Copyright Board Canada

Commission du droit d’auteur Canada

 

Date	2011-07-08
Citation	Files: Public Performance of Musical Works; Public Performance of Sound Recordings
Regime	Collective Administration of Performing Rights and of Communication Rights Copyright Act, subsection 68(3)
Members	Mr. Justice William J. Vancise Mr. Claude Majeau Mrs. Jacinthe Théberge
Proposed Tariffs Considered	SOCAN AND RE:SOUND TARIFFS 1.C (CBC-RADIO) 2006-2011

statement of royalties to be collected for the communication to the public by telecommunication, in Canada, of musical works and of published sound recordings embodying musical works and performers’ performances of such works

Reasons for decision

I. INTRODUCTION

[1] The issue in this matter is simple. Should the royalties paid by the Canadian Broadcasting Corporation (CBC) for the music it uses on its radio services continue to be linked to the amount of royalties paid by commercial radio stations? The answer is not so simple. The content as well as the technologies used to deliver it have evolved considerably in the last 20 years, both for CBC and commercial radio. We thus need to reexamine whether the link between the CBC and commercial radio is still appropriate today.

[2] The following are the Board’s reasons dealing with Tariffs 1.C (CBC – Radio) of the Society of Composers, Authors and Music Publishers of Canada (SOCAN) and of Re:Sound Music Licensing Company (Re:Sound) [1] for the years 2006 to 2011. Proposed statements of royalties were filed pursuant to subsection 67.1(1) of the Copyright Act. [2]

[3] On March 31, 2005, March 31, 2006, March 30, 2007, March 31, 2008, March 27, 2009 and March 31, 2010, SOCAN filed proposed statements of royalties for the communication to the public by telecommunication of musical and dramatico-musical works in 2006 to 2011.

[4] On April 2, 2002, [3] March 27, 2007, March 28, 2008 and March 31, 2009, Re:Sound filed proposed statements of royalties for the communication to the public by telecommunication of published sound recordings embodying musical works and performers’ performances of such works in 2003 to 2007, in 2008, in 2009 and in 2010 to 2011.

[5] Statements of proposed royalties were published in the Canada Gazette with a notice informing potential users and their representatives of their right to object to the statements. CBC filed a timely objection to each tariff proposal.

[6] On February 17, 2004, Re:Sound informed the Board that it had reached an agreement with CBC for the period 2003 to 2005. On October 7, 2004, CBC confirmed that if the Board agreed to certify Re:Sound Tariff 1.C for 2003 to 2005 according to the agreement reached between the parties, CBC would withdraw its objection for these years. For 2006 and 2007, since Re:Sound proposed that its tariff proposal remain as before, CBC maintained its initial objection.

[7] In December 2004, SOCAN reached an agreement with CBC for the years 2002 to 2005. On the following January 21, SOCAN informed the Board of that agreement. CBC confirmed this on January 27.

[8] Pursuant to the agreements, the Board certified Re:Sound Tariff 1.C (2003-2005) on January 14, 2005, and SOCAN Tariff 1.C (2002-2005) on March 20, 2008.

[9] The examination of the CBC tariffs for the remaining years was held in abeyance pending the decision of the Board in the re-determination of the commercial radio tariff. That decision was issued on February 22, 2008. [4]

[10] On September 25, 2009, the Board consolidated the examination of Re:Sound Tariff 1.C for 2010-2011 and SOCAN Tariff 1.C for 2010 with the examination of the same tariffs for 2006-2009. SOCAN and CBC agreed to a later request by the Board to include the SOCAN tariff for 2011 in the present hearing.

[11] Hearings were held over three days, in December, 2009 and the record of the proceedings was perfected on December 16, when CBC complied with undertakings made during the hearings.

II. THE POSITION OF THE PARTIES AND THEIR PROPOSED RATES

A. Collectives

[12] The Board last certified SOCAN Tariff 1.C on March 20, 2008. The annual fees, as agreed upon by the parties, increased from $1,380,675 for 2002 to $1,486,836 for 2005.

[13] Initially, SOCAN sought royalties of $2,750,000 for 2006 to 2009 and $3,750,000 for 2010 and 2011. After final data became available, SOCAN increased its request to $3,254,561 for 2006, $3,736,542 for 2007 and $4,027,525 for 2008 to 2010. [5]

[14] SOCAN requests that the formula used in 1991 to determine CBC radio royalties [6] be reinstated. The formula is composed of three elements: royalties commercial radio stations pay to SOCAN; CBC radio’s use of the SOCAN repertoire as compared to commercial radio; and CBC radio’s audience share as compared to commercial radio. The significant increases SOCAN proposed reflected, in part, proportional increases in music use and audience share.

[15] In support of this submission, SOCAN advanced three arguments. First, the reasons that resulted in an increase to the commercial radio rate apply with equal force to CBC. Second, the Federal Court of Appeal has found commercial broadcasters’ income to be “extremely relevant” in the determination of the CBC rate. [7] Finally, the magnitude of the proposed increase over the current tariff is irrelevant so long as the amounts proposed and ultimately certified are fair and equitable.

[16] The Board has certified a CBC radio tariff for Re:Sound twice since the inception of the tariff in 1998. The first time followed a hearing; the second decision reflected an agreement between the parties. [8] In each case, the rate was set at $80,000 per month.

[17] In its proposed tariffs, Re:Sound sought royalties of 6 per cent of operating costs [9] multiplied by 0.6139; [10] this would have resulted in royalties of $14,608,830 in 2008. In its statement of case, Re:Sound asked that the Board continue to set its royalties as a proportion of SOCAN’s, resulting in proposed amounts of at least $2,929,105 for 2006, $3,362,888 for 2007 and $3,624,773 for 2008 to 2011. [11]

[18] In support of its position, Re:Sound quoted judicial declarations according to which “[w]hen acting as a broadcaster, CBC is acting in its private aspect, its rights, obligations, powers and liabilities are the same as a private broadcaster and not those of a public body.” [12] Re:Sound contends that the 1991 formula is fair in part because it takes into account the amount of music CBC chooses to use; setting royalties as a percentage of operating expenses, as CBC proposes, does not. Should the Board favour CBC’s suggested approach, Re:Sound submits the proposed rate is too low, given that the SOCAN rate for non-commercial radio, which have considerably more modest operating budgets than CBC’s, is 1.9 per cent. Re:Sound also requests more stringent reporting obligations; CBC’s consistently inadequate reports hamper efficient and accurate distribution of royalties and impose a considerable administrative burden on Re:Sound.

B. CBC

[19] CBC submits that its SOCAN royalties should be calculated as a percentage of its radio operating expenses and proposes that this percentage be 0.44 per cent. This is approximately the average fraction of these expenses CBC radio paid to SOCAN pursuant to agreements from 1992 to 2005. This would entail a payment to SOCAN of $1,531,200 for 2006 and lower payments for the subsequent years, as expenses related to radio declined. CBC strongly opposed the use of the Board’s 1991 formula for determining royalties payable to SOCAN. Counsel for CBC summarized the position of CBC neatly in his opening argument:

[TRANSLATION] In fact, this hearing really turns on determining if the CBC should be subject for all time to the tariff formula of 1991, which links the annual royalties paid by CBC to SOCAN to the amount of royalties paid by commercial radio stations. [13]

[20] Counsel for CBC argued more forcefully in his closing argument that “[the] proposals [by SOCAN and Re:Sound] blindly adhere to the Board’s 1991 tariff formula.” [14]

[21] CBC’s opposition to the use of the 1991 methodology is based primarily on the argument that it is unfair to tie its royalty obligations to the income of a third party, which operates under a completely different mandate and business model. In addition, CBC contends that the formula is inadequate for a number of reasons and is leading to excessive royalties that bear no relationship either to the value derived by CBC for the use of music or to its financial resource and cannot provide the stability CBC seeks.

[22] In CBC’s submission, the 1991 formula also exaggerates the importance of the collectives’ repertoire by reason that it treats variations in music use on all four CBC services equally, without regard to each service’s audience share. Finally, the factors the Board used to justify the recent rate increase for commercial radio do not apply to CBC. Alternatively, if the Board uses its 1991 formula, CBC contends the calculations should be weighted according to the relative audience share of the four services.

[23] In addition, CBC submits that simulcasting of its over-the-air signals should be included in Tariff 1.C. In its most recent decision on the matter, the Board did not set a tariff for the simulcasting of CBC signals reasoning that CBC royalty payments already included the right to use SOCAN music on the Internet and declared that it would deal with the issue in the context of Tariff 1.C. [15]

[24] CBC agrees that royalty payments to Re:Sound should be the same as to SOCAN, adjusted for the use of the repertoire.

III. EVIDENCE

[25] The following summarizes the most relevant exhibits and oral testimony.

[26] All parties jointly filed a report prepared by Erin Research, a Canadian media research company, on music use by CBC radio in 2009. The report measures music broadcast between 6:00 a.m. and midnight on CBC’s flagship stations: Radio One and Radio 2 in Toronto, la Première Chaîne and Espace Musique in Montreal. Using recordings provided by CBC, Erin Research verified the accuracy of the log sheets also provided by CBC, added music played but not logged and removed music that was logged but not played. To assess the relative use of music in local programming compared to flagship stations, Erin Research sampled three additional radio stations for each of Radio One and Première Chaîne. [16]

A. Collectives

[27] The collectives provided other Erin Research reports on the use of music on CBC Radio in 2000 and 2005 and on commercial radio in 2008 [17] prepared for the purposes of earlier proceedings before the Board. The confidence intervals provided with each report confirmed a high degree of reliability of the analysis, [18] except possibly for the 2000 report; in that instance, some broadcast time was missing, leading Erin Research to first perform its analysis using the raw data and then redo it using an interpolative method. Collectively, the Erin Research reports paint a picture of the evolution of the use of music on CBC radio. Arguments about the changing nature and quantity of the use of music over time were based on these reports.

[28] Each collective filed a further Erin Research report documenting CBC’s use of its repertoire. Each performed the required music coding. How this was done was explained by Mr. Michael Lewin for SOCAN and Ms. Doris Tay for Re:Sound. First, each collective correlated CBC logs, as corrected by Erin Research, to its databases and other sources [19] to determine what it “owned”. Second, each collective classified music according to its own scheme. SOCAN coded music as in repertoire (protected and registered), public domain (referred to as “not in SOCAN repertoire” by Erin Research) or not yet registered (identified, probably protected but not reported to a collective society). Re:Sound coded music as “yes” (eligibility verified and confirmed), “no” (ineligible recording, live-to-tape and live music), “potential repertoire” (probably published sound recording with eligibility undetermined) and “N/A”(entries that did not appear to be music). Third, using the information supplied by the collectives, Erin Research calculated repertoire use separately for each radio service. For SOCAN, this was done first by using its three categories and second by ignoring music that was “not yet registered” (NYR). For Re:Sound, this was done using the first three of its categories and then by ignoring both the “potential repertoire” and “N/A” categories.

[29] Both collectives raised concerns about the reliability of CBC radio logs. Ms. Tay testified that they are not as detailed as commercial radio logs, making the process of determining both eligibility and distribution more difficult and excessively time consuming. In a witness statement filed by SOCAN, Mr. Richard Kimshooye also complained about lack of reliability and insisted on the need to report for a sufficient number of broadcast days per year to adequately reflect the diversity of music played on CBC radio in the distribution of royalties.

B. CBC

[30] CBC called as witnesses Ms. Christiane Leblanc, Director of Espace Musique and Chair, Multimedia Music Committee, French services; Mr. Mark Steinmetz, Director of Music, English Radio Services; Mr. Stan Staple, Senior Director of Research and Strategic Analysis; and Mr. Michael Mooney, Senior Director, Corporate Finance and Administration.

[31] Ms. Leblanc and Mr. Steinmetz discussed CBC’s special obligations under section 3 of the Broadcasting Act [20] and its mandate: to incorporate a wide range of programming that informs, enlightens and entertains, to be predominantly and distinctively Canadian, and to contribute actively to the flow and exchange of cultural expression. Both emphasized the non-commercial, national nature of CBC’s radio services and how this differentiates them from commercial radio. Music is used differently. Formats are not predictable, because musical choices can be made by hosts or interviewees. This is not an accident; CBC radio’s operations are intended to be an alternative to commercial radio, although Mr. Steinmetz conceded that CBC radio competes with commercial radio for listeners.

[32] Both witnesses emphasized what they consider to be a key difference between commercial broadcasters and CBC radio. Commercial radio uses automatic players to select music to be aired, whereas CBC radio hosts select the music they play. Attempts to use automatic players on both English and French services have failed, since the hosts override the selections by the computers about 80 per cent of the time.

[33] The testimony and evidence of Mr. Staple focussed on the relationship between music use and audience share on each of the four services. The audience data he provided was much more detailed than the collectives’, and covered a considerably longer period of time. He critiqued the use of unweighted averages as proposed by the collectives as meaningless. Using a variety of tables, he sought to demonstrate the existence of an inverse relationship [21] between music use and audience share on CBC radio, both at the service level and broken down for individual programs. Mr. Staple also relied on the decision of CBC to move music to its secondary services (Radio 2 and Espace Musique) and away from its primary services (Radio One and Première Chaîne) as further evidence that music has a lower value for CBC now than it did in 1990.

[34] Mr. Mooney offered evidence concerning CBC’s financial situation. He emphasized the substantial unpredictability in CBC’s revenues, since they come largely from the federal budget. He also noted that, to the extent CBC’s revenues remain constant, they are failing to keep pace with inflation. One solution to this problem, Mr. Mooney argued, is to set CBC’s royalty payments as a percentage of its expenditures. This would make payments far more predictable.

IV. HISTORY OF THE TARIFFS

[35] Since the parties are asking the Board to set the royalties on the basis of very different methodologies, it is useful to examine the history of the tariffs in order to contextualize them.

[36] From 1937 to 1946, CBC paid 8 cents per licensed radio receiving set to the Canadian Performing Right Society (CPRS). [22] From 1941 to 1946, CBC also paid 1 cent per set to BMI Canada. [23]

[37] In 1947, BMI and CBC agreed to a lump-sum tariff. The Copyright Appeal Board rejected the agreement. Thereafter and until 1967, the BMI and then PROCAN tariffs had no provision applicable to CBC.

[38] From 1947 to 1951, the Appeal Board set CAPAC’s Domestic Broadcasting Tariff at 14 cents per radio receiving set, 7 cents to be paid by private radio stations and the other 7 cents to be paid by CBC. In 1950, royalties were supplemented to account for the entrance of Newfoundland into Confederation, with CBC paying the lion’s share of the additional royalties.

[39] In 1952, the Appeal Board changed CAPAC’s tariff methodology and established a CBC radio tariff rate of 1 cent per inhabitant of Canada, plus an amount of 1.75 per cent of the gross income of CBC. [24] This tariff methodology was in effect until 1956.

[40] In 1957, the Board approved a single CAPAC tariff for CBC radio and television at 1.6 cents per capita. The formula remained the same until 1985, with the rate growing progressively to 8.159 cents. The same methodology was used in BMI’s tariffs from 1967 to 1985.

[41] In 1978, CAPAC requested that separate CBC tariffs be set at 2 per cent of program costs for radio and 1.6 per cent of commercial revenues for television. The Board rejected the proposal, stating that the adoption of an expense or revenue rate base would have no rational foundation with respect to CBC. [25] The fact that CBC was essentially conceived and carried out as a non-profit national operation and that collectives could achieve their objective of increasing their members’ income within the existing framework weighed heavily in the decision.

[42] In 1984, PROCAN requested that the CBC radio royalties be calculated as a percentage of expenditures. CBC countered that this would result in paying royalties based on expenditures that had nothing to do with music use. [26] The Board agreed that CBC should pay the same basic price for music as private broadcasters since “music is no different than any other property or commodity used in broadcasting.” [27] It also supported CBC’s opinion that audience was an important factor in any comparison between it and private radio. [28] In the end, the Board rejected the proposed formula primarily for its failure “to recognize the exceptional nature of many of the costs associated with CBC broadcasting as well as the questions of audience share and the use of public domain music […]” [29]

[43] On judicial review, the Federal Court of Appeal concluded that the Board was “required to consider the dramatic difference between the return to [PROCAN] from the [CBC] on the one hand and the return to the applicant from private broadcasters on the other.” [30] For reasons not relevant to this decision, the court remitted the matter to the Board for further consideration. In its reconsideration, the Board reaffirmed its prior decision, “repeating” that:

the tariff it fixed and the formula it selected to determine the amount of the fee were based on its assessment of the adduced evidence with regard to (1) the use by the CBC of the works of composers, authors and publishers represented by PROCAN; (2) the differences in audience size between private broadcasters and CBC; and (3) certain fundamental differences between the activities of the CBC and those of private broadcasters. [31]

[44] A further attempt to set separate CBC tariffs for television and radio for 1985 was rejected. [32]

[45] For 1986, PROCAN renewed its request to split the CBC tariff, with the radio tariff being set on a per capita formula. One of the objectors, the Société professionnelle des auteurs et des compositeurs du Québec (SPACQ), suggested 3.86 per cent of operating expenditures attributable to radio services. [33] For the first time, CBC did not oppose the introduction of a formula other than a per capita formula.

[46] Though the Board concluded, for a variety of reasons, that the per capita formula had outlived its usefulness, it still set a per capita formula determined according to CBC’s relative audience share:

The rationale for this comparison is that if, instead of assuming operating costs for its radio services without offsetting them through commercial revenues, the CBC were to capitalize on these services by selling advertising, it would calculate the price of this advertising on the basis of the number of potential listeners likely to hear the advertisement. This is a universally recognized principle applied in the field of communications and thus requires no further explanation. [34]

[47] The radio and television tariffs were split. CBC payments for its radio services doubled. An application for judicial review of the decision was dismissed. [35]

[48] For 1987, the collectives requested a 4 per cent increase in the per capita amount for CBC radio. CBC objected to the structure and amount. CBC requested a lump sum payment to “reflect its use of the societies’ repertoires and the general growth index in its Canadian context.” [36] SPACQ renewed its 1986 proposal. CBC argued that a tariff based on programming costs would be arbitrary, since such costs “are affected by a significant number of factors, most of which have nothing to do with music […]” [37]

[49] The Board acknowledged the unique nature of CBC and recognized that any analogy to commercial radio should take this into account. It also restated its obligation to determine a reasonable compensation for creators and the fundamental principle that subject to necessary qualifications, comparisons could and should be made among the various users of the societies’ repertoires. [38] The Board declined to use programming costs as the rate base. It found CBC’s proposal of a lump sum of some interest, but concluded that the amount proposed was unrealistic. As a result, the Board accepted the collectives’ proposal to increase royalties by 4 per cent. [39]

[50] The Federal Court of Appeal set aside the decision, essentially finding that the Board had dismissed CBC’s counter-proposal without properly considering its merit:

I think the board ought to have squarely addressed the question whether, as the applicant contended, it ought to be treated differently [...] because the value of that music to it is less than it is to private broadcasters, the volume of actual use of protected music in its radio programming being lower. [40]

[51] On reconsideration, the Board first examined the audience share of CBC relative to commercial radio and determined the royalties payable by CBC as a proportion of commercial radio royalties. The Board next determined that CBC used 60 per cent less protected music than commercial radio. However, it discounted the royalties computed by means of relative audience share by only 30 per cent, arguing that “audience share is and must remain the primary factor in establishing any comparison between the CBC and the private sector.” [41] Finally, it converted the lump sum to a per-capita figure.

[52] The tariff formula for 1988, 1989 and 1990 was the same as for 1987.

[53] In SOCAN 1991, [42] the Board refined the general approach used in CBC (1987 - Reconsideration) by setting the radio tariff at an amount calculated by multiplying the royalties commercial radio stations pay to SOCAN by CBC’s use of the SOCAN repertoire as compared to commercial radio and CBC’s audience share as compared to commercial radio. There were two important differences in this new approach. First, the relative repertoire use became an undiscounted factor in the formula. Second, the per-capita formula was abandoned in favour of a specified lump sum to be paid in equal monthly instalments.

[54] In 1992, the tariff was certified in accordance with an agreement between SOCAN and CBC. In doing so, the Board indicated that it was not abandoning the 1991 formula. [43]

[55] From 1993 to 2001 SOCAN did not file any CBC tariff for radio or television. As indicated earlier, the radio tariffs the Board certified for 2002 to 2005 reflected an agreement between SOCAN and CBC.

[56] The Board has certified Re:Sound tariffs for CBC twice, both times at $80,000 per month. The tariff for 1998 to 2002 was set “based on the amount of royalties that CBC pays to SOCAN, adjusted to reflect CBC’s relative use of the repertoires of these two collectives.” [44] The tariff for 2003 to 2005 was set at the same rate as a result of an agreement between Re:Sound and CBC.

V. ANALYSIS

[57] According to CBC, technology may have changed substantially but the mandate and core activity of CBC radio remains the same: “putting […] content through a Canadian lens for Canadians, about Canadians.” [45] While CBC competes with private radio stations for audiences, it also has a special public mandate including: to make all programming quintessentially Canadian; to reflect national and regional realities to national and regional audiences; and, to contribute to the flow and exchange of cultural information. Moreover, the French-language services play an important role in connecting francophone communities across Canada to one another.

[58] CBC argues that its unique mandate and nature require that it be treated differently than any other broadcaster. Commercial radio sells niche markets to advertisers; typically, private broadcasters specialize in a genre of music. Doing so is lucrative, since broadcasters can more easily sell their advertising time to marketers because they are able to deliver targeted audiences to marketers. CBC is operated differently, or, as Mr. Steinmetz put it, “[w]hat CBC does is totally in a way anti-radio, it’s the opposite of how the medium is supposed to be used.” [46] The CBC has a distinct mandate under the Broadcasting Act. “It creates a mosaic of programs for Canadians of all ages and cultural groups rather than programming for a specific format. It does not repeat music or have a readily identifiable sound with which to market itself to listeners.” [47] CBC seeks to attract as diverse an audience as possible. Its music services do not repeat programming or use rotational playlists. Repeat plays are rare, and used only to promote unknown Canadian talent.

[59] From the collectives’ perspective, however, CBC is no different than commercial radio. Counsel for SOCAN expressed this point in opening argument:

… [A]t the end of the day [CBC is] still a member of the radio broadcasting industry in Canada and [it] clearly competes with private radio for audiences. [It has] been very successful in doing so and that’s to their credit. [Its] audience share has risen considerably over time. One of the reasons [it has] been successful in this way is that they have used the Collectives’ music to broaden their audience base. In that respect [CBC is] no different at all from the private broadcasters. [48]

[60] Counsel for Re:Sound focused on some of the similarities between the CBC and commercial radio stations in his closing argument. “[T]he rights, obligations, powers and liabilities, and that would include liabilities to the Collectives, are the same as a private broadcaster and not those of a public body.” [49] Counsel for Re:Sound also explained that CBC radio is similar to commercial radio in its programming strategies. “Like commercial radio, which has embarked on branding through niche programming, CBC has also pursued branding its services. Its brand is diversity.” [50]

[61] We agree with the parties, in part, and disagree, in part. CBC does have a unique mandate. Having a unique mandate does not necessarily entitle CBC to be treated differently than commercial radio. The proposition that CBC should not be treated differently than commercial radio does not imply, however, that CBC should pay royalties using a formula based on what commercial radio pays. The judicial pronouncements the collectives quote to validate comparisons between commercial broadcasters and CBC are not helpful in this regard. To state, as in the decisions referred to in note 12, that CBC should be treated the same as a private broadcaster does not settle whether this can be best achieved by using what the latter pay to set the royalties for the former. As for CBC TV (1993), it clearly states that payments made by private broadcasters are “of course not conclusive.” [51]

A. SOCAN

[62] Since all parties agree that the SOCAN rate should be used to derive the Re:Sound rate, we first set the royalties for SOCAN.

[63] In SOCAN (1991), [52] the royalties for CBC radio were set as the product of commercial radio royalties to SOCAN, the relative audience share of CBC compared to commercial radio, and the relative use of SOCAN repertoire by CBC compared to commercial radio. Commercial radio royalties are, in turn, the product of commercial radio revenues and the commercial radio rate.

[64] In its statement of case, SOCAN updates the Board’s 1991 formula with 2008 figures. In 2008, royalties paid by commercial radio stations were $54,432,346. The audience share of CBC was 12.63 per cent while the audience share of Canadian private radio stations was 80.92 per cent. [53] The relative audience share of the CBC is 15.61 per cent (= 12.63 ÷ 0.8092). According to SOCAN, CBC uses its repertoire during 31.1 per cent of its airtime; Canadian private radio stations use SOCAN repertoire during 65.6 per cent of their airtime. The relative use of SOCAN repertoire by the CBC is thus 47.4 per cent (= 31.1 ÷ 0.656). Application of the 1991 formula yields royalties of $4,027,525 (= $54,432,346 × 0.1561 × 0.474). The formula has not been revised since 1991. Agreements between CBC and the collective societies have formed the basis for the certified tariffs.

[65] We do not take issue with the raw data SOCAN proposed to use with the 1991 formula, although they were subject to considerable discussion during the hearing. Nor do we take issue with the concept that CBC radio royalties should reflect its audience share and use of music; quite the contrary, these two figures are at the core of any equitable tariff for CBC Radio.

[66] Still we must consider the relevance of the 1991 formula to today’s realities. Commercial radio royalties are calculated at a rate and on a rate base that were selected for specific reasons. These reasons have changed over time. The 1991 formula relies on commercial radio royalties; that reliance is based on certain assumptions that must remain true for the formula to be useful. We conclude that most of the assumptions that underpin the 1991 formula have become either questionable or simply wrong.

[67] The 1991 formula implicitly relies on the rate the Board set for commercial radio. In 1991, that rate was 3.2 per cent. Starting in 2003, that rate increased to 4.2 per cent, for three reasons. [54] First, music is worth more for commercial radio than the Board previously thought. Second, commercial radio now uses more music than in the past. Third, commercial radio now uses music more efficiently. Using the increased rate in calculating CBC radio royalties assumes that all these factors are as relevant for CBC radio as for private broadcasters. This is not so.

[68] The applicability to CBC of the first factor, music’s increased value, is impossible to assess without reliable evidence. We do not have such evidence. Moreover, the fact that the value of music is higher to commercial radio does not imply that the value of music is higher to CBC radio. The value of music is not an abstraction that can be measured without regard to time and place. Context is essential.

[69] The second factor is relevant to CBC radio. The royalties it pays should fluctuate with its use of the protected repertoire just as they do for commercial radio. The 1991 formula is one way to account for this factor; it is however not the only one.

[70] With respect to the third factor, the efficiencies that putatively accrued to commercial radio did not accrue to CBC. Commercial radio uses branding and niche marketing to deliver specific listeners to advertisers. This, in turn, allows radio stations to charge higher advertising rates. Two comments are relevant here. First, because CBC radio does not sell advertising, it could not possibly obtain efficiencies associated with branding and niche marketing. Second, even if CBC could possibly obtain these efficiencies, CBC has actually moved in the opposite direction. CBC’s music stations have become more diverse over the past several years.

[71] To summarize, using the 1991 formula to set CBC radio royalties today would imply applying the three factors of Commercial Radio (2005). The first is not plainly relevant. The second is relevant but can be accounted for through other means. The third is not relevant. Consequently, the rate the Board set for commercial radio in 2005 cannot be used.

[72] Another assumption of the 1991 formula is that changes in the commercial stations’ revenues, whatever their cause, should be reflected in the royalties payable by CBC radio. Put another way, the 1991 formula posits that CBC radio royalties can reliably be derived by looking at how much private radio succeeds in getting paid to deliver audiences to advertisers, whether or not music played any part in that success. We disagree. It would be inappropriate to reflect in CBC royalties any increase in radio advertising revenues attributable to factors other than music. Yet clearly, such factors do exist. A significant increase in the number of stations is one such factor. Increased concentration in the Canadian commercial radio industry has led to increased advertising revenues through sales of national advertising and through offering a portfolio of stations that allow publicity to be better targeted to a variety of desirable audiences.

[73] Furthermore, linking what CBC radio pays for music and what commercial radio earns from the sale of airtime relies on the further assumption that commercial radio listeners are all worth more or less the same. This is not correct, if it ever was. Niche programming allows radio stations not only to increase the price per listener, but also to price-discriminate according to the targeted listener’s ability and propensity to spend, among other factors. [55] The implications of such price-discrimination are stark. The value of music varies by listener, making it difficult to determine the appropriate comparator.

[74] The situation which led the Board to conclude in 1991 that the amount of royalties commercial radio pays to SOCAN could be used as a starting point to set CBC radio royalties has evolved considerably, making the use of the 1991 formula inappropriate. The disconnect between commercial radio and CBC radio is simply too great today.

B. CBC’s Proposal

[75] CBC requests that its radio tariff be set at 0.44 per cent of its radio operating expenses. CBC notes that the amount the Board certified for the years 2002 to 2005 is about 0.44 per cent of CBC’s radio operating expenses at the time and argues that this may serve as useful precedent. We disagree.

[76] There are several conceptual difficulties with CBC’s proposed approach. First, it bears no relation to audience share or music use. CBC could increase or decrease one or the other at will and still pay the same amount to SOCAN. Second, just as commercial radio royalties are unrelated to the business of CBC, CBC expenses are unrelated to the value of music.

[77] Third, much of CBC’s argument for its proposed formula and rate is a thinly veiled plea related to ability to pay. Ability to pay is relevant, but only once an otherwise fair rate and formula have been determined. We cannot say it better than counsel for SOCAN did when she stated, “[…] what this case is really about, of course the ability to play, not the ability to pay.” [56]

[78] Fourth, these amounts were agreed upon by SOCAN and CBC. When it certified the tariff, the Board did not comment on the agreement, or how it was arrived at. Many factors other than value influence prices that become part of agreements. Some of these factors include: the bargaining skills of the parties, the time and trouble of not coming to an agreement, and the expectation that coming to an easy agreement on one issue may make it easier to come to an agreement on a second, possibly unrelated, issue between the same parties. Each agreement must be analysed to determine whether it is useful in helping us fix royalties; these are not, if only because we do not know whether factors we consider important, including audience share and music use, were taken into account.

C. The Tariff

[79] The tariff formulae proposed by each of the parties are unacceptable. As a result, we must use our own approach which requires selecting a starting point and making several further adjustments. The methodology we use only contains parameters, other than inflation, that are specific to CBC radio. Thus, only changes in its behaviour will influence the amount of royalties it pays.

i. Starting Point

[80] In 1991, the Board certified a SOCAN tariff of $1,117,323 for CBC radio. This decision reflected a variety of considerations, including earlier certified tariffs, admonitions from the Federal Court of Appeal, audience share and music use. The significant changes mentioned earlier make it inappropriate, for the reasons already stated, to use the 1991 formula today. That being said, we see no reason to conclude that the application of the formula in 1991 yielded an unfair result. As such, the amount can be used as the starting point for our analysis.

[81] This amount must however be adjusted to account for three principal factors: inflation, audience and music use. Since data is available for music use and audience for 2008, we use 2008 as the base year for comparison with 1991. Royalties are calculated for 2008 and then interpolated or extrapolated to the other years of the tariff.

ii. Accounting for Inflation

[82] The witness statement of Mr. Mooney [57] contained several references to inflation and constant dollars. First, he noted that parliamentary appropriations to CBC have fallen significantly in constant dollars. Second, expenses for CBC radio have also fallen in constant dollars. In tables submitted with Mr. Mooney’s witness statement, [58] CBC computed constant dollar measures of expenditures and revenues using the Consumer Price Index (CPI). CBC proposes a tariff based on a percentage of expenditures. Tariffs of this type automatically adjust for inflation. By making inflation a part of its case, CBC effectively asked two questions. First, should the CBC radio tariffs grow with inflation? Second, what measure of inflation should be used?

[83] The answer to the first question is yes. The purchasing power of a dollar in 1991 was substantially higher than it is today. To the extent that the royalty payments in 1991 reflected the value of music, failing to adjust those payments for inflation would allow inflation to erode the value of music.

[84] The answer to the second question is more complicated. Between January 1991 and December 2008, total inflation was 38.17 per cent, as measured by the percentage change between the CPI [59] for the two months in question. In order to calculate the percentage change, it is more accurate to use the monthly figures than the annual averages. [60] Using annual averages would in effect ignore two time periods – the period in the first year where inflation has not yet exceeded its average for that year and the period in the second year where inflation has already exceeded its average for that year. The 1991 starting point will be adjusted accordingly.

[85] Over the years, the Board has adjusted tariffs fixed in dollar terms for inflation by using the Industrial Product Price Index (IPPI), the CPI and various transformations of these indexes. The last time the Board addressed the issue was in 2004. [61] The Board then adjusted fixed amount royalties based on the average annual variation in the CPI less one per cent. Two reasons justified the subtraction:

First, the tariff inflation adjustment rule must ensure some balance between music users and copyright owners represented by SOCAN. Since this rule allows some tariff increases without any other change in market conditions, it seems fair to ensure that the gains are not necessarily attributed in full to the owners. In the economy as a whole, an individual’s compensation seldom increases automatically with inflation. When there is an automatic adjustment, it is usually a partial one.

Second, although SOCAN’s tariffs account for only a very small share of the overall CPI, the tariff adjustment rule should not itself be a source of increase or maintenance of inflation in a particular sector of the economy. If, for example, the inflation adjustment brings about a general increase in fees in a particular year, this increase could be entirely passed on to the costs of halls or ticket prices, which would soon be translated into a general increase in prices in that economic sector. The risks of such a situation occurring can be reduced by subtracting a percentage from the CPI. [62]

[86] Both considerations have become less relevant, for a number of reasons.

[87] First, the possible disincentive effects that full indexation may have on collectives are largely counterbalanced by the fact that real value has been eroded by inflation, however slowly. Failing to take into account the decreased purchasing power that comes with inflation leads to certifying tariffs whose fairness and equity themselves erodes over time.

[88] Second, the analogy between compensation in the labour market and payments made through a collective to individual rights holders is arguably a poor one. While an individual whose compensation in the labour market fails to keep pace with inflation is free to seek alternative employment, a rights holder whose collective is unable to preserve the real purchasing power of his royalties has no such alternative.

[89] Third, the fear of an inflationary spiral in Canada is lower today than it was in 2004. The inflation targeting regime by the Bank of Canada has weathered several expansions and one recent contraction in the Canadian economy, without evidence of being strained. To the extent that the cultural sector might want to pass on price increases to its customers, it is restrained by the law of demand: as the price of a good rises, customers demand fewer units of that good. The desire to maximize profits makes the likelihood of an inflationary spiral essentially non-existent.

[90] Fourth, subtracting any amount from measured inflation is inconsistent with the principle of replicating competitive markets: in competitive markets, prices are set in relative or real terms, and all nominal prices adjust in lockstep.

[91] In summary, neither reason advanced by the Board in 2004 for using less than the CPI to account for inflation has much currency, and several reasons favour using the CPI without correction. The CPI remains the best measure of inflation to preserve purchasing power, since it reflects the prices of the basket of goods and services purchased by Canadians on average. We will use the CPI from January 1991 to December 2008, without further adjustment.

iii. Accounting for Variations in Audience and Music Use

[92] The CBC Radio Tariff should reflect variations in listenership and music use. The 1991 formula achieved this by comparing these variables for CBC and for Canadian commercial radio. For the reasons already stated, we no longer wish to use this comparison.

[93] We have already stated that the 1991 formula, while no longer helpful, yielded a result that is prima facie fair. If the overall amount of royalties is fair, so is the amount of royalties per unit of music. Furthermore, since we use the 1991 royalties as the starting point, then all things being equal, CBC should pay the same amount per unit in 2008 as in 1991. Therefore, if we can compare the number of units of music for CBC radio in 1991 and in 2008, we can use these figures and the 1991 royalties to calculate the overall amount of royalties that CBC radio should pay in 2008, before adjusting for inflation.

[94] The number of units of music for CBC radio is a function of protected music use and audience. [63] We know to the minute how much protected music each service uses. We also know the number of hours tuned to each service. The product of both, which we term “music impressions”, gives us the precise amount of time Canadians listened to protected music on each service. The sum of the music impressions for each service, in turn, yields the total number of impressions used by CBC radio. The totals for 1991 and 2008 can be compared, and the amount of royalties adjusted accordingly.

[95] There is a strong analogy to be drawn between music impressions and distant signal viewing impressions, as that term was used in the second retransmission decision. [64] In that decision, the Board was counting the number of minutes viewers watched distant signals. Here, we are counting the number of hours listeners hear protected music on CBC radio. There is, however, one important difference. In the retransmission decision, viewing impressions were converted to shares, since impressions were used to allocate royalties. Here, there is no need to do so. It is sufficient to compare the number of music impressions in 1991 and 2008 to derive the amount of royalties for 2008 using the 1991 royalties.

[96] The approach we use offers three significant advantages. First, it avoids having to decide whether each service should weigh equally in the calculations (as SOCAN proposed), or whether its importance should reflect its listenership (as CBC proposed). Second, it reflects variations in the amount of music listened to on CBC radio, whatever the cause might be. If CBC programmers choose to include more music, the number of music impressions rises, holding the audience constant. Conversely, if the number of hours tuned increases (because of increased audience share, increased population or increased average number of hours tuned), the number of music impressions rises. Furthermore, the concept of music impressions is ideally suited for taking into account any inverse relationship between music and audience at the service level, as argued by CBC witnesses.

[97] The first step in calculating music impressions is to obtain an estimate of the number of minutes of SOCAN music used.

[98] Column D of Table 1 shows NYR music as a percentage of airtime, pro-rated on the conservative assumption that the use of SOCAN music in the registered repertoire is no more frequent than in the NYR repertoire. Column E shows the final figure of SOCAN music in minutes that includes a pro-rated adjustment for the NYR music. [65]

Table 1: Calculation of SOCAN Music with NYR Adjustment, 2008

Service	Registered Music   Musique déclarée	NYR Music   Musique NED	Registered Music over Total Programming Time   Musique déclarée par rapport au temps d’antenne total	NYR Music Adjusted to Percentage of Registered Music   Musique NED rajustée au pourcentage de la musique déclarée	SOCAN Music with NYR Adjustment   Musique de la SOCAN avec rajustement NED
	(A)	(B)	(C)	(D) = (B) × (C)	(E) = (A) + (D)
Table 1: Calculation of SOCAN Music with NYR Adjustment, 2008 Tableau 1: Calcul de la musique SOCAN avec ajustement NED, 2008
Radio One	968.3	93.2	100.00%	93.2	1061.5
Radio 2	3367.7	815.5	69.10%	563.5	3931.2
Première Chaîne	997.7	146.7	96.60%	141.7	1139.4
Espace Musique	4057.1	685.9	74.40%	510.3	4567.4

[99] The calculation of music impressions, detailed in Table 2, shows that even though both the total hours tuned to CBC and the total minutes of music played have increased substantially, the number of music impressions has only increased modestly from 108 million to 113 million. This only serves to confirm CBC’s argument that an inverse relationship exists between music use and listenership. Whether that relationship is linked to the service or to the program is irrelevant, since all music impressions on all services receive equal treatment.

Table 2: Calculation of 2008 Royalties based on Music Impressions

Service	Music Use (Minutes)1   Utilisation de la musique (minutes)2		Hours Tuned (’000s)   Heures d’écoute (milliers)		Music Impressions   Impressions musicales		Total Royalties   Redevances totales
	(A)		(B)		(C) = (A) × (B)		(D)	(E) = (D) × (C)2008 ÷ (C)1991
Table 2: Calculation of 2008 Royalties based on Music Impressions Tableau 2: Calcul des redevances pour 2008 à partir des impressions musicales
	1991	2008	1991	2008	1991	2008
Radio One	1,920.24	1,061.50	33,173	39,505	63,700,122	41,934,558
Radio 2	2,903.04	3,931.21	9,668	9,165	28,066,591	36,029,544
Première Chaîne	2,094.12	1,139.41	5,974	14,059	12,510,273	16,018,996
Espace Musique	1,837.08	4,567.41	2,149	4,262	3,947,885	19,466,300
Total	8,754.48	10,699.53	50,964	66,991	108,224,870	113,449,398	$1,117,323	$1,171,261

¹ The data for music use in minutes for the year 1991 was not filed in these proceedings. They are derived from Exhibit CBC/SRC-2 as filed in the 1991 proceedings, where we can find that Radio One, Radio 2, Première Chaîne and Espace Musique used music respectively 25.4%, 38.4%, 27.7% and 24.3% of total airtime. Multiplying these proportions by the 7560 minutes in a week consisting of 18-hour days, we obtain the minutes of music use. Even though this exhibit was not filed in these proceedings, we see no problem in using it, since all it contains is the data SOCAN used in the instant proceedings to calculate the total amount of music used by CBC in 1991.

² Les données relatives à l’utilisation de musique en minutes pour l’année 1991 ne sont pas au dossier de la présente affaire. On les dérive à partir de la pièce CBC/SRC-2 déposée dans l’affaire de 1991, dans laquelle on peut trouver que Radio One, Radio 2, Première Chaîne et Espace Musique utilisent la musique dans des proportions respectives de 25,4%, 38,4%, 27,7% et 24,3% du temps de diffusion. En multipliant ces proportions par les 7560 minutes comprises dans une semaine constituée de journées de 18 heures, nous obtenons les minutes d’utilisation de musique. Même si la pièce n’est pas au dossier de la présente affaire, nous n’avons pas de réserve à l’utiliser : elle contient uniquement des données dont la SOCAN s’est servie dans la présente affaire pour calculer le montant total de musique utilisée par la SRC en 1991.

[100] Total royalties in 2008 are $1,171,261. As discussed above, this figure requires an adjustment for inflation. Between January 1991 and December 2008, the cumulative rate of inflation was 38.17 per cent. Adjusting the royalties for inflation yields royalties of $1,618,322.

[101] The logical way to extend the tariff calculated for 2008 to the years 2006, 2007 and 2009 through 2011 is through the use of inflation and deflation. Table 3 shows the relevant calculations. Since we do not have inflation data for 2011, we use the Bank of Canada’s inflation target of 2 per cent. [66]

Table 3: Adjusting the 2008 Royalties for Inflation and Deflation

Year   Année	Jan/Dec CPI Growth (per cent)   Croissance de l’IPC de janv. à déc. (pour cent)	SOCAN Royalties   Redevances SOCAN
Table 3: Adjusting the 2008 Royalties for Inflation and Deflation Tableau 3: Correction des redevances de 2008 en fonction de l’inflation et de la déflation
2006	1.11	$1,563,358
2007	2.38	$1,580,711
2008	1.30	$1,618,332
2009	1.59	$1,644,063
2010	2.09	$1,678,424
2011	2.00	$1,711,993

[102] In the absence of annual data for both music use and audience for 2006, 2007 and 2009 to 2011, we can only adjust for inflation. But even if we had annual data for both, it is not evident that we would make annual adjustments on this basis. The Board makes adjustments to royalties based on permanent trends, not temporary fluctuations. Discerning which of the two is at play in any given set of data is the difficult task we would have before us, if we had annual data for both variables.

VI. THE RE:SOUND RATE

[103] For the years 1998 to 2002, the Board set the royalties CBC radio paid to Re:Sound equal to what it pays to SOCAN, discounted to reflect the relative airplay of the respective repertoires. Re:Sound proposes that this formula be used once again. CBC agrees. We use this formula once again.

[104] First, we compare the relative use of each repertoire to derive the rate. Throughout, the parties used numbers expressed as percentages. This approach is invalid, since the total minutes of music on each station are not the same in the SOCAN and Re:Sound music use reports. [67]

[105] This point is perhaps best clarified by use of an example. For Radio One, Erin Research documents 1,060.5 minutes of music played relevant for SOCAN and 1,467.7 minutes of music played relevant for Re:Sound. The difference between these two is more than 400 minutes. Where the Erin Research reports contain percentages of music used on Radio One, the percentages for SOCAN would be relative to the 1,060.5 minutes, whereas the percentages for Re:Sound would be relative to the 1,467.7 minutes.

[106] Using percentages to compare CBC radio’s use of the two collectives’ repertoire is not essential. More simply, we can compare the number of observed minutes of each repertoire’s use weighted by each service’s audience share. For SOCAN, we use the number of minutes adjusted to account for SOCAN’s share of NYR titles. The issue is then whether a similar adjustment should be performed with respect to Re:Sound.

[107] Re:Sound identified three categories of music in computing its repertoire use: [68] that which clearly is in its repertoire (eligible, published sound recordings), that which clearly is not (non-eligible recordings, live-to-tape, live music) and a “potential repertoire” consisting of recordings which Re:Sound could not determine whether they were in Re:Sound’s repertoire or not. The potential repertoire can be excluded altogether, included altogether or included for part, using some reasonable mechanism for determining how much should be. At first, Re:Sound merely expressed the opinion that its repertoire is at least equal to all the repertoire definitely identified, plus some or all of the potential repertoire. Mr. Staple argued that allocating the potential repertoire proportionately would be fairer. In closing argument, Re:Sound agreed.

[108] We apply a proportional adjustment. This warrants two comments.

[109] Apparently, Re:Sound considers that all eligible published sound recordings, whether or not copyright is owned by one of its member collectives, are within its repertoire. An argument can certainly be made that Re:Sound only “owns” what is in its repertoire and that this occurs when either the performer or the maker asks one of Re:Sound’s member collectives (or any foreign collective with which Re:Sound may have an arrangement) to administer the section 19 remuneration right on his or her behalf.

[110] There is no need to deal with it in this instance, for two reasons. First, the share of eligible repertoire that CBC uses and that Re:Sound does not represent probably is very small. Second, Re:Sound determines eligibility based on declarations by AVLA, SOPROQ or their members; it is highly improbable that a music label will file eligibility declarations on account of recordings that it does not own.

[111] For the same reasons, it also appears probable that the potential repertoire category contains a smaller proportion of Re:Sound repertoire than the known repertoire category. Again, it is not possible for us to calculate the difference, which is probably small and which we decide to ignore for the purposes of this decision.

[112] To compute the Re:Sound rate, we first created a proportional adjustment for potential repertoire and then made a correction to account for the use of public domain recordings that were incorrectly reported.

[113] We then compute the musical impressions of Re:Sound repertoire on CBC. The ratio of the musical impressions of Re:Sound repertoire relative to SOCAN’s repertoire is 77.32 per cent, as shown in Table 4. The result is that CBC will pay royalties to Re:Sound that are 77.32 per cent of what it pays to SOCAN.

Table 4: Re:Sound Music Impressions Relative to those of SOCAN

	Observed Minutes, Re:Sound Repertoire1   Minutes observées, répertoire de Ré:Sonne1	Observed Minutes, Potential Repertoire (PR)1   Minutes observées, admissibilité éventuelle (AE)‑	% Repertoire of Known Music1   Musique connue, en % du répertoire1	% Repertoire Applied to PR   % musique connue appliqué à AE	Potential Public Domain Minutes Deduction2   Déduction au titre du domaine public potentiel (minutes)2	Adjusted Observed Minutes, Re:Sound   Minutes observées rajustées, Ré:Sonne	Hours Tuned   Heures d’écoute	Music Impressions   Impressions musicales
	(A)	(B)	(C)	(D) = (B×C)	(E)	(F) = (A+D-E)	(G)	(H) = (F×G)
Table 4: Re:Sound Music Impressions Relative to those of SOCAN Tableau 4: Impressions musicales de Ré:Sonne relatives à celles de la SOCAN
Radio One	576.1	265.9	47.9	127.4	17.1	686.4	39,505	27,114,893
Radio 2	2985.8	372.1	52.8	196.5	5,0	3177.3	9165	29,119,669
Première Chaîne	904.6	100.6	82.6	83.1	0	987.7	14,059	13,886,012
Espace Musique	3993.9	259.3	67.7	175.5	40.7	4128.7	4262	17,596,716
Total impressions, Re:Sound   Impressions totales, Ré:Sonne								87,717,290
Total Impressions, SOCAN   Impressions totales, SOCAN								113,449,398
Re:Sound repertoire share   Part du répertoire de Ré:Sonne								77.32%

¹ Exhibit Re:Sound-4 at pp. 4-7. / Pièce Ré:Sonne-4 aux pp. 4-7.

² Testimony of Ms. Tay, Transcripts at p. 42. / Témoignage de M^me Tay, transcriptions, p. 42.

[114] Table 5 shows the royalties payable to SOCAN and Re:Sound.

Table 5: SOCAN and Re:Sound Royalties

Year   Année	SOCAN Royalties   SOCAN	Re:Sound   Royalties Ré:Sonne
Table 5: SOCAN and Re:Sound Royalties Tableau 5: Redevances payables à la SOCAN et à Ré:Sonne
2006	$1,563,358	$1,208,764
2007	$1,580,711	$1,222,181
2008	$1,618,332	$1,251,269
2009	$1,644,063	$1,271,164
2010	$1,678,424	$1,297,731
2011	$1,711,933	$1,323,686

[115] The relative use of Re:Sound’s repertoire as compared to SOCAN’s is much higher than on commercial radio. There are several reasons which contribute to this putative anomaly. First, CBC plays a considerable amount of “Canadian content,” far in excess of the statutory minimum. Since many commercial radio stations play Canadian content in amounts at or just above the statutory minimum, this means that CBC plays more Re:Sound repertoire than commercial radio stations. Second, classical music is generally in Re:Sound’s repertoire (since most recordings are less than 50 years old, they are protected) but not in SOCAN’s repertoire (since most composers of classical music died more than 50 years ago, their works are no longer protected). To the extent CBC radio plays classical music, it plays less SOCAN repertoire and more Re:Sound repertoire, thus increasing the fraction of relative use.

VII. SIMULCASTING

[116] All the proposed tariffs under examination, with the possible exception of Re:Sound’s proposed tariff for 2006 and 2007, target only CBC’s conventional, over-the-air radio broadcasting services. Nevertheless, CBC asked that the Board expand the scope of the tariff to include simulcasting. The collectives object to CBC’s request. Alternatively, Re:Sound contends that in the event the Board wished to include it, the tariff base for simulcasting should include Internet advertising revenues.

[117] In SOCAN 22.B-G, the Board stated: “When the time comes to revise SOCAN Tariff 1.C, the Board will be able to reestablish a link between CBC payments and its audience, including that derived from the Internet […]” [69] It stated that some payment should be made to SOCAN for CBC’s Internet simulcasting and those payments should, as much as possible, be derived from the income arising from these Internet simulcasting activities. However, as it noted in its reasons, the Board had no data on advertising or other revenues or additional audience on which to base an increase in SOCAN royalties. [70]

[118] We agree with CBC that the simulcasting of its radio services should be addressed in its main tariff. While separate uses, over-the-air broadcasting and simulcasting involve basically the same product for the same purposes; we see no reason at this time to deal with them in separate tariffs.

[119] Again, the record on which to base the simulcasting tariff is very thin. We do not know what CBC’s simulcasting income is. We are willing this time to accept that these are low enough now not to be of significant help in the matter. We are also willing to accept that segregating simulcasting income from all other Internet revenue may prove difficult. Therefore, we will not attempt to set simulcasting royalties as a function of simulcasting income for the time being.

[120] CBC’s assertion that BBM audience data may already account for the simulcast audience is based on no evidence whatsoever. The only shred of evidence available was provided by Mr. Steinmetz and Ms. Leblanc. Mr. Steinmetz testified that the daily peak simulcast audience is around 10,000 listeners while the combined peak over-the-air audience of the English services is around 1,080,000 listeners – a ratio of 0.93 per cent. Ms Leblanc testified that peak simulcast audience for the French services is about 2 per cent of their peak over-the-air audience. The average of the two numbers, weighted according to audience share, is 1.22 per cent. We increase the royalties accordingly for the right to simulcast the four CBC radio stations at issue in this decision.

[121] The collectives may wish to argue that since their proposals did not deal with simulcasting, the tariff we certify cannot address the issue. This would be incorrect. Both collectives have filed proposed tariffs for CBC’s simulcast of its radio services. The Board is free to deal with the proposed CBC simulcast tariff as it wishes, as long as it acts fairly with CBC and both collectives. CBC is a single user; as a result, everyone interested in the simulcast issue was present before us. CBC asked the Board to address the issue. The collectives knew of this. Fairness is satisfied.

VIII. ROYALTIES

[122] Table 6 below sets out the total royalties CBC will pay.

Table 6: Royalty Payments by Collective and Year

Year   Année	SOCAN Royalties Including Simulcasting   Redevances à la SOCAN, y compris au titre de la diffusion simultanée	Re:Sound Royalties Including Simulcasting   Redevances à Ré:Sonne, y compris au titre de la diffusion simultanée	Total Royalties   Redevances totales
Table 6: Royalty Payments by Collective and Year Tableau 6: Redevances à verser selon la société de gestion et l’année
2006	$1,582,430.75	$1,223,510.57	$2,805,941.32
2007	$1,599,995.74	$1,237,091.54	$2,837,087.27
2008	$1,638,075.63	$1,266,534.31	$2,904,609.95
2009	$1,664,121.04	$1,286,672.21	$2,950,793.25
2010	$1,698,901.17	$1,313,563.66	$3,012,464.83
2011	$1,732,879.19	$1,339,834.93	$3,072,714.12

[123] As a result of our decision, CBC radio will pay in 2006 to SOCAN and Re:Sound royalties that are respectively 5.3 per cent and 26.4 per cent higher than in 2005, excluding simulcasting royalties. While the increase for Re:Sound is much larger than for SOCAN, it rectifies what we view as an imbalance. Put simply, Re:Sound has some catching up to do.

[124] In 2008, CBC radio will pay to SOCAN 0.5 per cent of CBC’s radio operating expenses, including simulcasting royalties. We note, without reiterating our observations concerning CBC’s proposed formula, that this is close to the 0.44 per cent proposed by CBC and the implicit rate of 0.475 certified by the Board in 1991. The percentage is much lower than what applies to non-commercial radios, which is 1.9 per cent. We see no problem in this. There is no evidence that the music use patterns or business models of non-commercial broadcasters and CBC are similar.

IX. TARIFF WORDING

[125] Historically, CBC tariffs have not specified the music use information to be supplied to the collectives. The information CBC supplies to SOCAN has always been based on an informal agreement and does not appear in the record of the present case. Re:Sound tariffs have always provided that CBC provides it with the music use information that it supplies to SOCAN.

[126] The nature and extent of CBC’s reporting obligations was at issue in these proceedings. In a letter to the Board dated January 12, 2011, counsel for Re:Sound conveyed the agreement of the parties on this issue. We made a certain number of comments on the agreement, all of which received satisfactory responses from the parties.

[127] As a result, CBC shall provide each month, in electronic form, broadcast logs containing detailed information about the use of musical works and sound recordings by each of its stations. These reporting requirements are much closer to what we would like all radio broadcasters to supply to the collectives.

[128] Some provisions that are normally included in other tariffs are omitted here, given the special character of the CBC and the fact that this tariff targets a single user. These include the provisions dealing with confidentiality and notices.

X. TRANSITIONAL PROVISIONS

[129] The tariff contains transitional provisions made necessary because the tariff takes effect on January 1, 2006, while it is being certified much later. A table sets out interest factors or multipliers to be used on sums owed in a given month and not already payable on an interim basis starting on January 1, 2006 pursuant to the relevant CBC radio tariff.

[130] In the past, the Board did not apply interest factors to an increase in a pre-existing tariff; such factors were set only in tariffs of first impression.

[131] In our opinion, the practice of using interest factors should be generalized. This tariff is being certified more than four years after it takes effect. It increases the royalties payable. Collectives cannot distribute funds they have not collected. Rights holders cannot spend or invest royalties they have not received. These are opportunity costs; their losses must be accounted for. This reasoning applies equally to tariffs of first impression and to variations to existing tariffs.

[132] Of course, if collectives are to benefit from an interest factor every time royalties are retroactively increased, they should expect the same when royalties are retroactively decreased.

[133] Interest factors were derived as in the past, using month-end Bank Rates, simple interest, and no further penalties. Other interest rates could be used for this purpose. The fact that none of the parties could borrow at the Bank Rate is a reason to suspect that it may be inappropriate to use for interest factors. However, since no party has requested that a different interest rate be used to construct the interest factors, we will not disturb this convention for the time being.

XI. CONCLUSION

[134] Finally, we wish to point out a number of issues that may exist in the design and assumptions of the music logs and the Erin music use study.

[135] Music that should have been counted in may have been left out, and vice versa. Erin Research calculations appear based only on what was considered to be “relevant music” [71] but no indication was provided of what constitutes irrelevant music. Some music that Re:Sound treated as irrelevant (news themes) clearly is relevant to SOCAN, making comparisons between the two collectives unnecessarily complicated.

[136] The classification of music into the various categories used in the Erin Research reports also raised some concerns. Mr. Steinmetz testified that a “demo” is “either something we have commissioned and that we own or it can be something that was sent to us [electronically] without any deep information” and that “custom” are “vanity records” and self-released CDs. [72] CBC reported all these recordings as unpublished; Re:Sound treated them as such. Yet some “demo” recordings are indeed published and as such, potentially eligible for remuneration rights royalties. This would tend to understate the royalties. In the absence of data relating to this subject, we chose to leave well enough alone.

[137] Such ambiguities are not new. [73] In the future, parties may wish to seek Board input before defining the parameters of certain studies and surveys so as to ensure that the results are as reliable and as useful as possible.

Gilles McDougall

Secretary General