whether we should fix this portion in the Tariff; and
• How this portion should be determined.

A. WHY IS REVENUE ALLOCATION AN ISSUE?

[55] When it started operations in 2007, Stingray only provided one service to BDUs: the Karaoke Channel. Stingray acquired the pay audio service Galaxie, from the Canadian Broadcasting Corporation in 2009. The pay audio service provides programming with a number of channels, which play uninterrupted music 24 hours per day.

[56] From 2010 onward, [18] Stingray began offering additional services, including internet-based simulcasts of its pay audio programming, semi-interactive webcasts based on pay audio programming, and several services unrelated to pay audio programming, such as Stingray Ambiance, Stingray Music Videos and Concert TV. As described above at paragraph [6], these services are sold as a bundle to BDUs.

[57] The Proposed Pay Audio Tariffs use a royalty formula which multiplies a rate base (also referred to by the Parties as a revenue base or royalty base) by a percentage. This formula results in a dollar amount, which represents the royalties to be paid for a given period. For example, subsection 5(1) of proposed SOCAN Pay Audio Services Tariff, 2010 reads as follows:

[T]he royalties payable to SOCAN are 12.35 per cent of the affiliation payments payable during a month by a distribution undertaking for the transmission for private or domestic use of a pay audio signal.

[58] The affiliation payments referred to in this proposed tariff are therefore the rate base.

[59] The Proposed Pay Audio Tariffs do not apply to other communications of musical works and sound recordings made by Stingray or the BDUs. Communications to the public that occur in relation to other components of the bundle may be the subject of other tariffs—or no tariff at all (e.g., if no works or subject-matter in the repertoire of a collective society is used).

[60] The Objectors and Collectives disagree on what portion of affiliation payments received by Stingray from BDUs should be considered when determining the royalties to be paid for the pay audio service. The Collectives submit that the entirety of these payments be used in the formula; the Objectors submit that only the portion attributable to the pay audio service be used.

[61] Were Stingray to provide only pay audio programming services to BDUs, this distinction would be without a difference: the affiliation payment attributable to pay audio programming would comprise the entirety of the affiliation payment. However, the Objectors remitted royalties based on the entire affiliation payments even though the services received by the BDUs included more than pay audio programming.

B. WHAT PORTION OF AFFILIATION PAYMENTS SHOULD BE INCLUDED IN CALCULATING ROYALTIES?

[62] We conclude that only those affiliate payments attributable to the provision of the pay audio service should form the Rate Base for the purpose of the calculating royalties. We find that it would be unfair to include in the revenue base those revenues generated by activities not subject to the proposed tariffs under consideration.

[63] The Collectives’ main arguments against using a portion (as opposed to the entirety of) of the affiliation payments in the royalty formula for the pay audio service is based are:

• The wording of previously approved tariffs that were applicable to the pay audio service did not require an allocation;
• Expert opinion indicates that such an allocation would not be reasonable under principles of accounting and financial reporting; and
• Limiting the rate base to revenues attributable to the pay audio service would reduce the amount of royalties.

[64] The Objectors’ main argument is that the bundle of services for which affiliate payments are made includes more than only pay audio. The effect of using the entirety of the affiliation payments in the formula would result in either applying the wrong royalty rate to services not connected to the pay audio service, or even payments being owed in respect of services to which a tariff does not apply. [19]

[65] We consider these arguments in turn.

i. Previously Approved Tariff: 2007–2009 Pay Audio Tariff

[66] The Collectives point to the 2007–2009 Pay Audio Tariff as support for the proposition that it has previously been the case—and should therefore continue to be the case—that it is the entire affiliation payment that is used in calculating royalties.

[67] The 2007–2009 Pay Audio Tariff provides that

4. (1) […] the royalties payable to SOCAN and NRCC respectively are 12.35 per cent and 5.85 per cent of the affiliation payments payable during a month by a distribution undertaking for the transmission for private or domestic use of a pay audio signal.

[68] The Collectives argue that the term “affiliation payments” in the 2007–2009 Pay Audio Tariff refers to the “total of the wholesale per-subscriber fees that the distribution undertakings pay to Stingray for the licence to carry the pay audio signal.” [20] They go on to submit that there is no basis under the 2007–2009 Pay Audio Tariff to pay royalties on anything other than the entire affiliation payment. [21]

[69] The context of the phrase “for the transmission for private or domestic use of a pay audio signal” leads us to conclude that it is an adjectival phrase as part of the subjective completion “of the affiliation payments payable during a month.” Subsection 3(1) already provides that the royalty payments are “for the communication to the public by telecommunication of [works and sound recordings], in connection with the transmission by a distribution undertaking of a pay audio signal for private or domestic use.” Its repetition in subsection 4(2), where it is qualified by “where the distribution undertaking is [a small transmitter/low-power transmitter] reinforces this conclusion.

[70] In any case, while a previously approved tariff may constitute a useful proxy, it is not determinative.

ii. Accounting and Financial Reporting Standards

[71] Mr. Mak stated his expert opinion: “the allocation of revenue to individual services by the Objectors is not reasonable from a financial reporting or accounting perspective.” [22] Mr. Mak testified that, in his expert opinion, allocation of revenue amongst Stingray’s Suite of Services is not consistent with the historical business transaction documents; not consistent with the pricing, costing, accounting and financial reporting of Stingray’s services to the BDUs, as prepared by Stingray’s management; and not consistent with how Stingray’s Suite of Services are or may be treated for accounting purposes under International Financial Reporting Standards. [23]

[72] Further, Mr. Michelin gave the opinion that financial reporting considerations do not and should not dictate how a business should sell its products or services (i.e. as bundled services or individually). Here, Stingray has decided to sell its services as a bundle, presumably because it believes that this is in the best interests of its business. [24]

[73] While financial and accounting requirements may be instructive, they do not, in our view, decide the question. Instead, we agree with Mr. Michelin: reporting rules made for other contexts should not dictate the calculation of royalties in the context of a tariff proceeding.

[74] The Collectives also argue against revenue allocation on the basis that not including the entire affiliation payment in the royalty formula would decrease royalties. [25] There is not much to be said about this line of reasoning: the dollar difference between parties’ respective positions is not helpful in assessing the fairness of any particular position.

[75] Collectives argue that—were the Board to agree with the entirety o f the Objectors’ submissions—the difference between their position and that of the Objectors would amount to a “loss” to them. Using the same logic, the Objectors could argue this same figure demonstrates a “loss” to their industries—were the Board to agree with the entirety of the Collectives’ submissions. To them, it represents an overpayment. [26]

[76] This figure does not represent a “loss” for either: neither collective societies nor users are automatically entitled to pay or receive a particular pre-established amount under a tariff.

iii. Undesired effects without revenue allocation

[77] In Théberge v. Galerie d’Art du Petit Champlain inc., [27] the Supreme Court observed that “it would be as inefficient to overcompensate artists and authors for the right of reproduction as it would be self-defeating to undercompensate them.” [28] In this case, failing to adjust the rate base to limit it to those affiliation payments associated with pay audio would amount to the overcompensation the Court warned about.

[78] Moreover, the ill effects of including all affiliation payments in the pay-audio rate base go beyond this proceeding. Some of the affiliation payments pertain to activities for which other tariffs apply. It would not be inconceivable for the multiplicity of tariffs for which the affiliation payments serve as a rate base to cause the total royalty bill for all such tariffs to exceed the affiliation payments themselves.

iv. Conclusion

[79] For the foregoing reasons, we fix the rate base to the amount attributable to the provision of the pay audio service by Stingray to BDUs.

C. WHEN ALLOCATION SHOULD START

[80] We conclude that allocation of revenue is to be done for the entire period of the Proposed Tariffs, that is, for each year from 2010 to 2016.

[81] The Collectives submit that if revenue allocation is to apply to affiliation payments—revenue allocation should only begin in 2015, and apply to subsequent years. [29] They argue that this restriction is justified on the basis that Stingray itself only began implementing such revenue allocation for the purposes of calculating royalties for the pay audio service in 2015. [30]

[82] In their reply, the Objectors indicated that they

advised the Board that Stingray has or is capable of generating the requisite cost data to enable an objective allocation going back to 2014 […] and have advanced a principled and economically sound approach for dealing with prior years for which there is no data. [31]

[83] We conclude that it would not be fair to implement such a limitation. The fact that Stingray did not suggest to the BDUs that they retroactively apply the revenue allocation from 2010 does not prevent us from doing so. Indeed, if we did not do so, it would be equivalent to imposing a revenue allocation of 100% of affiliation payments to pay audio services for the years 2010 through 2014, despite the evidence that other services were included in the bundle purchased by BDUs during that time. [32]

D. SHOULD THE BOARD FIX THAT PROPORTION IN THE TARIFF?

[84] We conclude that the Board has the power to fix the proportion of revenues that are to constitute a rate base in a tariff, and that it is appropriate for us to do so here.

[85] Throughout this proceeding, the Objectors have maintained the position that the Board need not and should not fix a revenue allocation in the Tariff, but should only state whether a revenue allocation is permissible. The Collectives urge the Board to prohibit revenue allocation, or, in the alternative, fix the revenue allocation in the Tarif f.

[86] In closing arguments, the Objectors raised three reasons why the Board should not fix a particular revenue allocation in the Tariff. First, the content of the bundle of services changes every year. A correct allocation in one year may be wrong in another year. Second, the Proposed Pay Audio Tariffs are of general application, and there is a second entity that has received a pay audio licence from the CRTC. That licensee should not use an allocation designed by Stingray and based on Stingray’s business practices. Finally, the Objectors claimed that the question of revenue allocation is one of tariff implementation and enforcement; such matters are outside the Board’s jurisdiction. [33]

[87] The Collectives did not directly refute the points made by the Objectors but advanced the following reasons to fix a specific allocation in the Tariff. First, licensees should not be at liberty to change the rate base unilaterally every time they make a minor change to conduct their business. [34] Second, arbitrary allocations like the one Stingray produced in 2015 lead to too much unpredictability for the Collectives. [35]

[88] We consider the arguments raised, in turn.

i. Difficulties in implementation

[89] The first and the second arguments of the Objectors can be addressed by the following observation: the effective period of Tariff is entirely in the past. For the pay audio service, this period is from 2010 to 2016. As regards changes in the contents of a bundle of services received by a particular BDU, even if a bundle changes every year, those changes are known and have been filed as part of this proceeding. [36]

[90] Furthermore, as regards the possibility that another entity operates a pay audio service, no pay audio service has begun operations during the period of this tariff, other than the one operated by Stingray. An application by Kosiner Venture Capital Inc. (Kosiner) for the right to operate a national pay audio programming undertaking was approved by the CRTC on Octob er 1, 2015. [37] However, in a 2017 decision of the CRTC relating to satellite radio ownership, the CRTC described that undertaking as “licensee of an unlaunched national pay audio programming service.” [38] Since the Kosiner service was unlaunched on April 26, 2017, it follows that it was unlaunched throughout the relevant period of 2010 to 2016.

ii. Is it an enforcement matter, outside the Board’s jurisdiction?

[91] The Objectors are correct that the Board does not have jurisdiction to enforce its tariffs, once they are approved. In Private Copying Tariff Enforcement 2001–2003, the Board held that it does not have the power to order a non-compliant user of its tariffs to comply and to find that user in contempt if it does not comply. [39] In Re:Sound 6.A (Application to Vary), 2012, the Board declined to vary a tariff, stating that “to the extent the application raises issues of tariff enforcement, these generally are for the courts, not the Board, to decide.” [40]

[92] However, we do not agree that specifying the particular manner in which royalties are to be calculated amounts to illegitimately stepping into the domain of enforcement. Furthermore, the Copyright Act (the “Act”) states:

[t]he Board shall […] approve the proposed tariff after making any alterations to the royalty rates and the related terms and conditions, or fixing any new related terms and conditions, that the Board considers appropriate. [41]

[93] We therefore conclude that it is within the Board’s jurisdiction to do so.

iii. Is it appropriate to do so?

[94] In other approved tariffs, the Board has explicitly not fixed a particular allocation of revenues attributable to the activity under consideration. For example, in Online Music Services (2010– 2013), the Board fixed the royalties payable as a percentage of the “gross revenues attributable to the operation of that service.” [42] It did not specify how such attribution should occur, only explaining that revenues attributable to services not covered by the tariff would not form part of the revenue base for that tariff. [43]

[95] Were the issue of allocation of revenues not such a significant point of dispute among the Parties, we may have done similarly here. However, the issue of what portion of affiliation payments are to be used in calculating royalties for the pay audio service was so contested—to the point where Parties believed it was necessary to hear the testimony of expert witnesses —that we conclude that is appropriate and preferable to do so.

E. HOW SHOULD THE BOARD DETERMINE THE REVENUE ALLOCATION?

[96] Having concluded that it is appropriate to fix in the Tariff a particular portion of affiliation payments attributable to the pay audio service, we need to decide the manner in which this allocation will be determined. In order to do so, we consider the following questions:

• What approach should be used to determine revenue allocation;
• How should this approach to allocation account for Ubiquicast, a server technology used by Stingray; and
• How should the selected approach be adapted for cases to take into account the fact that different BDUs received different services in different years?

[97] We establish the allocation based on the approach initially developed by Stingray. We first adjust the approach to take into consideration the fact that Ubiquicast is used by the pay audio service. Second, where a BDU does not offer all services, the revenues that would otherwise be allocated to those unoffered services are proportionally redistributed among those that are offered.

i. What approach should be used to determine revenue allocation?

[98] We conclude that the approach initially developed by Stingray (the “Original” allocation) is the best among those considered in this proceeding.

a. Methodologies and Data Filed with the Board

[99] The four approaches for allocating affiliation payments to the pay audio service, filed by the Parties over the course of the proceedings, were:

• The “2015” allocation from the Objectors; [44]
• The “2016” allocation from the Objectors; [45]
• The “Usage” allocation from the Collectives; [46] and
• The “Original” allocation from the Objectors. [47]

[100] Both the 2015 allocation and 2016 allocation were filed in responses to questions from the Board. Both are also subject to motions by the Collectives. In the respective motions, the Collectives argue that they have not had an opportunity to test this evidence. Given that we conclude that there is other evidence on which we can determine an allocation, we rely on neither the 2015 allocation nor the 2016 allocation. These motions are therefore moot.

[101] Below, we analyze the relative merit of the “Usage” and “Original” allocations. We conclude that it is possible to use the “Original” allocation as a method for allocating affiliation payments to the pay audio service, and that it is better to use than the “Usage” allocation.

b. Usage-based allocation

[102] In response to Notice CB-CDA 2018-040, the Collectives proposed that the allocation be based on usage. Data are routinely collected by Stingray for various measures of usage, including the number of times an end-user turned on a pay audio channel, number of listeners, and duration of listening. [48] According to the Collectives, these measures of usage relate to Stingray Music, Stingray Web, and Stingray Mobile. [49]

[103] Based on the usage data, the Collectives proposed a “default allocation” for the following revenue categories, to be used whenever the data were insufficiently usable, which is summarized in Table 3.

Table 3: Usage-based Allocation

Table 3: Usage-based Allocation

Music	Mobile	Ambiance	Music Videos	Concert TV	Karaoke	Ubiquicast
92%	(nil)	1%	5%	1%	1%	(nil)

[104] The Objectors raised two principal arguments. First, it is not necessarily the case that end-user consumption reflects value to the user, namely the BDUs or Stingray. Second, because the various services in a bundle are so different one from the other, it is not clear that a measure of usage that treats them the same (an hour tuned to one versus an hour tuned to another) is appropriate. [50] We find these arguments convincing. Usage data do not reflect “demand,” as that term is defined by economists; that is, the choice of consumption when there is an opportunity cost thereof. [51]

[105] Furthermore, the price a consumer is willing to pay for an hour of radio is different from the price they are willing to pay for an hour of a movie, and different from the price they are willing to pay for an hour of a video game. Indeed, this price may differ by orders of magnitude.

[106] The question of the relative value of different services has been addressed by the Board before in a different context. In Online Music Services (2007–2010), [52] the Board set the price of a video-clip (that is, the video equivalent of a song) using the ratio of the mid-price audio track to the mid-price video-clip. [53] The specific ratio used by the Board in that decision does not matter, except insofar as it is not 1:1. Online Music Services (2007–2010) demonstrates that the “value” of a minute consumed of a particular medium is unlikely to be eq ual to the “value” of a minute consumed of a different medium. The greater the differences between the media, the greater this difference in value can be expected to be.

[107] We therefore conclude that we cannot use a methodology of measuring usage of audiovisual services and comparing it with usage of audio-only services as a means of determining revenue allocation of the affiliation payment.

c. The “Original” Allocation

[108] Another allocation to consider is the one which Stingray directed the BDUs to use in the context of their correspondence with the Collectives. This allocation has seven revenue categories for its services, and is shown in Table 4. [54]

Table 4: The “Original” Allocation

Table 4: The “Original” Allocation

Music	Mobile	Ambiance	Music Videos	Concert TV	Karaoke	Ubiquicast
40%	10%	20%	5%	5%	5%	15%

[109] We note that there is not necessarily a one-to-one mapping between these revenue categories and the services, as identified in these proceedings: it is possible for one or more of these categories to encompass more than one service. For example, in this allocation, it is not evident where simulcasts are placed: in the Music or Mobile revenue category. In either case, one of these revenue categories will include more than one service.

[110] All Parties had ample opportunity to test this evidence. The majority of criticism levelled by the Collectives related to the concept of revenue allocation in general, rather than this allocation in particular. Since we conclude above that there is to be revenue allocation and that we are to fix a revenue allocation in the context of setting the present tariff, those criticisms are not on point with the present issue.

[111] There is one criticism that bears mention, namely that “Stingray has not provided the data or analysis that it claims to have relied upon to develop its revenue allocation model.” [55] Those data and analysis are discussed in the oral evidence, but the data were never provided for the Collectives to test formally. We summarize the comments regarding those data below.

[112] In his report, Mr. Michelin writes that Stingray records costs for specific services and does maintain employee time sheets to track employee time spent on a project or service. [56]

[113] On direct examination, Jean-Pierre Trahan described the exercise of preparing the revenue allocation as a collaborative one between the lawyers, the salespeople, the engineers, and the accountants. [57] Among other things, the time sheets prepared by each engineer were examined. Thereafter, overhead costs were applied to each service on a prorated basis. [58] The Collectives did not ask Mr. Trahan any questions about this process of creating an allocation.

[114] On direct examination, Andrew Michelin described a hypothetical revenue-allocation exercise based on costs on a service-by-service basis. [59] ^{, [60]} ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| Notice CB-CDA 2017-098 asked for Mr. Michelin’s spreadsheet for the hypothetical exercise.

[115] In our view, an allocation based on costs, such as those described hypothetically by Mr. Michelin |||||||||||||||||||||||||||||||||||||||||||||||||||||| is the most appropriate way to undertake the allocation in the context of this proceeding. We find the Original Allocation credible and adopt it for the purposes of this proceeding.

ii. How should revenue allocation account for Ubiquicast?

[116] We find that Ubiquicast is used for the services in the Music and Ambiance revenue categories and that revenues must be reallocated from Ubiquicast to those categories.

a. How is Ubiquicast Used?

[117] As an expert witness, Dr. Murphy described Ubiquicast as a “specialized media stream ing playout server developed and supplied by Stingray that is located at the BDU premises (head end or equivalent termination point).” The purpose of this server is to avoid having to transmit multiple streams to BDUs in real-time from Stingray’s centralized servers, music files are stored on Ubiquicast hard drives. [61]

[118] According to Dr. Murphy, this has the effect of reducing ongoing BDU-Stingray networking costs, as files only have to be delivered once and can then be repeated locally as often as required. It has the added benefit of reducing downtime if the communication link fails or there is some other central technological failure, as the local Ubiquicast server is able to continue to operate independently for a period of time. [62]

[119] Dr. Murphy’s report states that in addition to the pay audio service, numerous other Stingray offerings also use Ubiquicast.

[120] However, Mr. Peloquin testified that Ubiquicast is not used for on-demand services. Furthermore, in Canada, only pay audio and linear Stingray Ambiance ch annels use Ubiquicast.

[121] Given Mr. Peloquin’s direct personal knowledge of these facts, we accept his testimony regarding which services actually use Ubiquicast.

[122] We therefore conclude that, during the effective period of the Proposed Pay Audio Tariffs, Ubiquicast was used in relation to the Stingray Music and Stingray Ambiance services.

b. Should Revenue be Allocated to Ubiquicast?

[123] Mr. Mak, an expert witness for the Collectives, testified:

Stingray’s allocation of revenue to Ubiquicast is particularly curious. We understand Ubiquicast is how Stingray’s services are distributed. Ubiquicast is not sold separately, nor does Stingray distribute its music services without Ubiquicast. The Ubiquicast server is necessary for a BDU to receive and distribute Stingray’s services, but it is presumably not a product that the BDUs would seek to buy independently. |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| Thus, it seems clear that Ubiquicast is not a distinct performance obligation from Stingray’s bundled services. In our view, it seems quite clear that Ubiquicast is highly integrated with other elements in Stingray’s bundle of services. It should not be treated as a distinct and separate good (or service). [63]

[124] The Objectors argue that this would not be fair. In their submission, the Collective’s approach would amount to paying royalties on revenues earned by providing non-copyright goods, hardware and related services. [64]

[125] We agree with the Collectives: inasmuch as Ubiquicast is used to transmit pay audio, affiliation payments associated therewith are “in connection” with pay audio. The manner in which the “Original” allocation was developed is largely based on costs. [65] We have accepted evidence that Ubiquicast is used to transmit the two linear services: Stingray Music and Ambiance. Thus, by creating a Ubiquicast cost category and an associated revenue category, the Original Allocation removes certain costs—in this case, those related to the Ubiquicast servers—and distorts the allocation. We therefore conclude that revenues allocated to Ubiquicast in the Original Allocation must be themselves reallocated among those services that use Ubiquicast. We now proceed to determine which portion of revenues attributed in the Original Allocation to Ubiquicast should be reallocated to pay audio.

c. How to Reallocate Revenues from Ubiquicast

[126] Under the Original Allocation, the Music revenue category has an allocation of 40%, Ambiance has an allocation of 20% and Ubiquicast has an allocation of 15%. In the absence of more precise evidence regarding the usage of Ubiquicast, we reallocate these 15 % among the Music and Ambiance categories, proportionally to their own allocation percentages.

[127] This results in the following:

• If the BDU used Ubiquicast and offered Stingray Music and Ambiance: the portion reallocated to Music is (40/60) 15% = 10%, resulting in a restated revenue allocation for Music of 50%.
• If the BDU used Ubiquicast and offered Stingray Music but not Ambiance: the 15% allocated to Ubiquicast is entirely reallocated to Music, resulting in a restated revenue allocation for Music of 55%.
• If the BDU did not use Ubiquicast and offered Stingray Music, whether or not with Ambiance, the revenue allocation for Music requires no restatement: it remains 40%.

iii. How should the Original Allocation be adapted where a BDU does not offer all six services?

[128] We implement a method in the Tariff to ensure revenues are distributed only amongst services that a BDU actually received.

[129] The Original Allocation was based on the hypothetical situation where a BDU received (and offered to its own customers) services in all six revenue allocation categories—and used Ubiquicast. It would not be fair to use the same allocation in instances where not all of the six categories of services were provided to a BDU. To do so would amount to allocating revenues away from Stingray Music towards a service that was not received by the BDU. The appropriate method of reallocation proceeds in two steps.

[130] In respect of a BDU, for a particular period,

• select the scenario in paragraph [127] corresponding to the combination of services it received from Stingray in that period; and
• redistribute revenues from those services that the BDU did not receive, to those that it did, pro-rata.

[131] Those revenues that are now attributed to the Music revenue allocation category are those that—for the purposes of this Tariff —will be treated as the portion of affiliation payments properly attributable to the pay audio service.

[132] In the Tariff, the second step from paragraph [130] is facilitated by breaking it down into three smaller sub-steps.

[133] First, treating the percentage for each category as dimensionless “category points.” This results in one of the category points tables below (Table 5, Table 6, and Table 7).

Table 5: Category – Pay Audio, Ambiance, and Ubiquicast provided

Table 5: Category – Pay Audio, Ambiance, and Ubiquicast provided

Category	Music	Mobile	Ambiance	Music Videos	Concert TV	Karaoke
Points	50	10	25	5	5	5

Table 6: Category points – Pay Audio and Ubiquicast provided, without Ambiance

Table 6: Category points – Pay Audio and Ubiquicast provided, without Ambiance

Category	Music	Mobile	Music Videos	Concert TV	Karaoke
Points	55	10	5	5	5

Table 7: Category points – Ubiquicast not provided

Table 7: Category points – Ubiquicast not provided

Category	Music	Mobile	Ambiance	Music Videos	Concert TV	Karaoke
Points	40	10	25	5	5	5

[134] Second, removing those categories not received by the BDU during the relevant period.

[135] Third, dividing the category points for the Music category (50, 55, or 40) by the total of the points for categories containing services received by the BDU during the relevant period. This fraction is called the pay audio affiliation multiplier in the Tariff. It represents the fraction of the affiliation payments that will be deemed to be attributable to the pay audio service.

[136] We implement this allocation methodology in the Tariff. It is important to note that the category points are only intended to permit the determination of the Rate Base for the pay audio service. In particular, the category points for the Mobile category are not intended to be used to calculate the rate base for near-simulcasts, as explained in detail at paragraphs 250 to 253, below.

iv. Example: A Hypothetical BDU

[137] We provide the following example in the hope that it may assist Parties with implementing the Tariff.

[138] Since we only have data for the largest BDUs, and since those data are highly confidential, it is preferable to show a hypothetical example. In our example, we consider such a hypothetical BDU, and the category of services it received from Stingray over time. Table 8 shows those data, with 1 indicating that it received a service in that revenue allocation category from Stingray, and 0 that it did not.

Table 8: Offerings of the Hypothetical BDU

Table 8: Offerings of the Hypothetical BDU

Year	Music	Mobile	Ambiance	Music Videos	Concert TV	Karaoke	Ubiquicast
2010	1	0	0	1	0	1	1
2011	1	0	0	1	0	1	1
2012	1	0	0	1	1	1	1
2013	1	0	0	1	1	1	1
2014	1	1	0	1	1	1	1
2015	1	1	1	1	1	1	1
2016	1	1	1	1	1	1	1

[139] The allocation for 2016 would be identical to that of 2015, since the hypothetical BDU offers the same services in both years.

[140] In 2014, the hypothetical BDU did not offer Ambiance and did use the Ubiquicast server. This combination leads us to use the category points in Table 6, just as it does for the remaining years from 2010-2013. Music has 55 category points. The sum of the category points for all services offered is 80. The pay audio affiliation multiplier in 2014 for the hypothetical BDU for Music would be 55/80 or 68.75%.

[141] In 2013, the hypothetical BDU offered neither Ambiance nor Mobile. The pay audio affiliation multiplier for Music would be 55/70 or 78.57%. This would also be the case in 2012, since the hypothetical BDU offered the same services in 2012 as it did in 2013.

[142] In 2011, the hypothetical BDU offered neither Ambiance, nor Mobile, nor Concert TV. The pay audio affiliation multiplier would be 55/65 or 84.62%. This would also be the case in 2010, since the hypothetical BDU offered the same services in 2010 as it did in 2011.

[143] Table 9 summarizes the revenue allocations calculated above.

Table 9: Allocations of Revenue to Stingray Music for the Hypothetical BDU

Table 9: Allocations of Revenue to Stingray Music for the Hypothetical BDU

Year	Allocation
2010	84.62%
2011	84.62%
2012	78.57%
2013	78.57%
2014	68.75%
2015	50.00%
2016	50.00%

IV. ISSUE 2: WHAT SHOULD THE ROYALTY RATE FOR PAY AUDIO BE?

[144] We fix the Gross Royalty Rate for pay audio by using the 2007–2009 Pay Audio Tariff as a proxy. We adjust this proxy to account for the significant increase in competition during the period of the Proposed Pay Audio Tariffs. We also fix the Gross Royalty Rate for small cable transmission systems as 50% of the rate applicable to other BDUs.

[145] The Board has relied on proxies (or “benchmarks,” as they were also referred to in this proceeding) on many previous occasions. The Board explained how a proxy can be used in Access Copyright – Tariffs for Post-Secondary Educational Institutions, 2011-2017:

We use the term “target” for the price (or tariff) the Board is tasked with setting. We use the term “proxy” for the price in a market related to the target. A proxy is said to be good if the proxy is determined in a pseudo-competitive market and the proxy market is “similar” to the target market. To the extent that these two conditions are not met, the proxy may have to be adjusted to determine a fair and equitable target tariff. The more uncertain the adjustments necessary to determine a fair and equitable tariff are, the less the proxy is usable. If the adjustments are subject to a great deal of uncertainty, the proxy may not be usable at all. [66]

[146] We adopt the same approach here. The two main positions advanced in this proceeding were that the rate for pay audio should be based on rates previously fixed for webcasts in other tariffs, and that the rate for pay audio should be based on the rate previously fixed for pay audio in previous pay audio tariffs.

[147] We consider these two sources for a proxy in turn. We conclude that the 2007–2009 Pay Audio Tariff should be used as a proxy for pay audio. Finally, we consider the question of the rate for small cable transmission systems.

A. SHOULD A TARIFF FROM THE WEBCASTING MARKET BE USED AS A PROXY?

[148] We conclude that a tariff from the webcasting market should not be used as a proxy for pay audio in this proceeding.

[149] The economic expert for the Objectors, Dr. Marx, states that the target, pay audio, is similar to non-interactive and semi-interactive webcasting, based on their listening experiences and their ability to cannibalize music sales. [67] As such, she proposes the royalty rate fixed in Online Music Services (2010–2013), a tariff that applies to webcast services, as a proxy for the pay audio rate.

[150] Mr. Vidler’s testimony supports the proposition that Stingray Music, broadcast radio, and non-interactive and semi-interactive webcast services are all more closely associated with delivering a lean-back background music experience, when compared to fully interactive services, which are much more likely to be linked to the ability to find songs or artists on demand. [68]

[151] However, while we agree the comparison of listening experiences may be a reasonable one, the question of cannibalizing music sales is ultimately an empirical one. Without additional evidence in this proceeding, we are unable to evaluate the strength of this claim. Dr. Boyer also notes the absence of data analysis in his critique of Dr. Marx’s report. [69]

[152] Dr. Marx’s analysis of the similarities between the pay audio and webcasting markets concludes that there is substantial similarity. In her opinion, the demand characteristics of pay audio, radio, and non-interactive or semi-interactive webcasting are comparable. On the supply side, delivery technologies differ, but they provide a similar end-user experience. [70]

[153] Dr. Marx’ analysis of the supply side is limited to the delivery technologies. She does not analyze the differences between the business models of pay audio and non - and semi-interactive webcasting. In Re:Sound – Tariff 8 (Non-Interactive and Semi-Interactive Webcasts), 2009-2012, [71] the Board noted that webcasters earn revenues primarily through advertising. [72] Pay audio earns revenue through affiliation payments from the BDUs. The transposition of a percentage rate from one business model to another can be conceptually difficult. [73]

[154] Dr. Boyer also sees that transposition as problematic. He testified that, in general, all industries should pay the same price for the same input. Using a revenue-based rate set for another industry will not achieve this: there is no reason to believe that the proper price to be paid for the same inputs, namely the right to communicate musical works and sound recordings, would correspond to the same percentage of revenues irrespective of the characteristics of the underlying industries in question. [74] If the two industries have different shares of value-added, Dr. Boyer argues, such equality is even less likely. [75]

[155] We do not accept Dr. Boyer’s critique in its entirety. While it is true, in general, that a percentage is not a price, [76] a proxy analysis need not only proceed in the domain of prices. Taking Dr. Boyer’s critique seriously would mean completely rejecting the use of proxy analysis to adapt a percentage rate from one market to another. We do not need to reach such a sweeping conclusion; nor do we do so.

[156] Dr. Marx emphasizes the question of the similarity between the markets for pay audio and webcasting. However, the supply-side issues, raised by Dr. Marx as a minor point, do not seem minor to us. Nor is the question of business model a minor difference either. Furthermore, because Dr. Marx did not discuss the possible adjustments (other than an adjustment for the profit of BDUs), substantial uncertainty remains as to how to adjust her chosen proxy to fit the target market.

[157] Given these deficiencies, along with those related to evidence of cannibalization, we find that webcasting is not a good proxy for pay audio in this proceeding.

B. SHOULD A TARIFF FROM THE PAY AUDIO MARKET BE USED A PROXY?

[158] We conclude that it is reasonable to fix the Gross Royalty Rates for pay audio based on the 2007–2009 Pay Audio Tariff as a starting point, it being the closest available proxy, and do so. We adjust the proxy to account for the substantial increase in competition during the period of the Proposed Pay Audio Tariffs. In addition, we consider two other methodologies—one based on Nash bargaining, and one based on the 2002 model—as put forward by the Collectives, rejecting them both.

i. Do previous tariffs represent a Nash Bargain?

[159] Dr. Boyer states that the principle of balance [77] is related to important economic concepts such as the concept of bargaining and in particular the Nash bargaining solution. He opines that it is possible to conceive of these negotiations as a game [78] between the Parties. He interprets the lack of opposition in 2005 and 2010 to the rate the Board set in 2002 as the outcomes of Nash bargains, which share surplus equally among the parties to the bargain.

[160] Dr. Marx critiques Prof. Boyer’s claims, noting that he does n ot calculate the surplus, does not implement the equal sharing rule in his report, and does not develop the Nash bargaining solution to determine any particular rate. [79]

[161] We agree with Dr. Marx; further, we are not able to determine whether the bargain is classical Nash. In addition, the reference to a Nash bargain without explicit modelling of bargaining power and threat points (the position that each party receives if no bargain is struck) is not meaningful. [80] While it is possible to interpret the withdrawals of objections in the previous two pay audio proceedings as a form of settlement, a settlement during litigation is unlikely to reflect a symmetry of bargaining power and threat points.

[162] Further elements that would inform bargaining power and threat poin ts include:

• The level of exclusivity of the rights administered by the Collectives;
• Whether the act to be authorized is prospective, or has already been done;
• Whether the results from this negotiation affect other negotiations and other relationships between the underlying copyright holders and the users (e.g., how else do they interact?); and
• At what point should/could the Collectives walk away (e.g., zero-marginal-cost goods versus frictions)?

[163] Moreover, we do not accept the assumption of the existence of a bargain between the relevant parties. The parties to the putative bargains in 2005 and in 2010 would have included the Canadian Broadcasting Corporation (CBC), which operated pay audio at the time through its subsidiary Galaxie—and would not have included Stingray, since Stingray only started operating pay audio in 2011. It is very improbable that a classical Nash bargain with CBC, would also amount to a classical Nash bargain with Stingray —their cost structures being very different. And so, even if there had been a Nash bargain previously, it is very improbable that it would remain so when the parties involved change.

[164] Accordingly, we reject the use of the Nash bargaining model in this proceedin g for two reasons, those related to bargaining power and threat points, and those related to the parties to the bargain.

ii. Can the Board apply the same methodology that it did in 2002?

[165] Dr. Boyer testified that there is another ground —one not developed in his expert Report— for maintaining the last-approved royalty rates for pay audio: he claimed that the model used by the Board in 2002 (though never spelled out), if applied again today, would result in the same (or higher) royalty rate.

[166] Dr. Marx points out the deficiencies in this argument. Prof. Boyer does not present any economic analysis that the 1997-2002 Pay Audio Tariff [81] remains a reliable benchmark in today’s marketplace. [82] In her view, the existing pay audio royalty rate is not grounded in economic principles, nor was any step used to arrive at that rate. [83]

[167] We note that, in effect, Dr. Boyer provided opinion on how to interpret the Board’s reasons in 1997–2002 Pay Audio Tariff. This evidence is essentially an attempt by Dr. Boyer to reconstruct the Board’s reasoning. The interpretation of domestic law, such as Board decisions, is squarely within the Board’s competence.

[168] Even were we to give this aspect of Dr. Boyer’s testimony some weight, we disagree that his interpretation is a justification for setting the same percentage rate today as the Board did in 2002.

[169] Dr. Boyer assumes that the model of 2002 remains valid today. However, Dr. Boyer did not test the model(s) of 2002 for continued validity: he did not do any econometric estimation or calibration or give other evidence to show that those models remain valid.

[170] The Board has previously explained the importance of validity [84] of models it considers. We agree this continues to be important when considering a particular model. Without additional evidence, we cannot conclude that the modelling used in 2002 remains valid today, and do not use it.

iii. Should we use the 2007–2009 Pay Audio Tariff as a proxy?

[171] In this proceeding, the closest proxy we are able to identify is the last-approved tariff for pay audio, 2007–2009 Pay Audio Tariff. [85] We use it as a starting point, and make adjustments thereto, to account for increased competition.

[172] The evidence before us demonstrates that, by and large, the marketplace in which pay audio transacted in 2010 was very comparable to that in the period of the last-issued tariff for pay audio, 2007–2009. This makes the 2007–2009 Pay Audio Tariff a good candidate proxy to consider.

[173] The royalty rate set for pay audio in that tariff is the same as was previously set in the 1997-2002 Pay Audio Tariff and the 2003-2006 Pay Audio Tariff. [86] We are aware of the critique that the initial decision that set this rate may not have been economically well founded.

[174] However, we accept a limited version of Dr. Boyer’s claim: that some parties bargained and agreed on the rates which were ultimately approved by the Board in its 2005 and 2010 decisions. We have no evidence to accept that these parties reached a classical Nash bargain, that is, a bargain in which surplus was shared on an equal basis. However, we have no reason to expect that any of the parties to those bargains did so inefficiently. An efficient bargain of this sort, while not necessarily a Nash bargain, is per se a “market price.”

[175] Thus, regardless of the potential deficiencies of the model that led to that rate initially, it is now a market rate, and is therefore a reasonable starting point for our analysis. [87] In 1997–2002 Pay Audio Tariff the Board set the (non-discounted) SOCAN rate at 12.35%. Prior to adjustments to account for eligible repertoire, this rate was 13%; [88] we term this the Proxy Rate.

[176] This approach is consistent with the Board’s objective of fixing tariffs that are “fair and equitable.” This Proxy Rate has been in effect, in substance, for some time. During the present tariff period, the evidence shows that the number of plays of sound recordings has continued to increase as has the number of BDU subscribers.

[177] |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| There is no hard and fast rule as to what is “fair and equitable,” but courts have accepted that it incorporates concepts of some stability and predictability and an element of reasonable expectations. [89] “Fair and equitable treatment” is also a principle that is applied in international treaty law (see the former North American Free Trade Agreement, for example) where the foregoing concepts are applied in that context as well. Here, to utilize the proxies put forward by the Objectors would lead to significantly higher decreases in royalties than those reflected in the gradual decreases we have implemented, something which, in the context we have just described, would not lead to a result that is “fair and equitable.”

a. Accounting for Effect of Competition

[178] However, there have been changes to the marketplace during the effective period of Proposed Pay Audio Tariffs. Most notably, pay audio has come under substantial competitive pressure.

[179] In his testimony, Mr. Vidler, a media researcher appearing as an expert witness for the Objectors, highlighted the changes to the neighbouring markets to pay audio, as many new services have entered Canada to compete with pay audio with a relatively similar offering. [90]

[180] Mr. Wright testified that the growth in competitive substitutes, such as streaming music services, has an impact on the relative valuation of Stingray’s pay audio service. In particular, the growing availability of substitutes diminishes the standalone value of Stingray’s pay audio service. [91]

[181] Stingray’s prospectus identifies several competitors, such as iHeartRadio, Pandora, iTunes Radio, LastFM, Google, Songza, YouTube, Spotify, and Slacker. [92] ^{, [93] , [94]} Some of these competitors appear routinely before us; some of them are large, multi-national corporations with substantial revenues.

[182] Economic theory suggests that the presence of such significant competitors drives down Stingray’s per-customer profits for its pay audio service. The exact amount of these profits is not in evidence.

[183] However, having adopted the allocation we do above, |||||||||||||||||||||||||||||||||||||||||||||||||||||| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| Simply stated, since approximately 2010, Stingray has been offering a greater number of services within its bundle of services, ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||

[184] While a revenue-based royalty rate should be self-adjusting, and thus resistant, to small fluctuations in the relevant market, significant changes require that the rate be re-assessed. In this case, we conclude that during the effective period of the Proposed Pay Audio Tariffs (i.e. 2010– 2016), the change in the relevant market is sufficiently significant that the proxy we use (the 2007– 2009 Pay Audio Tariff) requires adjustment.

[185] We conclude that the cumulative competitive effect of significant development of music offerings to consumers, coming from both direct and less direct competitors, has contributed to ||| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||

[186] There is evidence that at least certain costs increase with an increase in subscribers. Mr. Michelin notes that the direct costs of Stingray include webhosting and transmission costs, both of which seem to increase as do the number of subscribers. [95] In addition, the Prospectus [96] discusses how new subscribers may have different preferences than the previous subscribers; to reflect these preferences, Stingray has had to hire new staff to curate music for new channels.

[187] It may be that these increases in costs are comparatively small. Even so, ||||||||||||||||||||||||||||||||| |||||||||||||||||||||||||||||||||||| is of such a magnitude that even if costs were relatively fixed, we expect that |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||

[188] We therefore need to adjust the proxy we use for 2010 downward to reflect this.

[189] There is no model in this proceeding that would allow us to determine how large this decrease should be; neither do we have the data to establish such a model, in part because Stringray sells a bundle of services to the BDUs that includes pay audio and other services.

[190] We compare the situation of additional competition to the pay audio industry to the situations that led the Board to apply infant-industry (new-tariff) discounts the Board in the past. In such proceedings, the Board first calculated rates based on evidence unrelated to the profitability of the industry and then used evidence regarding the profitability of the industry to apply an infant-industry discount.

[191] In SOCAN – Tariff 22.A (Internet – Online Music Services) for the Years 1996 to 2006 [97] and CMRRA/SODRAC Inc. (Online Music Services) for the Years 2005 to 2007, [98] the Board applied a discount of 10%. In Tariff for Satellite Radio Services, 2005-2010, [99] the Board applied a discount starting at 25% at the start of the tariff period, and 10% later in the tariff period.

[192] In this proceeding, the evidence shows growing |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| This evidence is not surprising: when new competitors enter an industry, they also cause a decrease in profitability for the incumbent. Moreover, unlike the situations that required an infant-industry discount, this decrease in profitability is not temporary.

[193] We find that such a permanent decrease in profitability, caused by the entrance of mature, large, multi-national companies is inadequately benchmarked by the most common discount used in past cases: 10%. A discount of 20% is more appropriate. However, since |||||||||||||||||||||||||||||||||||||||||| ||||||||||||||||||||||||||||||||||||||||||||||||||||||, we linearly phase-in a discount of 20% on the starting royalty rate, beginning in 2011, for a resulting Gross Royalty Rate of 10.4% in 2016.

b. Conclusion

[194] Table 10, below, shows the adjustments we make to the proxy rate, based on the effects of competition. The resulting rate is the Gross Royalty Rate (i.e. unadjusted for repertoire use) applicable to each of SOCAN and Re:Sound for the pay audio service.

Table 10: Gross Royalty Rate

Table 10: Gross Royalty Rate

Year	Proxy Royalty Rate	Discount	Gross Royalty Rate
2010	13%	0%	13.00%
2011	13%	3%	12.57%
2012	13%	6%	12.14%
2013	13%	10%	11.70%
2014	13%	13%	11.27%
2015	13%	17%	10.84%
2016	13%	20%	10.40%

C. THE ROYALTY RATE FOR SMALL ENTITIES

[195] Subsection 70(3) of the Act requires the Board to set a preferential rate for small cable transmission systems.

[196] In its Proposed Pay Audio Tariffs, SOCAN proposed that “small entities,” namely:

1. a small cable transmission system,

2. an unscrambled Low Power Television Station or Very Low Power Television Station (as defined in Sections E and G of Part IV of the Broadcast Procedures and Rules of Industry Canada effective April 1997), or

3. a system which performs a function comparable to that of a cable transmission system, which uses Hertzian waves to transmit the signals and which otherwise meets the definition of “small transmission system,”

pay “6.175 per cent of the affiliation payments payable during a year by a distribution undertaking for the transmission for private or domestic use of a pay audio signal.”

[197] This rate amounts to one half of the rate proposed for other BDUs.

[198] In their submissions, the Objectors argued for a rate for small entities that amounted to one half of the rate for other BDUs.

[199] The last-approved tariff for pay audio, 2007–2009 Pay Audio Tariff, fixed the royalty rate for such entities at one half the rate for other BDUs.

[200] Given that the Proposed Pay Audio Tariffs imply a 50% discount, that this discount is supported by the Objectors, and that Board has previously fixed such a discount for small entities in other approved tariffs, [100] we conclude that a 50% discount for small cable transmitters remains appropriate.

[201] We therefore fix the royalty rate for pay audio applicable to small cable transmitters at 50% of the rate applicable to other BDUs.

V. ISSUE 3: WHAT SHOULD THE ROYALTY RATE FOR SIMULCASTS AND NEAR-SIMULCASTS BE?

[202] For simulcasts, we conclude that, for the purposes of the Tariff, revenues attributable to pay audio already include those attributable to simulcasts.

[203] With respect to near-simulcasts, we fix a royalty rate using Online Music Services (2010–2013) as a proxy. We do not specify the portion of affiliation payments that will constitute the Rate Base for near-simulcasts. Of course, whatever portion of affiliation payments that will constitute the Rate Base for pay audio must not also be included in the Rate Base for near-simulcasts; we say so in the Tariff. To do otherwise would be double counting.

[204] After describing simulcasts and near-simulcasts, we consider whether a theoretical model or a proxy approach can be used.

[205] We note that in Ruling CB-CDA 2018-205, the Board stated it would consider near-simulcasts in this proceeding specifically on the condition that no new evidence would be filed by the Parties. As such, expert evidence on this issue is necessarily limited.

A. WHAT ARE SIMULCASTS AND NEAR-SIMULCASTS?

[206] Simulcasts are transmissions of pay audio programming by internet simultaneously (or near-simultaneously) with the original broadcast. During the Tariff period, simulcasts were provided by means of a web-based player. Far fewer hours of simulcasts were played, when compared to pay audio proper. The Tariff characterizes a simulcast service as one which provides “simultaneous” or “nearly-simultaneous” transmission of pay audio programming. When the Board consulted the Parties on the wording of the Tariff, [101] this characterization led to some confusion. We note that “nearly-simultaneous” refers to the possibility of a few seconds lag between the transmission of the pay audio signal on television and the simulcast thereof. It does not qualify the contents of the programming, which will be the same for both.

[207] The term near-simulcasts was not used often in this proceeding. Several Objectors introduced the term “near-simulcasts” in mid-2018, seeking to ensure that the present proceeding would cover this activity—as opposed to it being considered in the ongoing Online Music Services (2007–2018) proceeding. They described “near-simulcasts” as transmissions provided through a BDU that starts as a simulcast, but becomes desynchronized as a result of end-user activity, such as pausing or skipping.

[208] Mr. Peloquin testified that the mobile app provided by Stingray for BDU customers includes semi-interactive features, offering end-users the ability to customize channels by “liking” songs and artists in order to hear these songs and artists more often, as well as block songs or artists from their music feed and skip songs they do not like (a limited number of times). [102] They are therefore a semi-interactive webcast.

[209] In-line with the way the term was used by the Parties, “ near-simulcasts” do not include— for the purposes of this proceeding—semi-interactive webcasts that Stingray provides directly to its own customers. Near-simulcasts are only those semi-interactive webcasts that Stingray provides to customers of its client-BDUs. At the request of the Parties, and as per Ruling CB-CDA 2018-205, semi-interactive webcasts that customers obtain directly from Stingray are not considered in this proceeding. This may include, for example, the Vibes channels, whose content is not based on pay audio programming. It also includes semi-interactive webcasts with content similar to pay audio programming, but to which consumers can only have access due to a direct subscription to Stingray.

[210] Confusingly, the term simulcast was used by some Parties to describe a webcast that— having not yet been manipulated by a customer—is still the same as the source pay audio program. The idea was that such a webcast would be a simulcast until manipulated by a customer, and become a near-simulcast upon manipulation.

[211] In Ruling 2018-149, the Board rejected this characterization, and held that so long as a service has the possibility of interaction, transmissions from those services are not simulcasts. Whether or not a customer interacts with the service, the mere possibility to do so makes these transmissions semi-interactive webcasts.

[212] We reaffirm that view here: whether or not an option for interactivity is actually used or not does not change the character of that particular service. Those services described as “near-simulcasts” by the Parties are, for the purposes of this Tariff, a specific form of semi-interactive webcasts.

B. ANALYSIS – CAN THE CLASSICAL NASH BARGAINING SOLUTION BE USED TO PRICE THE ANCILLARY SERVICES?

[213] After considering the Nash model put forward by Dr. Boyer, we conclude that we cannot use it in this proceeding to establish a royalty rate for the ancillary services.

[214] Dr. Boyer gave the opinion that, according to the classical theory of Nash bargaining, the price for information goods should be based on the surplus they generate, sharing it equally between users and rights holders.

[215] However, Dr. Boyer recognized that the Board does not have evidence to evaluate the surplus to be shared. Notably, Dr. Marx generally agrees with Dr. Boyer that classical Nash bargaining can be a methodology to find a price, but disagrees that all bargains automatically meet the criteria of a Nash bargain, including the requirement that all parties have equal bargaining power.

[216] There are two additional reasons why we do not use the classical Nash model here.

[217] First, even if the procedural history of past pay audio tariff proceedings would support a finding of implicit “bargaining,” any bargain would not have been related to simulcasts, nor near-simulcasts. Past pay audio tariffs have not set such rates.

[218] Second, according to data provided during the interrogatories, simulcasts began around 2011 and semi-interactive webcasts around 2013. Bargains struck prior to this time would have been unlikely to consider simulcasts and semi-interactive webcasts—or only very speculatively.

C. ANALYSIS – CAN A PROXY BE USED TO PRICE THE ANCILLARY SERVICES?

[219] Having concluded that we cannot use a theoretical model to determine royalty rates for ancillary services, we consider the use of a proxy.

[220] With respect to simulcasts, Dr. Marx states that technological neutrality implies that the price paid for the simulcast transmission of a service should be the same as the price paid for a listen on the original service, which—in her view—would itself be based on the rate in the Online Music Services (2010–2013) tariff. [103] In response to Dr. Marx’s expert opinion, Dr. Boyer advanced several reasons for rejecting non- and semi-interactive webcasting as a useful proxy for pay audio. [104]

[221] Even to the extent these reasons are valid in respect of pay audio proper, they are applicable neither to simulcasts nor to near-simulcasts.

i. Simulcasts

[222] For simulcasts, we consider two potential proxies: using the royalty rate that we set in this proceeding for pay audio, or using a royalty rate the Board has previously fixed for other webcasts in Online Music Services (2010–2013).

[223] We conclude that it is better to use the pay audio rate in this proceeding for two reasons: the pay audio proxy is a “closer” proxy, and —given difficulties of revenue allocation—it is simpler to do so. We further conclude that, for the purposes of the Tariff, revenues that we allocate to pay audio also include revenues attributable to simulcasts of pay audio programming.

a. The Closeness of the Two Potential Proxies

[224] Both proxies share many similarities with simulcasts; they are both meaningful proxies.

[225] However, pay audio differs from simulcasts primarily by means of transmission (cable or satellite, rather than internet) and consumption (by television, rather than by a mobile device or other computer). The first distinction is almost without a difference: the technologies used for such transmissions became more and more alike during the effective period of this tariff. The second is more meaningful: the way content is consumed can have a bearing on the price of that content. Thus, this consideration favours the use of previous webcasting tariffs as a proxy.

[226] Second, we consider the business models implied by the two proxies. The pay audio proxy is from a market where the service is provided uniquely by way of subscription. Rates that have been fixed for webcasting in Online Music Services (2010–2013) are from a market where the service is provided both by way of subscription as well as for free —but advertisement-supported. Indeed, among the largest webcasters, a significant amount of revenue continues to be from the sale of advertisements. The less interactive the webcasting service, the less likely that derived revenues will be from subscriptions. This consideration favours the use of the pay audio service as a proxy.

b. Practicability

[227] We also consider the practicability of using a rate other than that for pay audio for simulcasts.

[228] As we conclude above at paragraph [79], royalties fixed by a tariff should only be based on revenues attributable to the activity covered by the tariff. Here we need to identify which revenues are attributable to simulcasts.

[229] Given the evidence, we conclude that it is not possible to identify revenues attributable to pay audio separately from those attributable to simulcasts.

[230] In 2015, Stingray identified seven revenue categories. [105] Of these, there appear to us to be two reasonable categories where simulcasts may belong: Music and Mobile.

[231] Given the content of the various revenue categories, we conclude, for the purposes of fixing a royalty rate for simulcasts in the Tariff, that the content of these categories did not change significantly during the period of Proposed Tariffs (e.g., a service that was in one category later is classified in another category). As such, simulcasts cannot be in the Mobile category for the reason that simulcasts existed before the Stingray Mobile service even existed. Therefore, we conclude that revenues from simulcasts are already in the Music category.

[232] While it is theoretically possible to attempt to separate from the Music allocation category those revenues attributable to the web player, and then apply a different royalty rate, there are insufficient data to do this with any certainty.

[233] Given the lack of evidence, the comparatively small scope of simulcast activity, and given the difficulty of separating revenues attributable to simulcasts from those attributable to the pay audio service, we conclude that the royalties fixed in respect of this service already include royalties for these simulcasts.

[234] The effect of this conclusion is twofold. One, for the years of this tariff for which pay audio is under consideration (2010–2016), the royalties paid in respect of pay audio will also cover simulcast. Two, for the years of this tariff for which simulcasts are under consideration—but pay audio itself is not (2007–2009)—no additional royalties will be owed in respect of simulcasts. The royalties previously paid for pay audio are deemed to have been in respect of simulcasts as well. Thus, the royalty rate specified by this tariff for simulcasts for those years is nil.

[235] This situation is comparable to that in SOCAN, Re:Sound, CSI, Connect/SOPROQ, Artisti – Tariff for Commercial Radio, 2011-2017, [106] where a Collective sought additional royalties for uses not specifically identified in previously approved tariffs for commercial radio. The Board concluded that the manner royalties were determined under those previously approved tariffs already accounted for revenues derived from such activities. [107] We reach the same conclusion with respect to simulcasts for the years 2007–2009.

c. Changes to Web Player in December 2016

[236] According to Mr. Peloquin, as of December 1, 2016, a new version of the Stingray Music web player is made available to BDU subscribers through the Stingray Music website (http://music.stingray.com/). The service is only available to BDU subscribers who are authenticated by the BDU. The description provided by Mr. Peloquin leads us to con clude that this new version of the web player is a semi-interactive webcasting service.

[237] Nevertheless, given the difficulties in attempting to remove revenues attributable to simulcasts from the Music category, and given that this affects only one month, we make no changes to the royalty rate for pay audio.

ii. Near-Simulcasts

[238] For near-simulcasts, we consider again two potential proxies: using the royalty rate that we set for pay audio in this proceeding, or using the rate the Board has previously fixed for ot her webcasts in Online Music Services (2010–2013). We conclude that it is better to use the latter.

a. The Closeness of the Two Potential Proxies

[239] Near-simulcasts audio are semi-interactive webcasts, whose content is primarily based on programming offered through the pay audio service.

[240] Both potential proxies share many similarities with near-simulcasts; they are both meaningful proxies.

[241] Just as we did for simulcasts, we observe that the mode of transmission of pay audio differs from near-simulcasts, but that this is not a significant difference. However, the difference in mode of consumption is meaningful. While pay audio is primarily consumed through television, near-simulcasts are consumed primarily through mobile devices, such as a phone capable of running the necessary app. Perhaps most importantly, pay audio is a non-interactive service, while near-simulcasts are semi-interactive: they differ a fair bit in the manner they are consumed. Again, we observe that the way content is consumed can have a bearing on the price of the service.

[242] These considerations favour the use of the rate fixed in the Online Music Services (2010– 2013) tariff as a proxy.

[243] Second, we consider the business models used in the two proxies. The pay audio proxy is from a market where the service is provided uniquely by way of subscription. Rates that have been fixed for semi-interactive webcasts in previous tariffs, such as in Online Music Service (2010– 2013) tariff, are from a market where the service is provided both by way of subscription as well as for free—but advertisement-supported. As we noted above, the more interactive the service, the more likely that revenues would be generated from subscriptions. In this respect, near-simulcasts appear more similar to pay audio than webcasts covered by the Online Music Services (2010– 2013) tariff.

[244] However, unlike pay audio, an identical service can be purchased outside a bundle: Stingray offers an identical service directly to customers. Thus, near-simulcasts operate under a business model that is different from that of pay audio.

[245] In the end, the consideration of the business models favours neither the use of the rate we fix for pay audio as a proxy nor the rate in the Online Music Services (2010–2013) tariff as a proxy.

[246] Unlike for simulcasts, there appears to be no issue of revenues from near-simulcasts comingling with those of pay audio under the Music category.

[247] Overall, we conclude that the better proxy is the rate fixed in the Online Music Services (2010–2013) tariff. Moreover, it provides some degree of rate harmonization: at least in respect of certain years, the provision of a semi-interactive webcasting service will be subject to the same royalty rate, regardless of the identity of the provider.

b. The Rate

[248] The most recent tariff of general application applicable to semi-interactive services is the Online Music Services (2010–2013) tariff. This tariff fixed a royalty rate of 5.3% of gross revenues attributable to a semi-interactive service.

[249] Given the closeness of the proxy to the activity under consideration here, we conclude that no adjustment to the proxy is necessary—save an adjustment to account for repertoire-use—and use it for each of SOCAN’s and Re:Sound’s royalty rate for near-simulcasts.

c. Allocation

[250] We do not fix a figure or a formula for the allocation of revenues attributable to near-simulcasts. The Ancillary Proposed Tariffs (see para [21]) did not propose any such figure or formula, the Parties did not propose one, and the issue of revenue allocation for ancillary activities was not the subject of such debate as that for pay audio.

[251] Furthermore, it would be very challenging to do so given the evidence in this proceeding. Near-simulcasts do not include semi-interactive webcasts received by customers of Stingray— only those received from BDUs. Nor do near-simulcasts include semi-interactive webcasts whose content is not based on a pay audio service programming. However, the revenues derived from all of these webcasts appear to be grouped together in the Mobile revenue allocation category.

[252] There is no evidence that directly addresses into which of these allocation categories, or in what proportion, the ancillary activities under consideration are to be assigned. Unfortunately, the evidence is therefore ambiguous as to

• whether web players (whether provided by BDUs or by Stingray) fall within the Music allocation category or the Mobile allocation category;
• what proportion of Stingray Mobile is attributable to “stand-alone mobile services provided by Stingray directly to customers” (which are excluded from this proceeding under Ruling 2018-205), and which are attributable to those provided through customers’ BDUs (which are included); and
• the extent to which the Ubiquicast server (used to bring pay audio content “closer” in the chain to the customer, and reduce network costs between Stingray and the BDUs) is part of the ancillary activities.

[253] Drawing on the wording in the Online Music Services (2010–2013) tariff, we simply specify in the tariff that the royalty rate will be applicable to the portion of the affiliation payments attributable to those services. We impose no restrictions on this attribution, other than that the revenues already allocated to pay audio are to be excluded.

VI. ISSUE 4: THE REPERTOIRE-USE ADJUSTMENT

[254] Collective Societies are not automatically entitled to collect royalties with respect to all works or sound recordings used by a user: they must be both in the repertoire of the Collective Society, and eligible for royalties. [108] For this reason, the product of the Rate Base x Gross Royalty Rate is multiplied by the Repertoire-Use Adjustment: the proportion of works or sound recordings that are eligible and in a collective’s repertoire, to the number of works or sound recordings played.

[255] For example, if 100 sound recordings are played, and 60 of those are in a collective’s repertoire and eligible for royalties, then the Repertoire-Use Adjustment will be 60%.

[256] For SOCAN, we set the Repertoire-Use Adjustment as 95.8% for all years. For Re:Sound, we set the Repertoire-Use Adjustment as 45% for 2010 through 2013. The Re:Sound Repertoire-Use Adjustment for 2015 and 2016 is 63.5%. For 2014, the year in which there is a portion that is pre-WPPT [109] and a portion which is post-WPPT, we use the weighted average, namely, 51.9%.

A. HOW ARE REPERTOIRE AND ELIGIBILITY ESTABLISHED?

[257] A work or sound recording is in the repertoire of a collective society if they have been authorized by the owner of copyright, or their agent, to collect royalties in respect of those works and subject-matter. Therefore, the first issue is to determine the proportion of works or sound recordings used for which the collectives have this authorization.

[258] Furthermore, Collective Societies are only entitled to collect royalties in respect of eligible works and sound recordings—subject-matter for which the Act accords protection, and makes eligible for royalties.

[259] For works, the eligibility requirement is relatively easy to satisfy: it suff ices that copyright apply to the work in Canada. However, sound recordings are only eligible for remuneration if the maker of the sound recording was a citizen of Canada, a Rome Convention country, or a WPPT country; or where all fixations for the sound recordings were done in such a country. [110] The second issue is therefore what proportion of works or sound recordings can we conclude to be eligible.

[260] The addition of WPPT countries only came into effect as of August 13, 2014, when the WPPT formally entered into force in Canada. Since the US was not a Rome Convention country, sound recordings made by a citizen of the US, or those fixed in the US, were not eligible for remuneration before August 13, 2014. Prior to that date, US sound recordings were not eligible for equitable remuneration—even if they were in Re:Sound’s repertoire. The third issue is therefore what effect the eligibility of US sound recordings has on the Repertoire-Use Adjustment.

B. THE GATHERING OF EVIDENCE TO SET A REPERTOIRE-USE ADJUSTMENT

[261] As part of the interrogatory process, the Objectors put several questions to the Collectives related to the issue of repertoire, to which the Collectives objected. In Ruling CB-CDA 2016-047, the Board held in abeyance its rulings on these objections, on the basis that a properly conducted repertoire-use study would be better suited to determine the use of the music repertoire of each collective society.

[262] The Parties jointly submitted a methodology to be used for the study. [111] After several questions to the Parties, the Board approved the methodology, with only minor changes. [112] Broadly speaking, the Objectors would provide information on which works and sound recordings were used during a particular period, while the Collectives would provide information on which of those works and sound recordings were in the Collectives’ repertoire, and eligible under the Act for royalties.

[263] Repertoire-use studies have been conducted previously in Board proceedings. [113] The beginning of a repertoire study is inherently a cooperative exercise. After the user has supplied a list of works used, however, the user does not typically have the opportunity to challenge the claims made by the collective; more importantly, even where there is an opportunity, there is usually little evidence on which such a challenge could be made.

[264] To address this issue, this repertoire study contains the following clause:

17. The Objectors may audit up to 10% of items in the “yes” category reported by each of Re:Sound and SOCAN. For each of Re:Sound and SOCAN, the items to be audited will be randomly selected from the entire pool of “yes” responses provided by the Collective being audited. Each Collective would then provide whatever documentation and/or other information upon which it relies in claiming that the sound recording or musical work audited is eligible repertoire. The parties are free to make submissions on how the audit results should be interpreted.

[265] The repertoire study was begun on March 1, 2017 and was completed on September 19, 2018.

[266] On September 19, 2018, the Objectors wrote to the Board, informing it that they had agreed with SOCAN on a Repertoire-Use Adjustment figure of 95.8%. In its analysis prepared prior to the audit, Re:Sound claimed that 48% of sound recordings played for the period prior to August 13, 2014 were in its repertoire. For the period from August 13, 2014 and onwards, the corresponding figure is 72%. In terms of number of sound recordings, approximately 3,000 of the 4,000 identified sound recordings were identified by Re:Sound as being in its repertoire. However, Re:Sound and the Objectors did not agree on these figures.

[267] In their September 19 letter, the Objectors complained that “Re:Sound took six months to respond to the audit and even after all that time provided only three documents. […] Despite having had two chances to do so, Re:Sound has produced no documents which specifically track any audited title into its repertoire. Accordingly, none of the audited titles has been demonstrated to be within Re:Sound’s repertoire and, as a result, all of Re:Sound’s claims to eligible repertoire remain unfounded.” [114]

[268] The September 19 letter contained multiple motions, including one that the Board “direct Re:Sound to provide such additional documentation and/or information upon which it relies in asserting that the titles in the audit sample are eligible repertoire within 60 days.”

[269] The Board granted this motion, in part, providing for a 10-day deadline, with a further 5 days for the Objectors to reply to Re:Sound’s filing. [115]

[270] Re:Sound’s response consisted of two principal points. First, it has already supplied all the necessary documentation. Second, the request for more documentation is unprecedented. In support of its argument, Re:Sound cited three Board decisions in which the Board accepted the principle that “Re:Sound is best equipped to determine what is in its repertoire and what is not.” [116]

[271] In an attempt to advance a resolution to these differences, the Board arranged for a technical meeting to be held involving the Parties and Board staff. The Board described the purpose of the meeting as follows:

The Board proposes that an informal, technical meeting take place on the issue of the information used by Re:Sound to determine eligible repertoire. The object of the meeting would be to allow the Board staff to better understand the various forms of information, and the systems used to process that information, that Re:Sound uses to determine the eligibility and repertoire status of a sound recording. An additional benefit would be to minimize the future back-and-forth between the Board and the Parties regarding the evidence underlying the study. [117]

[272] Following the meeting, Board staff prepared a summary of the meeting and shared it with the Parties for their comment. [118] Re:Sound commented on the summary and designated the entire summary as confidential. [119] As such, we do not reproduce its content here. No other party provided comments. After Re:Sound’s comments, the summary became part of the confidential record of this proceeding. [120]

[273] After receipt of the staff’s summary, and Re:Sound’s comments thereon, [121] the Board ordered Re:Sound to seek specific information relevant to establishing whether a sound recording is in Re:Sound’s repertoire and eligible for royalties in relatio n to the audited sound recordings, approximately 300 in number.

[274] Re:Sound provided updates on the process of collecting this information on June 7, July 5, and August 2, 2019. Re:Sound was scheduled to provide another update on October 7, 2019. Instead, on that date, Re:Sound announced that it had settled with the Objectors on the issue of repertoire.

[275] The settlement consisted of two figures:

• 45% for the period up to August 12, 2014; and
• 63.5% for the period from August 13, 2014 onwards.

[276] In its October 7 letter, Re:Sound wrote: “The parties have agreed that the joint proposal outlined above completes the repertoire study including the audit stage and that no further documentary evidence, submissions or formal hearing process is required.” [122] On October 31, 2019, the Board rescinded its prior order to Re:Sound and stated that “no further submissions are required.” [123]

C. ANALYSIS OF THE SETTLEMENTS IN RESPECT OF THE REPERTOIRE

[277] In this proceeding, a repertoire study was conducted, and we received settlements from all Parties. The use of a settlement to resolve repertoire issues is not wholly novel in this proceeding: the Board accepted an uncontested filing on repertoire in 2010. [124]

[278] The settlements in this proceeding have an important feature: there is a common inf ormation base that informed the settlement discussions, namely the data from the repertoire study. Given the design of the study, we expect the quality of their common information to be good. As such, we expect the settlements to reflect the reality of the repertoire of the Collectives.

[279] We therefore accept the repertoire-use figures in the settlements to determine the royalty rates in the Tariff.

i. Calculating the Pay Audio Royalty Rates

[280] For pay audio and simulcasts, this requires taking the adjusted rate (see Table 10 at para [194]) and multiplying it by each collective’s Repertoire-Use Adjustment for each year. The results of this calculation are in Table 11.

Table 11: Calculating the Pay Audio and Simulcast Royalty Rates

Table 11: Calculating the Pay Audio and Simulcast Royalty Rates

Year	Adjusted Rate	SOCAN Repertoire-Use	SOCAN Royalty Rate	Re:Sound Repertoire-Use	Re:Sound Royalty Rate
2010	13.00%	95.8%	12.45%	45.00%	5.85%
2011	12.57%	95.8%	12.04%	45.00%	5.66%
2012	12.14%	95.8%	11.63%	45.00%	5.46%
2013	11.70%	95.8%	11.21%	45.04%	5.27%
2014	11.27%	95.8%	10.80%	51.90%	5.85%
2015	10.84%	95.8%	10.38%	63.50%	6.88%
2016	10.40%	95.8%	9.96%	63.50%	6.60%

ii. Calculating the Near-Simulcast Royalty Rates

[281] For near-simulcasts, we note that the proxy we use already incorporates a Repertoire-Use Adjustment. We have no reason to conclude that the use of SOCAN repertoire is substantially different as between near-simulcasts and those webcasts in Online Music Services (2010–2013) that we used as a proxy for pricing near-simulcasts in this proceeding (see para [248]).

[282] Accordingly, we conclude that the proxy of 5.3% already accounts for SOCAN’s repertoire use. Given the agreed Repertoire-Use Adjustment of 95.8% for SOCAN, this results in a royalty rate not adjusted for repertoire-use, the Gross Royalty Rate, of 5.53% for near-simulcasts. We multiply the Gross Royalty Rate by each collective’s Repertoire-Use Adjustment to obtain SOCAN’s and Re:Sound’s Royalty Rate for near-simulcasts. These calculations are shown in Table 12.

Table 12: Calculating the Near-Simulcast Royalty Rates

Table 12: Calculating the Near-Simulcast Royalty Rates

Year	Gross Royalty Rate	SOCAN Repertoire-Use	SOCAN Royalty Rate	Re:Sound Repertoire-Use	Re:Sound Royalty Rate
2010	5.53%	95.8%	5.3%	45.00%	N/A
2011	5.53%	95.8%	5.3%	45.00%	N/A
2012	5.53%	95.8%	5.3%	45.00%	N/A
2013	5.53%	95.8%	5.3%	45.04%	2.49%
2014	5.53%	95.8%	5.3%	51.90%	2.87%
2015	5.53%	95.8%	5.3%	63.50%	3.51%
2016	5.53%	95.8%	5.3%	63.50%	3.51%

VII. TERMS AND CONDITIONS OF THE TARIFF

A. SCOPE OF TARIFF: NEAR-SIMULCASTS

[283] We approve the Ancillary Proposed Tariffs with modifications to scope so that the Tariff covers only those semi-interactive webcasts with content similar to that of Stingray’s pay audio service.

[284] The Ancillary Proposed Tariffs do not have uniform scope. Importantly, SOCAN’s proposed 22.2 and 22.B tariffs from 2009 to 2013 cover “ communications from Sites or Services whose content is similar to that of a pay audio service subject to the [Pay Audio Services] Tariff.”

[285] However, SOCAN’s proposed 22.B tariffs from 2014 to 2016 are broader, applying to “communication of audio works on the Internet by a broadcaster that is subject to […] the Pay Audio Tariff.”

[286] Tariffs proposed by Re:Sound do not have such a restriction in their wording. However, in Ruling CB-CDA 2018-205, the Board only included in the present proceeding those portions of Re:Sound’s proposed tariffs that related to simulcasts and near-simulcasts.

[287] In order to avoid a Tariff that changes in scope during its effective period, and in the spirit of the request that led to the Board’s ruling 2018-205, we limit the scope of the activities covered by the Tariff to those webcasts described as “near-simulcasts” in this proceeding.

[288] As stated at paragraph [207], above, near-simulcasts were described by some Objectors as streams that start as “simulcasts”, but whose content becomes desynchronized through pausing or skipping. While we classify such a service as a semi-interactive webcast (see para [212] above), the description is useful to understand the scope of activities that will be covered by this Tariff. We term this service as a “semi-interactive webcast service” in the Tariff.

[289] We note that such semi-interactive webcasts do not include stand-alone mobile services provided by Stingray directly to customers (Ruling CB-CDA 2018-205), nor do they include semi-interactive webcasts whose content is not based on pay-audio programming. The latter may be the case, for example, with the Vibes channels: while they may be a semi-interactive service provided to customers through distribution undertakings, they do not appear to be closely based on the pay audio channels.

B. PAYMENT OBLIGATIONS

[290] The 2007–2009 Pay Audio Tariff, as well as the Proposed Pay Audio Tariffs implicitly allowed for the possibility that either Stingray or the distribution undertaking could pay the royalties. Except as it may affect reporting obligations, which we address below, this arrangement does not appear to have caused any issues. We therefore leave this structure in place in this Tariff.

C. REPORTING OBLIGATIONS

[291] We fix playlist reporting obligations, as per Re:Sound’s request, to require reporting for all days of the year, and to require extended sound recording information to be provided, where available.

[292] We expect playlist and sound recording information to be provided by Stingray, as this is most efficient. However, this information can be provided by either the distribution undertaking or Stingray.

i. Who provides which information

[293] In the 2007–2009 Pay Audio Tariff, the information to be provided depended on whether it was accompanying a payment being made by Stingray, or a payment being made by a distribution undertaking.

[294] The Collectives raised the issue that since payments were being made only by distribution undertakings, certain information, such as a full list of distribution undertakings that received the pay audio service, was not reported.

[295] We address this issue, in part. The Tariff will provide that a payment under the Tariff must be accompanied by certain information. The nature of the information is the same whether it is provided in connection with a payment made by Stingray, or a payment made by a distribution undertaking.

[296] However, we recognize that certain information, such as a list of pay audio signals, as well as playlist information could be provided by Stingray—perhaps once, in relation to all distribution undertakings. This would be more efficient than each distribution undertaking producing its own report for such information. Indeed, Stingray has been providing music use information to the Collectives. [125]

[297] This situation is contemplated in the Tariff: it will permit Stingray to furnish any information in the stead of any distribution undertaking, or all of them.

[298] Where Stingray makes a payment under the Tariff, it will be required to provide a list of the distribution undertakings to which it provided the pay audio service.

ii. Playlist information

[299] The reporting requirements of the Tariff will include a census of recordings played, instead of only a 7-day/month sample.

[300] The 2007–2009 Pay Audio Tariff included an obligation to provide a sequential list of all recordings played on each pay audio signal during a seven-day period in a month, and to provide detailed information for each sound recording.

[301] Re:Sound asks that this be expanded from a sample to a census: that the list is of all recordings played. SOCAN supports Re:Sound’s proposal. [126]

[302] In Commercial Radio (2011–2017), [127] the Board noted that the use of software has permitted many commercial radio stations to provide 365-day reports. [128] Based on the evidence in that matter, it also expressed the view that

[a] smaller sample […] will tend to favour those rights owners whose music is played more often while those rights owners whose music is played less frequently are more likely to be missed by such samples. [129]

[303] We are of the view commercial radio is sufficiently similar to pay audio in this respect that a similar issue is very likely to exist here: the more limited the sample, the less likely that unfrequently-played music will be captured.

[304] Furthermore, Stingray, as well as many of the distribution undertakings are large and sophisticated undertakings. They have the means to implement 365-day reporting.

[305] While this may be a challenge for smaller distribution undertakings, the Tariff includes a mechanism by which Stingray can provide such information on their behalf (see para [296], above). Furthermore, this is likely to be only a theoretical problem: in the past, it has been Stingray that provided the music use information. [130]

[306] For these reasons, the Tariff will require all played sound recordings to be reported.

[307] However, as with the sound recording information itself (see para [316], below), it is possible that with respect to events in the past, this information was either not collected, or not retained. This is addressed by the Tariff requiring sound recording information for such playlists “where available.” In respect of days for which no playlist information was recorded or retained, no sound recording information will be available.

iii. Sound Recording Information

[308] We accept Re:Sound’s request to expand the kind of information provided in respect of a playlist entry. The Tariff will require all the requested information to be provided, where available.

[309] Re:Sound asks that the information that must be reported in respect of a sound recording be expanded from that which was required in the 2007–2009 Pay Audio Tariff. In particular, Re:Sound requests that the following sound-recording–related information be added:

1. the name of the sound recording;

2. the date and time of the broadcast;

3. the catalogue number of the album;

4. the track number on the album;

5. the duration of the sound recording broadcast, in minutes and seconds;

6. the duration of the sound recording as listed on the album, in minutes and seconds;

7. the Universal Product Code (UPC) of the album;

8. the International Standard Recording Code (ISRC) of the sound recording;

9. whether a track is a published sound recording; and

10. any alternative title used to designate the sound recording.

[310] Ms. Mbuoben, the Finance Manager in the Finance Department of Re:Sound, testified that the additional information

assists Re:Sound’s Distribution Department in identifying eligible and represented repertoire. While not all of this information is required to identify and determine eligibility for each sound recording, it is not possible to know in advance precisely what information will ultimately be required for each recording. Having all of the proposed categories of music use information available increases the likelihood that Re:Sound will be able to properly identify and determine eligibility of the particular sound recording and reduces the time spent by Re:Sound staff researching individual recordings. [131]

[311] SOCAN is generally supportive of Re:Sound’s proposal. [132]

[312] We accept this testimony and agree that the provision of additional information assists collectives in more precisely to whom royalties should be distributed, and the amounts of those royalties. These are both desirable outcomes. In this proceeding, the effort is not disproportional to the quanta of royalties being distributed; we therefore add the additional information to the other reporting requirements.

[313] However, users cannot provide information that they do not have.

[314] In Commercial Radio (2011–2017), the argument was made that a requirement to provide information “where available” was not sufficient, and that the reporting of all listed information be mandatory. The Board took the view that

given that such information is not always available, there is no practical way to make the reporting of such information mandatory. A radio station that is not providing information when it is readily available to it is an issue of compliance with the terms of the tariff, not of wording.

We wish to emphasize here that the standard of “where available” is not intended to be, and should not be, interpreted as conveying any level of discretion upon the person or entity having the obligation to provide the music-use information. All the listed information in its possession or under its control, regardless of the form or way in which it was obtained, must mandatorily be provided for the station to effectively be in compliance with the requirement we set. [133]

[315] We agree with the Board’s statements in that case, and use the same standard of “where available” in this Tariff as well.

[316] We note that given the retroactive nature of the Tariff, it is entirely possible that information beyond that required in the Proposed Tariffs was not collected, or not retained. In such instances, that information will not meet the “where available” criterion.

iv. Other Submissions of the Parties

[317] The Parties provided comments on a draft version of the Tariff we approve today; we thank the Parties for these. Many were of a minor nature, where slightly different wording, or a clarification was preferred, or where a small correction was required. Where there was no significant disagreement, we do not address those comments in these Reasons, and make adjustments in the Tariff as appropriate.

D. TRANSITIONAL PROVISIONS

[318] The Tariff relates to the period from 2007 to 2016; we are approving the Proposed Tariffs retroactively. As the Board has done on numerous occasions, dating back to 1999, [134] we have included transitional provisions in the Tariff.

[319] Usually, these provisions contain two elements: provisions for the payment of royalties— together with interest as set out in a table of interest factors—at a set date after the publication of the tariff, and provisions for the filing of reports relating to the periods for which the royalties are due, typically on the same date as the set date for payments. The Tariff includes these two elements.

[320] Since the royalty-rate calculation requires information (which services a BDU received from Stingray) that the Proposed Tariffs did not, it is possible that this information was not retained. We therefore include a transitional provision that takes this possibility into account.

[321] The Tariff provides that royalties owed in respect of a BDU depend on the services that BDU received from Stingray. As such, the correct calculation of the royalties can only be performed if that BDU or Stingray has retained a record of which services Stingray provided to that BDU during the relevant period. While we expect that most BDUs have these records (such information was provided during interrogatories), since the Tariff requires some calculations relating back to 2010, we need to consider the possibility that some records relating to some years may no longer exist.

[322] To address this possibility, we include a two-part rule in the Tariff:

• If records are not available for a particular year, revenue-allocation calculations are based on the following year for which records are available. For example, if records are not available for 2012 or 2013, but are available for 2014, the records for 2014 are used for both 2012 and 2013.
• It would not make sense to calculate revenue allocation by including services in the calculation that had not yet been deployed in Canada. Accordingly, we constrain the calculation not to include those services.

[323] We used the responses to Question 1 of the Board’s Notice CB-CDA 2017-098, along with other evidence filed by the Objectors about the history of the services provided by Stingray to BDUs in Canada to determine which year should be deemed as the earlier year in the Tariff for which each category of services was of fered (as such, an entry of 2010 denotes that the service was introduced in 2010 or earlier).

[324] When the Parties were given the opportunity to provide comments on the completeness, comprehensibility, and correctness of the text of a preliminary version of this Tariff, [135] the Collectives submitted that in the absence of sufficient information, all revenues—that is, the entire affiliation payment made by a BDU to Stingray—should be attributable to the Music Category. [136]

[325] The Objectors “strongly” objected to this submission, arguing that it goes beyond what the Board permitted when it provided the Parties an opportunity to comment, and is an attempt to reargue the substance of the matter. [137]

[326] It is not necessary for us to determine the appropriateness or admissibility of the Collectives’ submission on the default revenue allocation. For all the reasons we already expressed above, we are of the view that it would be unfair to include in the revenue base those revenues generated by activities not subject to the proposed tariffs under consideration. As such, some mechanism for allocating revenues is necessary—even if it is limited by the data that are available.

APPENDIX: OBJECTORS

At earlier points in the proceeding, the following companies were also part of the Objectors: Braggs Communications Inc., the Computer & Communications Industry Association, MTS, and Yahoo! Canada Co. All of these latter companies withdrew their participation prior to the oral hearing, for various reasons.

In addition, the following companies objected to one or more of the tariffs under consideration in this proceeding, but also withdrew their participation: Apple Canada Inc., Apple Inc., Canadian Association of Broadcasters, Canadian Broadcasting Corporation, Canadian Satellite Radio, Cineplex Entertainment LP, CKUA Radio Network, EMI Music Canada, Entertainment Software Association, Entertainment Software Association of Canada, Federation of Calgary Communities, Hotel Association of Canada, Iceberg media.com, L’Alliance des radios communautaires, L’Aréna des Canadiens inc., L’Association des radiodiffuseurs communautaires du Québec, Music Canada, National Campus and Community Radio Association, Pandora Media Inc., Pelmorex Media Inc., Restaurants Canada, Saskatchewan Telecommunications Inc, Sirius XM Canada Inc., SONY BMG Music (Canada) Inc., Universal Music Canada Inc., and Warner Music Canada Co.