Copyright Board Canada

Commission du droit d’auteur Canada

 

Date	2019-08-02
Citation	CB-CDA 2019-056
Regime	Retransmission of Distant Television Signals Copyright Act, section 70 (formerly subsection 73(1))
Members	The Honourable Robert A. Blair Mr. Claude Majeau Mr. J. Nelson Landry

statement of royalties to be collected for the retransmission of distant television signals, in canada, for the years 2014 to 2018

Reasons for decision

I. INTRODUCTION 1

II. BACKGROUND 2

III. CHRONOLOGY OF THE PROCEEDINGS 4

IV. POSITIONS OF THE PARTIES 7

A. The Collectives 7

B. The BDUs 10

V. EVIDENCE 12

A. The Collectives 12

i. Peter Grant, Forum Research 12

ii. Dr. Gerry Wall 14

iii. Professor Jeffrey Church 15

iv. Barry Kiefl 15

v. Carol Cooper 16

B. The BDUs 17

i. Dr. Tasneem Chipty 18

ii. Debra McLaughlin 19

iii. Lori Assheton-Smith 24

iv. Suzanne Blackwell 26

v. BDU Industry Witnesses 28

vi. David Purdy, Rogers Communications 28

vii. Geoff Wright, Bell 29

viii. Gary Pizante, Shaw Communications 30

ix. Ann Mainville-Neeson, TELUS 31

x. Marie Ginette Lepage, Vidéotron G.P. 31

VI. LEGAL ANALYSIS 31

A. The Copyright Act’s legal framework with respect to the retransmission regime 31

B. Legal Issues to be considered 33

C. The Retransmission Regulations 33

i. The Interpretation of the Statute and Regulations 34

ii. Position of the Parties Respecting the Status of the Retransmission Regulations 36

iii. Analysis Respecting the Status of the Retransmission Regulations 37

iv. The Collectives’ Argument on the Application of the Retransmission Regulations 38

v. The BDUs Argument on Application 39

vi. Analysis Respecting the Application of the Retransmission Regulations in the Present Proceeding 40

vii. The effects of the Broadcasting Act and regulations made thereunder 42

a. The Collectives 42

b. The BDUs 43

c. Analysis 44

viii. The Board’s Power to Vary and the Collectives’ Revised Tariff Proposal 44

ix. The effect of Broadcasting Regulatory Policy CRTC 2015-96 on the current proceedings 48

a. The Collectives 48

b. The BDUs 48

c. Analysis 49

d. Technological neutrality 49

VII. ECONOMIC ANALYSIS 50

A. Professor Church’s proxy approach 51

i. Description 51

ii. BDU’s comments 53

iii. Analysis 56

a. Representativeness of his selected 24 U.S. specialty services 56

b. Presence of niche services 57

c. Are prices of the U.S. services competitive? 57

B. Dr. Chipty’s proxy approach 58

i. Description 58

a. Adjusting for the difference in popularity 61

b. Substitution opportunities 62

ii. Collectives’ comments 63

iii. Analysis 68

C. Dr. Wall’s approaches 69

i. Description 69

ii. BDU’s comments 71

iii. Analysis 74

D. The trend-analysis approach 76

i. Description 76

ii. Collectives’ comments 76

iii. Analysis 77

E. The Board’s Approach 79

i. Rate for 2014 79

a. Cost of programming 81

b. Market power of specialty services 82

c. Program substitutability 83

ii. Rates for 2015-2018 86

iii. Rates for Retransmitters with no more than 6,000 subscribers 86

iv. Discounts 88

F. Rates approved and total royalties generated 88

VIII. ALLOCATION 89

IX. TARIFF WORDING 89

I. INTRODUCTION

[1] On March 28, 2013, nine Collectives [1] jointly filed a proposed tariff for the retransmission of distant television signals for the years 2014-2018 (“Proposed Tariff”). The Proposed Tariff was filed pursuant to section 71 of the Copyright Act, [2] and was published in the Canada Gazette on June 1, 2013. Prospective users or their representatives were thereby advised of their right to object to the Proposed Tariff.

[2] On July 31, 2013, Bell Canada, Bragg Communications Inc. (operating as Eastlink), Rogers Communications Inc., Shaw Communications Inc., Cogeco Cable Inc., Videotron G.P., TELUS Communications Company, MTS Inc., and the Canadian Cable Systems Alliance (CCSA) (collectively the broadcast distribution undertakings or “BDUs”) jointly filed timely objections to the Proposed Tariff.

[3] At the time of filing, the Collectives’ proposed rates for large Retransmitters with more than 6,000 subscribers ranged from $1.06 per subscriber per month in 2014 to $1.38 in 2018, as described in more detail later in these reasons. This compared to the previous rate of $0.98 per subscriber per month which had been in effect in the last year of the previous period, 2009-2013.

[4] In May 2015, however – after the exchange of interrogatories and in conjunction with the filing of their Statement of Case – the Collectives proposed significantly higher royalty rates rising from $2.00 per subscriber per month in 2014 to $2.38 in 2018. Lower rates were proposed in both cases for smaller-sized Retransmitters. The manner and timing of these proposed increases, and their implications, were a source of contention in the proceedings and will be dealt with in more detail below.

[5] For the reasons that follow, we have concluded that the following retransmission royalties are fair and equitable for various sizes of Retransmitters for the years 2014-2018, and approve the tariff accordingly:

Table 1: Monthly Rate for each premise receiving one or more distant signals (in dollars), 2014-2018

Monthly Rate for each premise

Number of premises	2014	2015	2016-2018
Up to 1,500	0.49	0.57	0.60
1,501 - 2,000	0.54	0.62	0.65
2,001 - 2,500	0.60	0.68	0.71
2,501 - 3,000	0.66	0.74	0.77
3,001 - 3,500	0.71	0.79	0.82
3,501 - 4,000	0.77	0.85	0.88
4,001 - 4,500	0.83	0.91	0.94
4,501 - 5,000	0.89	0.97	1.00
5,001 - 5,500	0.94	1.02	1.05
5,501 - 6,000	1.00	1.08	1.11
6,000+	1.06	1.14	1.17

II. BACKGROUND

[6] The royalties payable under the tariff are a part of the retransmission regime established under the Act. This regime permits BDUs to retransmit over-the-air (OTA) broadcast signals by capturing, packaging and selling them to their subscribers without the consent of the broadcasters or the owners of the broadcast programs. As a condition of this retransmission regime, where the retransmitted signals are “distant” signals, BDUs must pay royalties set by the Board to various Collective Societies that have filed tariffs. [3]

[7] A distant television signal is a signal that is not “local”. Local signals are defined as follows. [4] A local analog signal is a TV signal that covers an area within the radius of 32 km from the Grade B contour of the station. A local digital signal is a TV signal that covers an area within the radius of 32 km from the Noise-Limited Bounding Contour (NLBC) of the station.

[8] In 1990, the Board certified a retransmission tariff at $0.70 per subscriber per month for the first time. [5] In its decision that followed a lengthy hearing, the Board indicated that the tariff should satisfy the following six criteria: [6] (i) be fair and equitable; (ii) reflect Canadian circumstances; (iii) given a choice of approaches, that equally compensate copyright owners, be the one that results in the least possible disruption to the cable services available to subscribers; (iv) be based on a set of statistics for a test year; (v) reflect the actual retransmission of programs and recognize that some programs may be more valuable than others; and (vi) be simple to administer, transparent and comprehensible.

[9] In its 1990 Decision, the Board also took into account the effects of the tariff on three groups: the Retransmitters (the users), the subscribers (the end-users) and the collecting bodies (the rights owners). As the Board explained, first, “a fair and equitable tariff should impose a royalty consonant with the benefits Retransmitters receive from the use of distant signals.” Moreover, “Retransmitters should pay to copyright owners no less than the value of the harm caused to them by the use of their works.” Second, all Canadian subscribers should be treated in a similar fashion. In other words, a fair and equitable royalty should not magnify the existing variance in cable fees. Third, a fair and equitable royalty scheme should generate appropriate compensation and distribute it fairly among the collecting bodies. [7]

[10] The Act, however, requires the establishment of a preferential rate for small retransmission systems. As explained in the 1990 Decision, a fair and equitable tariff may treat Retransmitters in different circumstances differently. Systems differ as to their size, their location and the number of distant signals that they carry. In particular, while small systems tend to carry many distant signals, they tend to have higher average fixed and operating costs, and as a result, tend to charge higher monthly fees. As such, Parliament legislated preferential treatment for some small systems. [8] Accordingly, in its 1990 Decision, the Board set the royalty at $100 per annum for each small retransmission system. [9]

[11] On November 28, 1991, pursuant to s. 70.63(4) of the Act, the Governor in Council adopted the Retransmission Royalties Criteria Regulations (the “Retransmission Regulations”). [10] The Board must have regard to these criteria in establishing the amount of royalties to be paid under the tariff. The Retransmission Regulations require the Board to take into account: (i) royalties paid for the retransmission of distant signals in the United States under the retransmission regime in the United States; (ii) the effects on the retransmission of distant signals in Canada of the application of the Broadcasting Act and regulations made thereunder; and (iii) royalties and related terms and conditions stipulated in written agreements in respect of royalties for the retransmission of distant signals in Canada that have been reached between collecting bodies and Retransmitters and that are submitted to the Board in their entirety. One issue in this proceeding is whether those Regulations continue to govern the Board’s deliberations, given amendments to the Act in 1997.

[12] A second hearing took place in 1993. In its 1993 Decision, the Board noted that there was no reason to abandon or modify the rate-setting principles adopted in its 1990 Decision. As a result, the Board left the rates for 1992, 1993, and 1994 unchanged at the level set for 1990 and 1991. [11]

[13] There has not been a hearing relating to the tariff rates since 1993. However, pursuant to a series of agreements between the Collectives and the BDUs, the Board certified tariffs in 1996, 2000, 2003, 2008, and 2013. The first three of these agreements did not change the tariff rate, which means the tariff rates established in the 1990 Decision remained the same through to 2003.

[14] In 2008, the Board certified tariff rates for 2004 to 2008 pursuant to an agreement between the parties. In its Decision, the Board approved an annual increase of $0.03 in rates per subscriber per month. As a result, the retransmission tariff was gradually increased from $0.73 per subscriber per month in 2004 to $0.85 per subscriber per month in 2008. [12]

[15] In this decision, the Board noted that the increase in rates was justified for several reasons. [13] First, the retransmission market had evolved considerably since 1990 and, in particular, the number of distant signals available to the average subscriber had grown substantially. Second, the rates had remained the same since 1990, whereas the Consumer Price Index (CPI) had increased by more than 45 per cent.

[16] For the period 2009 to 2013, retransmission rates were agreed upon by the parties and certified by the Board in 2013. The rate increased by $0.05 per subscriber per month in 2009 and, over the subsequent four-year period, by $0.02 per subscriber per month annually. Consequently, as of 2013, the tariff rate was $0.98 per subscriber per month for BDUs having more than 6,000 subscribers. [14]

III. CHRONOLOGY OF THE PROCEEDINGS

[17] On October 31, 2013 – after the filing of Objections – the Collectives and the BDUs jointly requested the certification of an interim tariff for the retransmission of distant television signals for the period commencing January 1, 2014. On December 6, 2013, the Collectives and the BDUs jointly proposed that the 2014-2018 Interim Tariff continue the terms of the recently certified 2009-2013 Retransmission Tariff.

[18] On December 19, 2013, the Board issued an interim decision [15] in regards to Television and Radio Retransmission for the years 2014-2018. According to the interim decision, the Television Retransmission Tariff, 2009-2013 would remain applicable, unless modified, until the final tariff was certified for the years 2014-2018.

[19] On March, 14, 2014, counsel for MTS Inc. and CCSA informed the Board that they withdrew their objections to the Proposed Tariff and would not be party to the proceeding. Moreover, effective March 17, 2017, MTS Inc. was wholly acquired by Bell Canada. On December 12, 2017, counsel for Bell Canada informed the Board that Bell Canada withdrew any and all statements of objection filed on behalf of MTS prior to the acquisition.

[20] On March 25, 2015, Bragg Communications (operating as “Eastlink”) withdrew its objection to the Proposed Tariff. It also requested that Eastlink’s interrogatory responses be destroyed and not made part of the record. On March 27, 2015, the Collectives (except CRRA) objected to Eastlink’s request, arguing that a party should not be permitted to prevent the Collectives and the Board from considering relevant information if it no longer objects to the Proposed Tariff. On April 10, 2015, the Board denied Eastlink’s request. [16]

[21] On April 2, 2015, further to CRC’s request, the Board revised the hearing schedule and set a date for the hearing beginning Tuesday, November 24, 2015. [17]

[22] The oral hearing was conducted over fifteen days, spread over four hearing sessions in the months of November and December 2015, and January, March, and August 2016.

[23] On March 11, 2016, the Board received comments from CCSA with respect to the application of the principles of non ultra petita doctrine and procedural fairness, relating to the manner and timing of the Collectives’ introduction of new proposals for royalty rates in this proceeding (the “CCSA letter”). CCSA argued that while some of its smaller members are subject only to the annual, flat-fee royalty, many other systems operated by CCSA’s members pay the monthly, per-subscriber rates set out in the retransmission tariffs. On March 14, 2016, the Board gave notice [18] that the CCSA letter would be made part of the public record in the file, and that parties could either provide a written response before oral arguments or address it during oral arguments which were scheduled for March 22-23, 2016.

[24] On March 15, 2016, the Collectives wrote to the Board to express their concerns about the CCSA letter. The Collectives submitted that the CCSA letter should not have been accepted by the Board and should be removed from the public record. The Collectives also requested that, if the Board was not willing to remove the CCSA letter from the public record, the scheduled oral argument be delayed until the CCSA’s evidence could be tested by the Collectives.

[25] On March 16, 2016, the Board ruled [19] that the CCSA letter would remain part of the public record, and that oral arguments would proceed as planned on all issues except on the issues of non ultra petita and procedural fairness. It further ordered that any party claiming that interrogatories and/or cross-examination of CCSA’s representatives or members are required to test factual allegations made in the letter provide the Board with a proposed process to that effect by March 24, 2016.

[26] On March 24, 2016, the Collectives informed the Board that they were prepared to serve limited interrogatories on CCSA no later than April 4. On June 6, 2016, the Collectives requested the right to cross-examine CCSA if its letter was to remain part of the record. On July 18, 2016, the Board informed the parties that the hearing would resume on Monday, August 29, 2016, with the cross-examination of CCSA. Parties were asked to file any additional evidence that might arise from the CCSA letter no later than Monday, August 15, 2016. Oral arguments proceeded on the only issues of non ultra petita and procedural fairness on August 30, 2016.

[27] On November 6, 2014, the Collectives proposed a procedure for addressing the allocation of royalties under the tariff separately from the issue of the quantum of the royalties, and that the hearing for the 2014-2018 television retransmission tariff address only the quantum of royalties. [20] The request was made on behalf of all of the Collectives except DRTVC, which had not taken a position on the issue. On November 7, 2014, the Board accepted the proposal, without further comment from the parties.

[28] On March 23, 2016, the Board requested that the Collectives provide a status report on the negotiations regarding the allocation of royalties among them no later than April 15, 2016. The Collectives reported back to the Board on the status of their negotiations several times since then. On October 19, 2018, following a request from the Board, the Collectives provided a description of the issues that remained in dispute and proposed a procedure leading to a hearing for dealing with those issues. Accordingly, the Board issued the schedule of proceedings and set the date for the hearing to June 18, 2019.

[29] On April 7, 2016, the parties responded to the Board’s questions posed in Notice 2016-024 dated March 7, 2016. On October 27, 2016, the Board issued Notice 2016-088 to the parties requesting them to comment on some evidence with respect to the profit margin of U.S. specialty services and viewership of substitution opportunities of TV programming. Specialty television services are services that provide niche programming (music, sports, public affairs) or target a specialized audience (children, specific cultural groups, etc.). They are only available through cable or satellite. They are never broadcasted over-the-air and, as such, do not fall under the realm of the retransmission regime.

[30] The Board also requested the disaggregated viewing data the BDUs’ expert had used in her analysis. On November 18, 2016, the Board issued Notice 2016-094 to clarify some points raised by the Collectives in respect of the Board’s questions in Notice 2016-088. On December 9, 2016, the parties responded to the Board’s questions as set out in Board Notice 2016-088 and elaborated upon in Board Notice 2016-094.

[31] On December 12, 2018, the Board received a letter from the Collectives to this proceeding, requesting that the Board render a decision with respect to quantum as soon as possible. They submitted that a decision on quantum might assist the parties in reaching an agreement on the allocation issues as well as alleviating the burden of having to maintain significant monetary reserves and/or holding back on the distribution of royalties. The BDUs did not object to the request.

[32] On December 18, 2018, the Board issued its decision in respect of the quantum of the television retransmission tariff, with reasons to follow. [21]

[33] On January 31, 2019, the Collectives confirmed to the Board that an agreement on allocation among all of them had been reached.

[34] These reasons deal with the tariff rate and the allocation of royalties among the respective Collectives, the only matters at issue in this proceeding. No structural changes have been proposed to the retransmission tariff by either party, to the way the payments are triggered, and to certain long-standing discounts such as for the BDU’s size, for institutional consumers and for francophone markets.

IV. POSITIONS OF THE PARTIES

A. The Collectives

[35] As noted above, the Collectives’ Proposed Tariff for the years 2014-2018 was filed with the Board in March 2013. At the time, the Collectives’ proposed rates were ranging from $1.06 in 2014 to $1.38 in 2018, as shown in Table 2 below. These proposed rates were published in the Canada Gazette on June 1, 2013.

[36] In May 2015 and after the exchange of interrogatories, the Collectives filed their Statement of Case in which they proposed a further increase of the retransmission royalty rate to $2 per subscriber per month for 2014. The Collectives also proposed an annual adjustment factor of 4.4 per cent, to be applied to the subsequent years 2015 through 2018. This annual adjustment factor was the average annual increase in the cost of a basic cable subscription from 2010 to 2014. This adjustment leads to a rate of $2.38 per subscriber per month in 2018, as shown in the following Table.

Table 2: Monthly Rates Proposed by the Collectives, for Retransmitters with 6,001 premises and over (in dollars) /

Tableau 2 : Taux mensuels proposés par les sociétés de gestion pour les retransmetteurs possédants 6001 locaux et plus (en dollars)

Tableau 2 : Taux mensuels proposés par les sociétés de gestion pour les retransmetteurs possédants 6001 locaux et plus

	2014	2015	2016	2017	2018
As published in the Canada Gazette / Tels que publiés dans la Gazette du Canada	1.06	1.14	1.22	1.30	1.38
As provided in the Statement of Case / Tels que proposés dans l’énoncé de cause	2.00	2.09	2.18	2.28	2.38

[37] To justify the higher rates in their request, the Collectives argue that the current proceeding is the first one in 20 years to examine the retransmission rates in depth. They also note that this is the first time in over a decade that BDUs have had to respond to the Collectives’ interrogatories.

[38] The Collectives state that the interrogatory process, which occurred after the Collectives filed the Proposed Tariff for the years 2014-2018, provided them with critical confidential information to assess the value of distant signals. The Collectives explained that this detailed financial information shows the new and highly lucrative ways the BDUs have developed to sell distant signals to their subscribers and their profit therefrom. According to the Collectives, this information was not previously available to them in any form.

[39] According to the Collectives, the evidence shows that significant changes occurred in the industry since the last time the Board held a hearing on this issue in the early 1990s. The Collectives claim that while these changes increased the overall value of distant signals to BDUs, this increased value is not reflected in recently certified tariffs.

[40] The Collectives provide several examples of these changes. First, there has been a significant increase in the number of distant signals retransmitted by BDUs over the last several years, in the form of both in-time-zone distant signals and time-shifted distant signals. Consequently, the average BDU subscriber, who received 4.56 distant signals in 1990, received 55.3 distant signals in 2014. [22] The Collectives argue that the growth in distant-signal delivery reflects the BDUs’ assessment that subscribers want distant signals, and that BDUs are economically better off by carrying and selling them. [23] According to the Collectives, this behaviour reflects the continued importance and economic value of distant signals to the BDUs. [24]

[41] Second, the nature of distant signals has changed. For example, the new feature of time-shifting means that many distant signals duplicate local signals but from a different time zone. The Collectives argue that time-shifted signals provide additional value to subscribers in a variety of ways including by offering different types of programming and additional opportunities to view their favorite programs. [25]

[42] Third, the value of distant signals has changed. For example, BDUs now promote distant signals heavily to their subscribers; some BDUs sell packages consisting entirely of distant signals.

[43] Using these newly available data and expert opinion, the Collectives revised their proposed rate of $1.06 per subscriber per month filed in 2013 to a new rate of $2.00 per subscriber per month in 2014, subject to a variety of discounts discussed below.

[44] For small retransmission systems, [26] the Collectives propose to continue the $100 annual, flat-rate royalty that has been in place since 1990 with an exception for systems with 2,000 or fewer subscribers that are located within the service area of other systems with more than 2,000 subscribers similar to past retransmission tariffs certified by the Board. They also propose to maintain the discounts for retransmission systems that serve francophone markets, certain non-residential premises, etc. [27]

[45] Similarly, the Collectives propose to maintain the discounts previously set by the Board, including those for retransmission systems that serve fewer than 6,000 subscribers. Those discounts were set in the 1990 Decision at five cents per subscriber per month for each 500 subscribers fewer than 6,000 subscribers. The Collectives propose to maintain both the five-cent discounts and, except for the smallest category, the 500 subscribers “tiers”.

B. The BDUs

[46] In their Statement of Case, the BDUs argue that the Collectives failed to establish that there was any increase in the value of distant signal programming since 2013, especially of an order that would justify the increases in the existing rates of the revised request. The BDUs argue that, on the contrary, the value of distant signals has steadily been declining for the past several years and will continue to decline in the future.

[47] The BDUs submit that the rate for 2014 should remain unchanged at $0.98 per subscriber per month and should decrease by $0.02 per subscriber per month for each subsequent year to reach a rate of $0.90 per subscriber per month in 2018. BDUs agree with the Collectives that the graduated rates for mid-sized systems and other discounts should continue. [28]

[48] To support this rate proposal, the BDUs suggest that the broadcasting industry is in a period of technological transition, and that the demand for, and value of distant signals is declining due to the emergence of new and better substitute sources of programming. More specifically, the BDUs argue the following.

[49] First, the proportion of viewing of distant-signal programming to total television viewing has decreased since the Board first certified the tariff in 1990. Also, average minutes viewing of distant signals continues to decrease despite the number of distant signals being carried. [29]

[50] Second, most households only watch programming on approximately three distant signals to any significant degree, despite the fact that the average BDU retransmits 55 distant signals. Most households do not watch any of the programming on the majority of distant signals available to them.

[51] Third, on a per-service basis, the revenue and programming expenditures of over-the-air and comparable specialty television services have either declined or remained flat over the last several years.

[52] Fourth, the same programming that is carried on distant signals is increasingly available through on-demand services available from BDUs and over the Internet, as well as “over-the-top” (OTT) services such as Netflix, Shomi [30] and CraveTV. The BDUs argue that these more-convenient sources of television programming have diminished and will continue to diminish the importance and value of distant signals as a source of TV programming.

[53] Fifth, the widespread adoption of personal video recorders (PVR) is replacing the need for BDU subscribers to rely on distant signals from different time zones to access programming at alternative times of the day.

[54] Sixth, among consumers, there is a shift away from traditional sources of television programming, including distant signals, towards more flexible and convenient technologies including mobile streaming applications, video-on-demand services (VOD), and Internet services.

[55] Seventh, as the BDUs argued in the 2016 hearing, recent changes to broadcasting regulations mean that by the end of 2016, subscribers will have much greater control over the television signals and services they receive. (In fact, these changes came into force on December 1, 2016). [31] The BDUs claim that this change will lead to a decrease in the retransmission of distant signals which are the least watched and least valuable category of television services offered by BDUs.

[56] Accordingly, the BDUs submit that, instead of revisiting the proxy analysis adopted by the Board more than 20 years ago, the Board should take as the starting point the existing rate of $0.98, which was certified in 2013, and consider whether there is any evidence of changes in distant signal retransmission since 2013 that would justify a change to the rate.

[57] To support this approach, the BDUs make three arguments. First, the 2013 rate was agreed to by the nine Collectives and the eight objectors representing hundreds of individual retransmission systems in Canada. As such, it represents what the parties believed to be a fair and equitable rate in 2013. Second, pursuant to the Retransmission Regulations, the Board must have regard to the agreement on the 2013 rate in determining what constitute a fair and equitable rate. Third, the 2013 rate is the result of the latest in a series of freely negotiated settlements between the parties over the period 1994-2013.

[58] Moreover, the BDUs state that the analyses prepared by the two Collectives’ experts lack appropriate adjustments to account for differences between distant-signal programming and their selected benchmark, and hence, are incapable of supporting the Collectives’ requested rate increases. [32]

[59] The BDUs also object to the “revised” Proposed Tariff of the Collectives. The BDUs do not represent all Retransmitters subject to the tariff. The BDUs state that, as a matter of procedural fairness, the Board should not even consider the revised rates since the Collectives had filed with the Board their initial rates, which the Board published in the Canada Gazette two years before filing the new ones.

[60] Notwithstanding the fact that the Board is considering the Proposed Tariff many years after its publication in the Canada Gazette, the BDUs argue that if the Collectives were permitted to request increased rates on the basis of new information filed during the hearings, it would render meaningless the 60-day statutory notice period for objecting to the Proposed Tariff, demonstrating that the Collectives’ conduct is a disregard for the Act’s statutory procedure.

[61] The Collectives, however, argue that the non ultra petita principle does not apply to the Board; the Board may assess a revised tariff proposal that seeks a greater royalty than the original published tariff. The Collectives state that the fairness concerns surrounding any upward revision to the original published tariff are satisfied since the affected stakeholders have been given adequate notice of the revision, and representative stakeholders had the opportunity to participate in the proceedings on behalf of all who are similarly situated. The Collectives present an analysis [33] which shows that the participating BDUs represent collectively the vast majority of all royalty payments made by large and medium BDUs. The Collectives also state that there was no evidence that the remaining BDUs, all of whom the Collectives argue received notice of the revision, wanted to participate.

V. EVIDENCE

A. The Collectives

[62] In this proceeding, several Collectives retained experts to analyze the marketplace for, and the economics of, the retransmission of distant signals. The Collectives’ expert evidence, and the conclusions to which it leads, are summarized below.

i. Peter Grant, Forum Research [34]

[63] Mr. Grant explains that, in 1990, the average subscriber received 4.56 distant signals. Most of the distant signals were signals from the U.S. border stations associated with the ABC, CBS, NBC and PBS networks, also known as the “3+1”. About a quarter of all distant signals were Canadian. However, the total number of distant signals has risen significantly since then.

[64] The Collectives and BDUs jointly commissioned a study [35] from Mediastats. The study was to provide information on the average numbers of distant signals per residential subscriber for the period from 2004 to 2014. It concluded that the average number of distant signals had risen from 4.56 in 1990 to 25.2 per subscriber by 2004, and to 55.3 per subscriber by 2014. The Collectives and the BDUs each interpreted the study separately; in the Collectives’ case, that interpretation came from Mr. Grant.

[65] In Mr. Grant’s view, the increases since 1990, and subsequently since 2004, are the result of various factors including the commencement of retransmission of the Fox network by Canadian BDUs (creating the “4+1”) in 1994, the advent of direct-to-home (DTH) satellite services in 1997, the availability of digital (versus analog) services starting in 2000, and the emergence of Internet Protocol television (IPTV) offerings after that. These factors, in turn, led to the introduction of time-shifted distant signals and high-definition (HD) signals.

[66] According to Mr. Grant, much of the increase in the number of distant signals is because of the introduction of time-shifted distant signals. He further indicated that surveys of BDU subscribers over the past 12 years consistently show that time-shifted distant signals are one of the most valuable services on cable or satellite. [36]

[67] With respect to the predicted impact of the Canadian Radio-television and Telecommunications Commission’s (CRTC) “Let’s Talk TV” proceeding and the introduction of “skinny basic” in 2016, Mr. Grant states that this change may result in fewer distant signals on average being delivered to Canadian subscribers over time. Mr. Grant also points out that all other packages offered by BDUs would still contain distant signals and concludes that it would still take a number of years before the changes can be implemented and any effects assessed. [37] The Collectives add [38] that even a potential reduction in the number of distant signals cannot be equated with a drop in the value of distant signals since BDUs would continue to develop new and lucrative ways to package and sell distant signals to maximize their profits. [39]

[68] Mr. Grant concludes that the TV retransmission industry and marketplace changed significantly since 1990, and that this change requires an update to royalties first set in 1990.

ii. Dr. Gerry Wall [40]

[69] Dr. Wall uses three methods to estimate the value of distant signals. His main methods are based on a direct-market approach, which estimate the market price of distant signals based on the retail prices charged by BDUs to the subscribers (Methods 1 and 2 below). As Dr. Wall explains, there were no market data to estimate the value of distant signals directly when the Board held the first two Retransmission hearings in 1990 and 1993. As a result, the Board used a proxy approach based on the wholesale price of the A&E signal. However, since the retail price of distant signals is now available in Canada, this information provides “the most relevant and direct means” [41] of valuing distant signals to set a fair and reasonable retransmission royalty.

[70] In Method 1, Dr. Wall first selects packages that consist mostly or entirely of time-shifted distant signals. By using the average retail price of these distant signals and financial data provided by TELUS and Rogers during the interrogatory process, Dr. Wall estimates an average wholesale price for distant signals. By multiplying this average wholesale price by the average number of distant signals received in 2013 (i.e. 54.0), Dr. Wall estimates a price of $4.97 per subscriber per month for distant signals.

[71] In Method 2, Dr. Wall considers the implicit prices of distant signals included in basic and extended basic service packages offered by BDUs. As Dr. Wall explained, these packages typically include a large number of non-distant signals. As a result, the estimates of distant signal values are not as clear cut. Dr. Wall uses the basic and extended-basic service packages offered by Bell, Rogers, Shaw and TELUS in 2013 to estimate the value of distant signals in this method. After estimating the implicit, weighted-average retail price per signal of the packages under consideration and by using the financial data provided by the BDUs and the mark-ups for services estimated in Method 1, Dr. Wall estimates an implicit wholesale price range for distant signals. By multiplying the estimated price range by the quantity of distant signals received by subscribers in 2013, Dr. Wall estimates a price range for distant signals. Method 1’s estimated price of $4.97 per subscriber per month for distant signals falls roughly in the middle of the confidential price range estimated in Method 2. As such, Dr. Wall argues that the results of Methods 1 and 2 confirm one another.

[72] Dr. Wall’s proxy approach (Method 3) builds on the approach used by the Board in its 1990 Decision. In this decision, the Board used A&E as a proxy for a distant signal. The price of A&E was $0.25 in 1990. This price was discounted to $0.15 to adjust for simultaneous substitution of distant signals and possible market power of A&E as a specialty service. This value (i.e. $0.15) was then multiplied with by the average number of distant signals a subscriber received in 1990 (i.e. 4.56) to obtain a price of $0.70 for distant signals. Dr. Wall used the growth of the price of A&E and the growth of the number of distant signals per subscribers in 2009-2013 to update the last certified price of $0.98 for 2014-2018. [42]

[73] Details of Dr. Wall’s analysis and the BDUs’ responses thereto are reviewed in the economic analysis section of this decision.

iii. Professor Jeffrey Church [43]

[74] Professor Church conducts a proxy analysis to examine the market price paid by BDUs for a comparable set of services supplied in a competitive market. This analysis includes three tasks as follows. [44] First, it identifies a group of programming (services) which are qualitatively comparable to distant signals. Second, it verifies that the comparable set of channels is supplied competitively. Third, it calculates payments to the comparable set of channels.

[75] Professor Church chooses a set of U.S. specialty services as the proxy group and concludes that these U.S. specialty services have comparable value to distant signals.

[76] In the next step, Professor Church calculates total payments made by four English-Canadian BDUs (Bell, Rogers, Shaw, and TELUS) for a group of 24 U.S. specialty services he selected. By dividing this total payment for U.S. specialty services by the total number of subscribers to those BDUs, Professor Church obtains a monthly per-subscriber price of $2.76 for U.S. specialty services. Under the assumption that U.S. specialty services have a similar value to distant signals, this is also the price of distant signals. Professor Church does not make adjustments to this rate, since he argues that this rate is a conservative proxy for a competitive-market valuation of distant signals.

[77] Details of Professor Church’s analysis and BDUs’ responses are reviewed in the economic analysis section of this decision.

iv. Barry Kiefl [45]

[78] Mr. Kiefl testified about issues related to various viewing and tuning measures used by other witnesses.

[79] Mr. Kiefl explained that there are three ways to measure TV audiences: Diary, People Meters, and Portable People Meters (PPMs). Diary is a printed booklet in which the respondents record their TV viewing for a one-week period in 15-minute blocks. People Meters use a hard-wired device to all TV sets in a household which sends viewing data to the measurement company (Numeris). People Meters record the channel received on a TV set minute by minute. A hand-held device is used by household members to indicate their presence in the room. PPMs are small, pager-size devices that respondents carry with them. PPMs detect inaudible codes embedded in TV programs by networks and stations and automatically measure individual viewing by the minute throughout the year. PPMs also detect whether a respondent is within earshot of a TV set. They are equipped with a cellular modem that sends minute-by-minute viewing data to Numeris.

[80] According to Mr. Kiefl, PPMs have been widely accepted as the gold standard in Canada. As Mr. Kiefl explained, diaries suffer from many limitations and methodological problems, including completion-error problems and the fact that they only measure viewing for a few weeks a year. As such, national advertisers, broadcasters, industry groups and the CRTC now rely on PPM data, and not Diary data, for measuring general national viewing trends in Canada.

[81] Mr. Kiefl also commented on the reports Ms. McLaughlin and Dr. Chipty prepared for the BDUs. His comments will be dealt with in our discussion of Ms. McLaughlin’s and Dr. Chipty’s expert reports.

[82] The BDUs note that Mr. Kiefl admits that the Numeris PPM system does not detect roughly half of the U.S. 4+1 signals in Canada. [46]

v. Carol Cooper [47]

[83] Ms. Cooper calculates the percentage of annual royalties attributable to the Objectors who are or have been party to the tariff. The purpose of her calculation is to assess the degree to which retransmission systems affected by the revised rates are, or were, Objectors and already before the Board in this proceeding.

[84] Ms. Cooper states that the vast majority of royalties from medium and large Retransmitters in 2014 was owed by the Objectors. This proportion [48] becomes higher if past objectors, who objected to the Proposed Tariff and then withdrew their objections, are included. In her calculation, Ms. Cooper excluded the royalties relating to small retransmission systems, as they are subject to a flat rate that was not proposed to be amended in the proceeding.

[85] Ms. Cooper also explains that, in order to ensure that the non-objecting retransmission systems were made aware of the revised requested rates, the Collectives engaged Mediastats Inc. to send out a notification regarding the revised royalty rates to non-objectors on its mailing list. As a result of this effort, a letter was sent to all medium and large non-objecting retransmission systems on the mailing list of CRC.

[86] Second, Ms. Cooper admits that while the letter notifies BDUs that there is a Copyright Board proceeding underway to consider the revised tariff which are higher than the originally published ones, the letter did not give BDUs any information about whether it was possible to intervene in the proceeding or how to intervene or file comments in relation to the revised rate proposals. [49]

[87] Finally, Ms. Cooper further acknowledges that, prior to receiving the letter from Collectives’ counsel, a BDU that was not a party to the Copyright Board proceeding would have been expecting, at most, an eight per cent increase in the rates between 2013 and 2014, rather than the 104 per cent increase reflected in the Collectives’ revised rates. [50]

B. The BDUs

[88] The BDUs undertook separate analyses on the value of distant signals which are summarized below. The BDUs state that their analyses strongly suggest that the existing rate is too high, and that the value of distant signals programming has been declining for a number of years and will continue to decline. [51] As a result, they submit that the existing rate should be maintained for 2014 and decrease for the following years.

[89] With respect to the Collectives’ expert reports, the BDUs submit that the Collectives rely primarily on Professor Church’s and Dr. Wall’s export reports to support their proposed rate increases of more than 100 per cent. However, the BDUs argue that the Collectives’ own witnesses contradict each other on a number of key issues, calling into question the soundness of their approaches.

[90] The BDUs provide the following examples to support this argument. First, Dr. Wall testified that an adjustment for simultaneous substitution is required in a proxy analysis [52] but Professor Church does not make such an adjustment. Second, Professor Church says that Canadian Category B specialty services cannot be used in a benchmark group for distant signal programming, but Dr. Wall includes those and other Canadian specialty services in his analysis. Category B services are a class of Canadian specialty television services that focus on a specific genre (for example, music, children’s programming, weather, comedy programming). These services may be carried by all digital cable television and direct broadcast satellite providers, as opposed to category A services which must be carried (with the exception of general interest national news and sports specialty services which are designated as Category C services). Third, Mr. Kiefl criticizes the use of set-top box data and Numeris Diary data, but Professor Church relies exclusively on these two sources for his analysis. Fourth, Professor Church counts all viewing of U.S. 4+1 signals in the Toronto area as distant viewing, despite the fact that Mr. Grant explained in detail how these signals are local to large portions of the Toronto area. [53]

[91] Finally, the BDUs add that the Collectives have provided the Board with three different benchmark analyses that only provide a snapshot with absolutely no evidence on changes in the value of distant signal programming in recent years.

i. Dr. Tasneem Chipty [54]

[92] Dr. Chipty, of the Analysis Group, prepared an economic analysis of the value of distant-signal programming to BDUs in order to suggest an appropriate royalty. The BDUs also asked Dr. Chipty to assess the arguments and analyses put forth by the Collectives’ economic experts.

[93] Dr. Chipty’s approach to estimate a reasonable distant signal royalty rate for distant signal is also based on proxy analysis. Dr. Chipty uses both U.S. specialty services and Canadian Category B specialty services as a proxy for distant signals as, in her opinion, a combination of the two specialty services is a better benchmark than the U.S. specialty services alone.

[94] Among the services provided to her by the BDUs, Dr. Chipty selects a benchmark group of 18 U.S. specialty services [55] and 47 Canadian Category B specialty services offered by the six BDUs Bell, Cogeco, Eastlink, Rogers, Shaw, and TELUS. Dr. Chipty excluded certain services of these BDUs from her benchmark if she lacked information on their price or number of subscribers or if their relevant information was missing in the set-top box data she received from the BDUs.

[95] To obtain a starting benchmark rate, she divides total available payments by the six BDUs for the benchmark group by the number of subscribers to these BDUs. This yields a benchmark price of $3.29 per subscriber per month for 2013. After adjusting for inflation, she obtains a benchmark price of $3.38 per subscriber per month for 2015. [56] After isolating the cost of programming by removing the profit from the services, her benchmark rate becomes $2.85 per subscriber per month.

[96] Dr. Chipty then adjusts this rate for the relative value of programming and simultaneous substitution. She uses the difference in the viewership minutes of her benchmark group and distant signals to make the adjustment. These viewership minutes are estimated using the set-top box data from Rogers, Bell, and Shaw for selected programs in Toronto, Montreal, and Vancouver during May 4, 2015 to May 17, 2015. She re-weights these viewership minutes so that her split of U.S. local and distant viewing in Toronto and Montreal matches that of the Numeris data. The adjustment for simultaneous substitution is also based on set-top box data. [57]

[97] After adjusting for the relative value of programming and simultaneous substitution, she obtains a rate of $1.20 per subscriber per month. Additionally, as part of a sensitivity analysis, Dr. Chipty uses the 57.6 per cent PVR penetration rate as an extra measure to account for substitution opportunities of distant signal programs. This adjustment yields a rate of $0.87 per subscriber per month.

[98] As such, Dr. Chipty argues that the rate should be between $0.87 and $1.20; she recommends a rate of $1.00 per subscriber per month.

[99] Details of Dr. Chipty’s analysis and Collectives’ responses are reviewed in the economic-analysis section of this decision.

ii. Debra McLaughlin [58]

[100] Ms. McLaughlin, of Strategic Inc., analyzes consumer behaviour and attitudes related to the viewing of television programming in general and distant signals in particular.

[101] Ms. McLaughlin explained the difference between Diary data and PPM data as follows: Numeris produces television reports on 37 Diary Markets and 6 PPM Markets. [59] For large markets like Ottawa, Kitchener-London, Winnipeg and Halifax; mid-size markets like Saskatoon, Regina, Thunder Bay; and small markets like Kenora, Dawson Creek, Kingston, and Peterborough, diaries are used as reliable data for economic activities worth of hundreds of millions of dollars annually. While PPM includes more records in a one-year timeframe, Diary has a more robust weekly information. Finally, both PPM and Diary fail at capturing a sizeable sample among the young, new Canadians, and people who have a mother tongue other than English or French.

[102] According to Ms. McLaughlin, both PPM and Diary approaches to examining tuning of distant signals show a reduction of tuning of distant signals. As measured by Diary, tuning drops from 59 million hours of distant signals in 2009 to 49 million in 2014. As measured by PPM for Anglophones in Toronto, Vancouver and Montreal, tuning drops from 17 million hours tuned to distant signals to 14 million hours over the same period.

[103] Ms. McLaughlin also draws attention to a downward trend in the viewing of distant signals between 2009 and 2014 in both Diary and PPM data, a drop of about 11 per cent in 2012 and 18 per cent in 2014, relative to 2009. The decline is even more pronounced among viewers aged 18 to 34 years, where viewing is 33 per cent less in 2014 than in 2009.

[104] By analyzing Numeris data, Ms. McLaughlin found that BDU penetration peaked at 95 per cent of Canadian households in 2012 and declined to 92 per cent in 2014. Ms. McLaughlin’s analysis of PVR penetration also shows that, in most markets, [60] penetration rose from less than 10 per cent in 2009 to 50 per cent or better in 2014.

[105] Using Numeris data, Ms. McLaughlin estimates that live tuning had dropped from 95.2 per cent of total tuning in 2009 to 86.8 per cent in 2014, and that the shift is even greater among adults aged 25 to 54 such that it dropped from 94.1 per cent to 83.8 per cent.

[106] According to Ms. McLaughlin, new technologies have disrupted the trends in how and when people consume television programming. She explained that while once the ability to time-shift programming increased consumers’ ability to schedule their tuning to “must-see” TV, new technologies such as PVR, streaming from online sources and the on-demand programming feature of most BDUs allow consumers to view their program at any time they want, and that this has lessened both the appeal and use of distant signals.

[107] By reviewing the results of the 2014-2105 Media Technology Monitor (MTM) study, Ms. McLaughlin looked at the prevalence of “cord cutting” among Canadians. Cord cutters are those who cancel their BDU subscriptions and obtain television programming from other sources. According to Ms. McLaughlin, slightly more than half of respondents who were not BDU subscribers were cord cutters, and that among those, 44 per cent had been without a BDU subscription for more than three years.

[108] As Ms. McLaughlin explains, the MTM study suggests that the number of cord cutters will continue to grow: among current BDU subscribers, eight per cent said it was very likely they would cancel their service in the next year while 13 per cent said they were somewhat likely to do so.

[109] Ms. McLaughlin also shows that these trends away from traditional forms of broadcasting, including distant signals, and towards newer forms of on-demand technology are confirmed by the study Ekos Research Associates conducted on Canadians ages 18+ to develop an understanding of how people consume television. [61]

[110] Ms. McLaughlin argues that these consumer attitudes and behaviours explain the steady decline in viewing of distant signals. As Ms. McLaughlin explains, Canadians, and in particular adults under the age of 34, are increasingly turning to newer, more convenient sources of television programming which are displacing the need for, and thus reducing the value of, distant signals.

[111] Drawing on Ms. McLaughlin’s report, the BDUs argue [62] that there is a clear trend away from live viewing of prime-time programs and toward playback viewing of those programs between 2009 and 2014. [63] “Playback viewing” refers to prime-time programs being viewed later on a specialty channel or a pay-tv service. It does not require recording of the programs as such. As Numeris data show, conventional television has lost market share to specialty and pay television in Canada since 2005.

[112] The BDUs further argue [64] that, since Ms. McLaughlin’s study confirms that viewing of distant-signal programming has been continuingly decreasing for at least the past decade, an increase in the rate is unwarranted. On the contrary, the rate should decrease to reflect the decline in value.

[113] In response, the Collectives argue that Ms. McLaughlin uses data that are inaccurate and incomplete. [65] The Collectives elaborate on their argument by explaining that while Ms. McLaughlin uses Numeris Diary data for a large portion of her analysis on distant-signal viewing, Numeris PPM data, which is the “gold standard”, is available nationally, and there are enough respondents to also permit individual market analyses in the largest markets. The Collectives point to Mr. Kiefl’s testimony, where he explained that Diary data have more shortcomings than PPM data, and that PPM data are used by the entire broadcasting industry and the CRTC. [66] According to the Collectives, [67] there are significant differences between the Diary and PPM data; these differences lead to different results.

[114] While based on Diary data, Ms. McLaughlin reports a notable drop in general television viewing from 2006 to 2014. [68] Mr. Kiefl explains that the bulk of the reported drop occurred in a single year (2012) when a methodological change affected the data. [69]

[115] Mr. Kiefl refutes Ms. McLaughlin’s assertion of a decrease in live viewing. Ms. McLaughlin reports that Numeris PPM data showed “live” viewing had declined to below 90 per cent. Mr. Kiefl points to the weekly update [70] of the Numeris PPM data, which shows that live viewing remains at over 92 per cent in 2013-2014 and 2015. [71] Mr. Kiefl further indicates that TV viewing levels changed little in the past decade according to PPM data from Numeris, except among children and teens, which declined by 10-15 per cent since 2009. Based on the Diary data, Mr. Kiefl also concludes that the distant signal viewing share as a percentage of all TV viewing remained unchanged over the past decade. [72]

[116] As Mr. Kiefl argues, Ms. McLaughlin’s conclusion that there is a decline in penetration rates is also inconsistent with data reported by other sources, including the TVB and Numeris itself, which both show BDU penetration remaining roughly constant. Mr. Kiefl also points out that while Ms. McLaughlin reports a recent decline in DTH and cable penetration based on Numeris data, industry data published by the CRTC show no substantial change in the number of cable/satellite subscribers. He further explains that the data shows that only growth has slowed but few people are cutting the cord.

[117] Using the data from the Mediastats study, the Collectives state that the average, residential, retail price for BDUs’ basic packages almost doubled from 2004 to 2014. Combining this fact with Ms. McLaughlin’s report of a decline in overall viewing, [73] the Collectives argue that the market value of all viewing is rising despite reported declines in viewers’ consumption.

[118] Regarding the PVR, the Collectives argue that Ms. McLaughlin only considered reported PVR usage in selected markets, for only part of the year and at only limited times of day, such that Ms. McLaughlin only examined PVR usage during nine per cent of actual broadcast hours.

[119] To elaborate on this comment, the Collectives note that Ms. McLaughlin ignores the full-year, national data available from Numeris. They draw on Mr. Kiefl’s testimony, where he shows that, contrary to Ms. McLaughlin’s conclusion based on restricted data, using the full data shows that more than 92.3 per cent of all viewing is live viewing. [74] Moreover, the Collectives state that Ms. McLaughlin seems to assume that PVR viewing is an alternative to distant signal viewing. The Collectives disagree: a distant signal recorded by a PVR remains distant when viewed from a PVR. The Collectives also state that Ms. McLaughlin provides no analysis or quantification of what amount of the 7.7 per cent of PVR viewing is to distant signals. [75] Finally, the Collectives state that Ms. McLaughlin does not address the issue that certain “must-see” types of distant signal programming are not recorded for later viewing. [76]

[120] The BDUs draw attention to Mr. Kiefl’s 2004 viewing study, which determined that “there had been, as of 2004, a pretty significant decline in the amount of distant signal viewing from the mid-’90 period to the mid-2000 period”. [77]

[121] The BDUs state that Mr. Kiefl’s earlier report found that distant signal viewing accounted for 14 per cent of all English TV viewing in 1992 and just 10.7 per cent of viewing in 2004. [78] In other words, the share of viewing captured by distant signal programming decreased by almost 25 per cent in the 12 years following the Board’s second retransmission decision. The BDUs state that in his earlier report, Mr. Kiefl concluded that: “Despite the fact that more distant signals were available to cable and especially DTH subscribers in 2004, the relative audience of distant signals has declined significantly.” [79]

[122] The BDUs also state that during his testimony about the earlier report, Mr. Kiefl explained that the number of other kinds of programming could affect the decline in distant signal viewing. [80]

[123] Another issue the Collectives raised [81] with respect to Ms. McLaughlin’s report was that distant signal viewing recorded only from Toronto, Montreal, and Vancouver is not nationally representative. According to the Collectives, by segregating certain viewing data between the three biggest markets and all other markets, Ms. McLaughlin reveals that the relative viewing of Canadian and U.S. distant signals varies significantly. For example, her data shows that that while in the big three markets Canadian distant signal viewing was about 170 per cent higher than that of U.S. distant signals in 2014, it was only about 27 per cent higher in the rest of Canada. [82] The Collectives also argue that, by listing all of the stations identified by Mediastats to be local in each of the 37 Canadian markets, Ms. McLaughlin’s report shows that the big three markets not only have uniquely high numbers of local U.S. stations available (because of their proximity to the U.S. border), but because of their size, they also have higher numbers of Canadian local signals, which result in the share of U.S. and Canadian distant signal viewing being much lower in these markets compared to the rest of Canada. [83]

[124] Finally, the Collectives point out that Ms. McLaughlin’s report confirms the importance of time-shifted signals to BDU subscribers by indicating that 32 per cent of respondents said they would consider switching BDUs if their current provider stopped providing time-shifted signals. [84]

[125] The viewing data and the points the parties and their experts raised will be addressed later in this decision.

iii. Lori Assheton-Smith [85]

[126] Ms. Assheton-Smith prepared a report on the effect of broadcasting regulations on the retransmission of distant signals in Canada.

[127] By reviewing the CRTC’s regulations of distant signal retransmission from the early days of cable television in Canada, she identifies a connection between the regulation of distant signals and the carriage of distant signals by BDUs.

[128] Ms. Assheton-Smith concludes that, although the increase in the average number of distant signals carried by BDUs from 2004 to 2014 may partially reflect the value to consumers of time-shifting, particularly in the early part of this period, the overall increase in the number of distant signals carried by BDUs over the years reflects, in many cases, the encouragement or requirement of the CRTC to expand distant and local signal offerings, due to various broadcasting policy objectives.

[129] As Ms. Assheton-Smith explains, some significant regulatory developments that influenced the carriage of distant signals include: the licensing and launch of the DTH satellite providers in 1995; the transition from analog to digital technology around 2000; the launch of time-shifted distant signals in 2002; the CRTC’s order that the DTH BDUs retransmit the signals of an additional thirteen small market stations after 2003; the CRTC’s order that DTH providers increase their basic service to include signals from the large private broadcasters, CBC, independent broadcasters, and provincial television services from each province, and that those BDUs carrying a second set of U.S. 4+1 stations carry at least one Canadian distant television signal from each English ownership group within the same time zone in 2008; and the CRTC’s regulatory framework for DTH distribution in 2011.

[130] Ms. Assheton-Smith, however, explains that many of the regulatory incentives that encouraged the increase in the number of distant signals available to a BDU subscriber may no longer be applicable or relevant. As such, she expects a gradual reduction in the carriage of distant signals. Additionally, she also explains that the current regulatory framework, that emphasizes consumer choice and flexibility – for example, the increase in the flexibility of VOD programming undertakings – provides incentives and requirements that will reduce the BDUs’ distant-signal carriage and consumers’ demand for distant signals and time-shifting in the future. [86]

[131] Drawing on Ms. Assheton-Smith’s report, the BDUs argue that while the Collectives focus on the increase in the average number of distant signals being carried, they have not considered the extent to which these increases have been in direct response to CRTC’s regulatory policies. According to the BDUs, the number of distant signals that are retransmitted has more to do with addressing the concerns of Canadian broadcasters and CRTC’s regulations than it does with consumer demand for multiple signals from the same network. [87]

[132] The Collectives refute Ms. Assheton-Smith’s testimony as follows. They note that she confirms that the CRTC does not require BDUs to carry any distant signals except for the mandatory carriage of TVA, [88] and that BDUs are not required by regulation to package and sell all of the distant signals they carry. Also, DTH BDUs are only required to carry three signals nationally (CBC, SRC and CTV), and within a province a limited set of up to 13 additional signals that may be distant in parts of the province. [89]

[133] The Collectives state that BDUs, however, choose to sell over 55 distant signals, on average, to subscribers. [90] The Collectives argue that this indicates that as profit-maximizing, rational economic actors, BDUs have concluded that they are better off by selling distant signals to their subscribers than not doing so.

[134] With respect to the Let’s Talk TV decision, the Collectives also highlight that Ms. Assheton-Smith acknowledges that she did not know if the CRTC decision would result in any reduction of distant-signal carriage, was unaware of any CRTC study suggesting such a thing, and “would not want to hazard a guess” as to the impact of the decision. [91] According to the Collectives, Ms. Assheton-Smith acknowledges that the impact of Let’s Talk TV was “difficult to predict at this time” and that the “timing and effect of [regulatory change] is difficult to assess at this time.” [92]

[135] In response, the BDUs state that the Collectives’ claim [93] that a “BDU witness confirmed, for example, that (except for mandatory carriage of TVA), the CRTC does not require BDUs to carry any distant signals” is a completely false statement. According to the BDUs, [94] the portion of the transcripts cited by the Collectives in support of this assertion only deals with “cable” systems, not all BDUs as claimed by the Collectives in their statement. The BDUs note that contrary to the Collectives’ statement, Ms. Assheton-Smith actually testified that DTH BDUs are required by regulation to carry distant signals and to distribute those signals as distant signals to subscribers that do not have local signals. The BDUs further argue that even with respect to just the cable systems, the Collectives have misstated Ms. Assheton-Smith’s evidence, since she testified that there are circumstances in which a cable BDU is required by CRTC regulations to carry and distribute a distant signal in addition to TVA. According to the BDUs, under these circumstances, BDUs have no choice but to pay the royalties approved by the Board because they are required by the CRTC to distribute distant signals to subscribers. [95]

iv. Suzanne Blackwell [96]

[136] Ms. Blackwell, of Giganomics Consulting Inc., prepared an overview of the Canadian broadcasting industry focusing on the competitive environment in which BDUs operate, the factors that drive BDUs’ basic-service price increases, and the financial performance of over-the-air and specialty broadcasters over the past decade or so.

[137] Ms. Blackwell argues that there is no basis for increasing the distant signal royalty rate based on trends in the broadcasting industry, for two reasons. First, broadcasters of television programming experienced no increase in program spending, revenue or wholesale rate for a core group of non-sports specialty services, no increase in program spending or revenue for conventional OTA services since 2011, and weak financial performance of conventional OTA services relative to specialty services. Second, there has been increased competition among BDUs for fewer customers; an increased consumer adoption of broadband Internet and OTT video services; and an increase in the retail price of basic BDU service.

[138] After reviewing the detailed financial information collected by the CRTC from OTA broadcasters and specialty services, Ms. Blackwell observed that the core group of non-sports specialty services had not increased program spending since 2008 on a per-service basis, and that these same services had not increased their revenue from subscribers when measured on a per-service basis, net of inflation.

[139] According to Ms. Blackwell, if the specialty-programming services that are used as the benchmark for distant signals have not increased either their program expenditures or subscriber revenue on a “per-service basis” since 2008, there is no reason to expect that the value of programming in retransmitted distant signals has increased by 146 per cent during that same period, as claimed by the Collectives.

[140] Ms. Blackwell explains that the comparison with the revenue earned by local OTA signals is even more compelling, considering that distant signals are the same signals as OTA with the same programming. As she notes, between 2004 and 2014, the per-service revenue of OTA signals decreased by 24 per cent in constant-dollar terms. Similarly, OTA services’ program expenses, per service, decreased between 2004 and 2014. If the ability of local broadcasters to generate revenue from their programming fell by 24 per cent between 2004 and 2014, and the amount spent on that programming also declined, Ms. Blackwell argues that it is unreasonable to suggest that the value of the same programming when retransmitted by BDUs would increase by 174 per cent over that same time period.

[141] Ms. Blackwell pointed out that the percentage of Canadian households that subscribe to a BDU has decreased from 85.8 per cent in 2011 to 83.3 per cent in 2014, while the number of residential high-speed subscribers increased from just two million subscribers in 2000 to more than 11 million in 2015. According to Ms. Blackwell, these statistics demonstrate the increase in “cord cutting” and “cord shaving”, where BDU subscribers cancel or reduce their subscriptions and use online sources instead to access the same programming.

[142] An example Ms. Blackwell provides to corroborate her argument is Netflix. While Netflix only launched in Canada in 2010, it grew to more than one million subscribers in 2011 and almost four million subscribers in 2015. She also points out to other online sources of programming such as Apple, YouTube, Google, Vimeo and Canadian services such as Illico, Shomi and CraveTV.

[143] Ms. Blackwell concluded that the significant increases in distant signal royalty rates proposed by the Collectives are out of line with the trends she observed among the conventional and specialty television industry. [97]

[144] In response, the Collectives argue the BDUs’ conclusion based on the OTA television industry analysed by Ms. Blackwell is incorrect, because unlike distant signals, OTA services earn revenue from advertising, and hence, they do not exhibit the same market structure. [98]

[145] The Collectives also argue that Ms. Blackwell’s proxy contradicts Dr. Chipty’s proxy group. The Collectives specifically point out that Ms. Blackwell states that Category A and C services can be used as a proxy for the price of distant signals; and point out the price of these services has increased significantly over time. The Collectives state that Ms. Blackwell chooses to exclude Category C sports services because of the increase in their prices even though she knew that major-league sports programming is available on distant signals. [99] The Collectives state that by excluding Category C sports services, Ms. Blackwell was able to change an upward-trending price line artificially and arbitrarily into a flat or downward-trending price line. [100]

v. BDU Industry Witnesses

[146] The BDUs called five industry witnesses from Rogers, Shaw, Bell, Vidéotron and TELUS. These witnesses explained the state of competition in the industry, recent changes and trends in viewership of distant signals, and the impact of an increase in tariffs on the TV broadcasting industry. Most stated that the industry faces significant competition, that distant signals are losing their values and viewership, and that, in the event of an increase in tariffs, BDUs would either remove distant signals from their basic offering or transfer the costs to consumers. [101]

vi. David Purdy, Rogers Communications [102]

[147] According to Mr. Purdy, Senior Vice-President of Content for Rogers Communications (Rogers), the level of competition among BDUs is very significant, as the number of competitors in the regulated system increases while the number of television subscribers is shrinking. Mr. Purdy also explained that both the number of alternative viewing platforms/sources available to subscribers and actual viewing of these alternative sources would increase.

[148] Mr. Purdy pointed out that the value of live, linear channels of local and distant signals declined significantly, while the value Rogers’ customers associate with on-demand viewing increased both in usage and in “overall perceptions of value”. [103]

[149] He also explained that Rogers’ customers can now access distant-signal programming, past and present episodes, in a number of ways such as: [104] live on linear television channels containing local or distant signals; live over-the-air using a digital antenna; pirated online; on a BDU or broadcaster’s mobile application; streaming on major broadcasters’ websites such as CTV.ca, both live and on-demand; the Rogers free on-demand service included in basic cable on the set-top box which includes all Canadian broadcasters’ programming; subscription VOD available on the set-top box or online such as Shomi and Netflix; and, on a transactional rental basis. [105] In addition to these services, digital-cable subscribers, who are the vast majority of Rogers’ subscribers, receive PVR capability in the set top box, which allows them to record programs live or in advance, and on up to eight different channels simultaneously. [106] As Mr. Purdy explained many of these alternatives were not available when Rogers launched time-shifted distant signals. [107]

[150] Mr. Purdy stated that the Collectives’ proposed increasing royalty rates were entirely inconsistent with the trends in the cost of other programming he observed at Rogers. [108]

[151] With respect to vertical integration of Category B specialty services and BDUs, Mr. Purdy did not agree with the assertion of the Collectives that the price of these services would be different from the market price. [109]

vii. Geoff Wright, Bell [110]

[152] Mr. Wright explained that, besides its TV packages, Bell also offers a variety of VOD services to its subscribers, [111] namely Bell’s proprietary subscription-VOD platform, CraveTV, available on the set-top box to Bell’s BDU subscribers and online to all; [112] a VOD “storefront” on the set-top box where a library of VOD programming is accessible; [113] the ability to record live programs after they have aired (the Restart and Look Back functions) [114] and while they air for later viewing (the PVR); [115] free VOD on the set top box of programming from Canadian broadcasters and specialty service providers including “catch up rights”, [116] and transactional VOD, like a movie rental store where you can also get television shows on an episode by episode basis. [117] The majority of these services are at no extra cost to Bell customers. [118]

[153] Mr. Wright also testified that while there was an increase in the number of both specialty services and distant signals offered by Bell since the mid-1990s, there has also been a shift from conventional viewing to specialty viewing, and that growth in specialty services viewing has outstripped conventional viewing. [119]

[154] According to Mr. Wright, subscribers’ demand for both Canadian distant signals and U.S. 4+1 signals declined over the last three years. This decrease can be attributed to the opportunity to use alternative viewing platforms such as PVRs, on-demand services, and over-the-top providers of on-demand content such as Netflix, Shomi and CraveTV. [120]

[155] He also asserted that the Canadian BDU market is highly competitive. Mr. Wright disagreed with the Collectives that the price of vertically integrated Category B specialty services would be different from the market price.

viii. Gary Pizante, Shaw Communications [121]

[156] Mr. Pizante testified that the value of distant signals to Shaw is declining, [122] and there is no growing demand for distant signals. He attributed this change to the increased availability of other content sources. As he explained, customers are looking for convenience, ease of use and technology that goes well beyond what the time-shifting packages offer. [123]

ix. Ann Mainville-Neeson, TELUS [124]

[157] Ms. Mainville-Neeson testified that the Collectives’ proposed rates do not reflect the value of distant signal programming to TELUS, and that TELUS considers the television distribution market to be highly competitive. [125]

x. Marie Ginette Lepage, Vidéotron G.P. [126]

[158] Ms. Lepage testified that specialty services are the most in-demand by customers, [127] and that in the last five years, Vidéotron has not offered additional distant signals to subscribers. [128] Also, she explained that the number of subscribers subscribing to time-shifted signals, whether Canadian or US, decreased in 2015. [129] According to her, there is a decrease in overall subscribers to Vidéotron’s BDU services and further decreases are anticipated. [130] She also stated that the Canadian BDU environment is currently highly competitive.

[159] With respect to these witness statements, the Collectives state that Dr. Chipty does not rely on any of them in formulating her proposed royalty rate, the statements are not capable of supporting BDUs’ proposals, and that the proposed evidence from BDU managers is speculative, alarmist, self-serving and inconsistent with the actual behaviour of subscribers and BDUs themselves. [131] The Collectives state that despite the testimony of BDU managers, BDUs’ own surveys commissioned for the Let’s Talk TV proceedings demonstrate the importance of distant signals to their customer base. [132]

[160] Relevant arguments of the parties will be addressed later in this decision.

VI. LEGAL ANALYSIS

A. The Copyright Act’s legal framework with respect to the retransmission regime

[161] Retransmitters, in this case referred to as BDUs, are required to pay the retransmission tariff to compensate for the communication to the public by telecommunication of the copyright-protected works carried by distant signals. These works include the programs carried by the signals and the “broadcast day” compilations created by the broadcasters. As a matter of law, the BDUs are not required to compensate broadcasters for the retransmission of the signals themselves.

[162] The retransmission right contained in section 31 of the Act was enacted in 1988 pursuant to the Canada-United States Free Trade Agreement Implementation Act to give effect to Article 2006 of the Canada-United States Free Trade Agreement. [133] Article 2006 required that:

1. Each Party’s copyright law shall provide a copyright holder of the other Party with a right of equitable and non-discriminatory remuneration for any retransmission to the public of the copyright holder’s program where the original transmission of the program is carried in distant signals intended for free, over-the-air reception by the general public […] [134]

[163] The Supreme Court of Canada has stated that the Copyright Act’s objectives − of “encouraging creativity and providing reasonable access to the fruits of creative labour” − are furthered through “a carefully balanced scheme that creates exclusive economic rights for different categories of creators” while giving due weight to the limited nature of these rights through specific exemptions. [135] Section 31 of the Act maintains this careful balance by creating a class of users’ rights applicable to BDUs in conjunction with a compulsory licence regime that compensates the owners of copyright for use of their works. [136]

[164] The copyright owner’s right to communicate the work to the public by telecommunication pursuant to paragraph 3(1)(f) of the Act captures the activity of retransmission. [137] The Supreme Court of Canada, in Reference re Broadcasting Regulatory Policy CRTC 2010-167, stated that subsection 31(2) is directed at narrowing the scope of the owner’s right under paragraph 3(1)(f), by “circumscrib[ing] the right of copyright owners to control the retransmission of literary, dramatic, musical or artistic works carried in signals.” [138] Subsection 31(2) of the Act entitles BDUs to simultaneously retransmit the literary, dramatic, musical or artistic works carried in a local signal, without authorization by or payment to the copyright owner, and in the case of distant signals, allows simultaneous retransmission of the works contained in the signals subject to the payment of royalties. [139]

[165] Thus, subsection 31(2) of the Act does not create an exception to the rights associated with the signals themselves, nor a corresponding right to receive royalty payments with respect to the retransmission of the signals themselves. The copyright held by broadcasters in communication signals under section 21 of the Act is limited and does not create the right to authorize or prohibit the retransmission of communication signals by BDUs. [140]

[166] The Supreme Court of Canada has distinguished the respective scope and function of sections 21 and 31 of the Act by noting that the “Copyright Act seeks to regulate the economic rights in communication signals, as well as the retransmission of works by BDUs.” [141] As such, the retransmission regime only engages the rights of broadcasters in their capacity as owners of copyright in the works contained in distant signals. [142]

B. Legal Issues to be considered

[167] Through its Notice CB-CDA 2016-011 dated February 5, 2016, the Board asked the parties to address three issues in their respective legal briefs:

1. Whether the Retransmission Royalty Criteria Regulations (SOR/91-690) remain in force, and, if so, how should the Board deal with s. 2 of the Retransmission Royalties Criteria Regulations?
2. Whether the principle of non ultra petita finds any application in this case, and whether the change in proposed royalty rates raises issues of procedural fairness, particularly for non-participants who may nonetheless be affected by the tariff. In the event procedural fairness is an issue, how can the Board proceed to remedy the situation?
3. Whether the Board should account for Broadcasting Regulatory Policy CRTC 2015-96 set to come into force in March 2016 and if so, how?

[168] We turn to those issues now.

C. The Retransmission Regulations

[169] The Retransmission Royalties Criteria Regulations were adopted by the Governor in Council on November 28, 1991, pursuant to the then s. 70.63(4) of the Act. Section 2 of these Regulations requires the Board to take into account three particular criteria in determining royalties that are fair and equitable in the retransmission regime. Those mandatory criteria are the following: (i) royalties paid for the retransmission of distant signals in the United States; (ii) the effects of the application of the Broadcasting Act and its regulations on the retransmission of distant signals in Canada; and (iii) the terms and conditions of existing agreements respecting retransmission royalties in Canada.

[170] In 1997, however, the Act was amended. Subsection 70.63(1) was replaced by the new subsection 73(1) which removed the reference for the Board to have regard to criteria established under subsection 70.63(4) in determining retransmission royalties. At the same time, subsection 70.63(4) was repealed and section 66.91 was added so as to continue to provide the Governor in Council with the power to make regulations establishing general criteria to which the Board “must have regard” in establishing fair and equitable royalties to be paid pursuant to the Act or general criteria to be applied by the Board in doing so.

[171] The Collectives argue that the Retransmission Regulations are no longer in effect, as they are inconsistent with the current legislation, focusing on the removal of the mandatory language in s. 73(1). The BDUs take the opposite position, relying on the continuing regulation-making power in s. 66.91.

[172] For the reasons that follow, despite the seeming ambiguity in the effect of the legislative changes in question, we are of the view that the continuing operation of the Retransmission Regulations is not inconsistent with the new enactment when read as a whole. Even if we were to accept that the Retransmission Regulations were no longer in effect, it is well-established that the Board has a broad discretion to take into account factors that it deems relevant and appropriate in establishing fair and equitable royalties and tariffs in any event.

[173] Our analysis of these issues follows.

i. The Interpretation of the Statute and Regulations

[174] Section 2 of the Retransmission Regulations sets out the three required considerations. It provides that:

[t]he criteria to which the Board must have regard in establishing under paragraph 70.63(1)(a) of the Copyright Act a manner of determining royalties that are fair and equitable are the following:

(a) royalties paid for the retransmission of distant signals in the United States under the retransmission regime in the United States;

(b) the effects on the retransmission of distant signals in Canada of the application of the Broadcasting Act and regulations made thereunder; and

(c) royalties and related terms and conditions stipulated in written agreements in respect of royalties for the retransmission of distant signals in Canada that have been reached between collecting bodies and retransmitters and that are submitted to the Board in their entirety.

[175] Former s. 70.63(1) of the Act, under which the Retransmission Regulations were enacted, and the former s. 70.63(4) which set out the regulation-making power of the Governor in Council at the time, read as follows:

Certification

70.63(1) On the conclusion of its consideration of the statements of royalties, the Board shall

(a) establish, having regard amongst others to the criteria established under subsection (4),

(i) a manner of determining the amount of the royalties to be paid by each class of retransmitter, and

(ii) such terms and conditions related to those royalties as the Board considers appropriate;

(b) determine what portion of the royalties referred to in paragraph (a) is to be paid to each collecting body;

(c) vary the statement accordingly; and

(d) certify the statements as the approved statements, whereupon those statements become for the purposes of this Act the approved statements. [underlining added]

[…]

Criteria

70.63(4) The Governor in Council may make regulations establishing criteria to which the Board must have regard in establishing under paragraph (1)( a) a manner for determining royalties that are fair and equitable. [underlining added]

[176] In 1997, section 70.63 of the Act was repealed as part of the Bill C-32 reform of the Act. Its functional replacement, s. 73, does not require that the Board “have regard” to any particular prescribed criteria when establishing “a manner of determining the royalties to be paid […] by retransmitters within the meaning of subsection 31(1).” At the same time, however, the 1997 amendments also enacted s. 66.91, a substituted regulatory-making power for the Governor in Council, which continued the Governor in Council’s authority to establish general criteria to which the Board must have regard. The relevant portions of those sections state:

73(1) On the conclusion of its consideration of proposed tariffs, the Board shall

(a) establish

(i) a manner of determining the royalties to be paid by educational institutions and by retransmitters within the meaning of subsection 31(1), and

(ii) such terms and conditions related to those royalties as the Board considers appropriate;

66.91 The Governor in Council may make regulations issuing policy directions to the Board and establishing general criteria to be applied by the Board or to which the Board must have regard

(a) in establishing fair and equitable royalties to be paid pursuant to this Act; and

(b) in rendering its decisions in any matter within its jurisdiction.

[177] In the legislative summary of Bill C-32 by the Library of Parliament, section 73 of the Act was described as a “modified version” of the former section 70.63. [143]

[178] Given these changes, we must now assess whether the Retransmission Regulations are still in force. We take into account the following provisions of the Interpretation Act in doing so:

2 (1) In this Act,

enactment means an Act or regulation or any portion of an Act or regulation;

3 (1) Every provision of this Act applies, unless a contrary intention appears, to every enactment, whether enacted before or after the commencement of this Act.

44. Where an enactment, in this section called the “former enactment”, is repealed and another enactment, in this section called the “new enactment”, is substituted therefor,

(g) all regulations made under the repealed enactment remain in force and are deemed to have been made under the new enactment, in so far as they are not inconsistent with the new enactment, until they are repealed or others made in their stead. [underlining added]

[179] Further to the Board’s Notice CB-CDA 2016-011 of February 5, 2016, the parties provided their submissions respecting the current status of, and the applicability of, the Retransmission Regulations. We address those now.

ii. Position of the Parties Respecting the Status of the Retransmission Regulations

[180] The Collectives argue that the requirement that the Board consider certain criteria was deleted in the 1997 amendments and submit that all that is required is that the Board establish a manner of determining the royalties to be paid by Retransmitters. The Collectives proffer language that could have been used by Parliament had it intended to continue to require the Board to have regard to the regulatory criteria.

[181] The BDUs submit that the Retransmission Regulations were not repealed at the time of the 1997 amendments. As such, pursuant to paragraph 44(g) of the Interpretation Act, the BDUs contend that the Retransmission Regulations are not inconsistent with the new enactment and therefore remain in force and are deemed to have been made under section 66.91 of the Act. [144]

[182] The BDUs further submit that only minor modifications were made to sections 66.91 and 73 of the Act and that, while the Governor-in-Council’s regulation-making power was arguably expanded, the Retransmission Regulations remain compatible and consistent with the rest of the amended Act and continue to apply within Canadian law. In support of this proposition, the BDUs refer to the decisions in R. v. National Grocers Co. Ltd. [145] as well as R. v. Parrott. [146]

iii. Analysis Respecting the Status of the Retransmission Regulations

[183] While we do not view the 1997 modifications to the Act as “minor”, we do agree with the BDUs that the Retransmission Regulations made under the now repealed s. 70.63 are not inconsistent with the new ss. 73(1) and 66.91, which must be read together and interpreted in a manner that gives the new enactment a large and liberal, and purposive, meaning. [147]

[184] It may appear somewhat incongruous that, in the general provision setting out its directives to the Board about establishing the manner of determining the royalties and their terms and conditions, Parliament would remove a requirement that the Board have regard to certain particular criteria set out in a regulation, but at the same time leave that very regulation – which also governs the Board’s deliberations – in place. That is why we observe that this interpretation is not free from doubt. Even so, as long as that regulation has not been repealed, it “remain[s] in force and [is] deemed to have been made under the new enactment, in so far as [it] is not inconsistent with the new enactment”: Interpretation Act, s. 44(g). The Governor in Council has chosen not to repeal the Retransmission Regulations. We cannot conclude they are inconsistent with the new enactment, having regard to the scheme of the Act, the retransmission regime that is set out therein, and the requirement that the Board set retransmission royalties that are fair and equitable.

[185] It makes sense to us that the Board should have regard to the criteria set out in the Retransmission Regulations, and even if not required to do so, could, and in appropriate cases would, take them into account in any event. Their application is therefore in keeping with the purpose of the Act.

[186] Although the retransmission regimes in Canada and the United States are somewhat different, information regarding retransmission royalties in the United States could, depending on the circumstances, be helpful to the Board in its deliberations. As the Collectives observe, in their submissions summarized more fully below, the Regulatory Impact Analysis Statement, SOR/91-690 prepared by the Government in relation to the Retransmission Regulations when they were enacted in 1991, state that

[t]he criteria do not direct the Board to a particular conclusion. The Board may, in its discretion, attribute such weight as it sees fit to each criterion, to the criteria in their entirety and to the evidence and the arguments of the parties.

[187] We make the same observations with respect to the effect of the application of the Broadcasting Act and its regulations, and the relevance of existing agreements between the Collectives and the BDUs regarding retransmission royalties.

[188] It may be said that, by removing the mandatory requirement to consider particular criteria in the newly enacted s. 73(1), Parliament sent a signal that it intended regulation-making powers, or at least going-forward regulation-making powers, would no longer provide for such requirements. By enacting a new regulation-making power in s. 66.91 that continues to provide for the ability to set requirements for which “the Board must have regard”, however, Parliament also signaled a contrary intention. Given the Governor in Council’s power to adopt a regulation setting out mandatory criteria, was it Parliament’s intention that the Governor in Council should repeal the existing Retransmission Regulations and replace them with a new regulation to the same effect? Or was it Parliament’s intent that the Governor in Council simply leave the existing Regulations in place if it chose to do so? The Governor in Council appears to have chosen to do the latter.

[189] For the foregoing reasons, we conclude that the Retransmission Regulations are “not inconsistent with the new enactment”, have not been “repealed or others made in their stead”, and therefore “remain in force and are deemed to have been made under the new enactment”: Interpretation Act, s. 44(g).

[190] Having determined that the Retransmission Regulations remain in effect, we now turn our mind to the application of the three criteria in the present proceeding.

iv. The Collectives’ Argument on the Application of the Retransmission Regulations

[191] The Collectives submit that even if the Board must have regard to the considerations enumerated in the Retransmission Regulations, these considerations are not exclusive, and – moreover – should not be given much weight.

[192] First, as noted above, the Collectives point to the Regulatory Impact Analysis Statement, SOR/91-690 prepared by the Government in relation to the Retransmission Regulations:

The criteria do not direct the Board to a particular conclusion. The Board may, in its discretion, attribute such weight as it sees fit to each criterion, to the criteria in their entirety and to the evidence and the arguments of the parties. Consideration of the criteria will take place in a manner that allows each party to conduct its own case and present its evidence as it sees fit. Finally, the criteria are flexible enough to endure over time.

[193] Second, the Collectives submit that this understanding was also obvious on the face of the now repealed section 70.63(1)(a) of the Act, which instructed the Board to have regard “amongst others” to the regulatory criteria. Indeed, in its 1993 Decision, the Board found that the adoption of the Retransmission Regulations was not an attempt to override the six guiding principles set out in its 1990 Decision. [148]

[194] Thus, according to the Collectives, since the parties have filed extensive economic evidence establishing the value of distant signals in the Canadian marketplace, which in their view is of core importance, and since all parties have clearly focused their submissions on it, the Board should give this evidence the most weight; the three factors set out in the Retransmission Regulations are of little or no importance.

[195] In support of this last assertion, the Collectives submit that: [149]

• The fact that neither party provided sufficient evidence of the U.S. retransmissions regime indicates that this factor has a diminishing importance in the Canadian retransmission setting;
• The regulatory evidence relating to the Broadcasting Act is, at best, inconclusive, and – in any case – is already integrated in the tariff structure; and
• Any agreements considered pursuant to this factor must relate to the tariff period under consideration, and – in any case – no written agreements stipulating royalties and related terms and conditions have been submitted to the Board in their entirety.

v. The BDUs Argument on Application

[196] The BDUs submit, in relation to the first factor, that, in the absence of any evidence from the parties, it is not reasonable for the Board to attempt to make any adjustments to the Canadian rates to reflect rates charged in the United States. Neither party urges us to give effect to the U.S. retransmission-rates criterion.

[197] In relation to the second factor, the BDUs submit various ways how the effects of the application of the Broadcasting Act may be considered, including the facts that:

• The increase in the number of signals carried since 1990 can be traced to the CRTC’s decisions and regulations;

• There is a downward trend in the average number of signals carried; and

• The Broadcasting Act treats standard definition and high definition versions of the same channel as one signal, as opposed to two separate signals.

[198] In relation to the last factor, the BDUs submit that the appropriate rate for the years 2014-2018 continues to be the rate that the BDUs, the Retransmitters and the Collectives agreed upon for the year 2013. The BDUs state that the written agreement to jointly propose this rate and its related terms and conditions to the Board was submitted to the Board in its entirety as required by the Retransmission Regulations.

vi. Analysis Respecting the Application of the Retransmission Regulations in the Present Proceeding

[199] Although we have concluded that the Retransmission Regulations remain in effect, the matter of how the Board should deal with them in this proceeding must still be weighed.

[200] We deal with criteria 2(a) and 2(c) of the Retransmission Regulations at this point in the analysis. Our discussion of the effect of the Broadcasting Act and regulations comes later.

[201] This exercise is complicated by the following. Neither party adduced any evidence with respect to criterion 2(a) (the U.S. retransmission regime). Indeed, both the Collectives and the BDUs urged us not to give weight to this factor. In addition, with respect to criterion 2(c) (written agreements), the BDUs refer to and rely upon a written agreement that was entered into with the Collectives jointly to propose the rate for 2013 with respect to an earlier tariff. That agreement is not filed in this proceeding but, rather, was submitted to the Board in the previous proceeding.

[202] Given these complications, three questions arise in relation to the requirement that the Board “must have regard to” those criteria in establishing the manner of determining royalties that are fair and reasonable in this proceeding: Retransmission Regulations, s. 2. First, is there an inference to be drawn against or in favour of either party as a result of the lack of evidence? Second, is it necessary or appropriate for the Board, itself, to embark upon further inquiries to seek out such evidence? Third, is it open to the Board to consider the written agreement referred to above? In the circumstances here, we answer the first two questions in the negative and the third in the affirmative. The weight to be given to the written agreement is dealt with in the economic analysis portion of these reasons.

[203] Having regard to the criteria, and in responding to these questions, we agree with the Collectives that the Regulatory Impact Analysis Statement, cited above, provides a helpful guide to Parliament’s intention concerning the application of the criteria. The BDUs do not contend to the contrary. It is a matter of assessing the weight the Board “sees fit [to attribute] to each criterion, to the criteria in their entirety and to the evidence and the arguments of the parties.” With that in mind, we conclude as follows.

[204] With respect to the first question, there are situations where a lack of evidence may permit a decision-maker to draw an obvious inference in favour of or against a party. This does not appear to us to be the case here, however. The information that criteria 2(a) and 2(c) require the Board to consider may favour one party, or neither. It is therefore not possible to assign a “burden” in relation to these factors and draw an adverse inference against one party or another.

[205] With respect to the second question, we do not see it as a necessary or fitting aspect of the Board’s role in this proceeding to seek out further evidence in order to supplement the record regarding these two criteria.

[206] All parties have urged us to give little, if any, weight to rates in the U.S. Retransmission regime. We accept the Collectives’ point that the fact no parties have adduced any evidence on that issue but, rather, have all concentrated on the core importance of the Canadian market, is a realistic indication that matters in the US regime are of diminishing importance in the Canadian context. The BDUs as well, urge us not to attempt any such adjustments. In these circumstances, we see no need for the Board to seek out further evidence on criterion 2(a). We have considered this factor, but having regard to the foregoing, give no weight or effect to it for the purposes of this proceeding.

[207] The parties disagree on whether the earlier written agreement on which the BDUs rely – the joint agreement proposing the rate for 2013, filed in a previous proceeding – should be considered. The Collectives take the technical position that the agreement was not filed “in [its] entirety” and therefore does not meet the requirements of criterion 2(c). They argue as well that such agreements must be submitted to the Board in the course of the proceeding in which they are to be considered. We do not find either submission compelling in the circumstances.

[208] In the Board’s view, if another written agreement between Collectives and Retransmitters is relevant and important, the Board may consider it, whether or not the Retransmission Regulations remain in effect and whether or not criterion 2(c) requires the agreement to have been submitted in the course of this proceeding. The Board has a broad discretion to take into account factors that it deems to be relevant to the determination of fair and equitable royalties [150] and has the authority in appropriate circumstances to take steps itself to supplement the record – as long as procedural fairness is satisfied. [151]

[209] The written agreement in question is on record with the Board, albeit in a previous proceeding and is not controversial in itself. We see no prejudice to the Collectives in our considering it. On the view we take of the BDU’s trends analysis approach, as set out later in the economic analysis portion of these reasons, the settlement agreement is of little importance to the determination of the royalties here. Consequently, while we accept that we may take it into account, we need not afford it much, if any, weight.

[210] There is no need for the Board to search for other written agreements that might fall within criterion 2(c), in our view. Having regard to the way the Collectives and the objecting Retransmitters have conducted their respective cases and presented their evidence, there is no indication that there are other relevant written agreements to be submitted to the Board in this case.

vii. The effects of the Broadcasting Act and regulations made thereunder

[211] We now turn our attention to factor 2(b) of the Retransmission Regulations, namely, “the effects on the retransmission of distant signals in Canada of the application of the Broadcasting Act and regulations made thereunder.”

a. The Collectives

[212] According to the Collectives, with respect to the second factor, the regulatory evidence relating to the Broadcasting Act is, at best, inconclusive.

[213] The Collectives note that the BDUs’ own expert, Lori Assheton-Smith, conceded that TVA is the only station whose distant OTA signal must be carried by the BDUs pursuant to CRTC regulations. The Collectives add that Ms. Assheton-Smith conceded that, within the current regulatory context, DTH providers are already carrying “pretty much all the Canadian over-the-air signals,” presumably because they view them as being attractive to customers. [152]

[214] The Collectives argue it is further worth noting that considerations relating to the Broadcasting Act are already embedded into the tariff structure they proposed. As noted in the Board’s 1993 Decision, the different rate for the Francophone market reflects carriage rules in that marketplace, as does the preferential rate provided to small systems. Consequently, should the Board elect to consider this factor, it may take comfort in the fact that the priorities of the Broadcasting Act remain an integral part of the proposed tariff structure. [153]

b. The BDUs

[215] The BDUs filed the report of Lori Assheton-Smith [154] to address subsection 2(b) of the Retransmission Regulations. Ms. Assheton-Smith’s report, along with her testimony relating to that report, establish several contextual factors which, they argue, refute the Collectives’ claim that an increase in the average number of signals carried since 1990 indicates that distant signals as a whole have increased in value, or that an increase in the bare number of signals carried justifies an increase in the royalty rate.

[216] First, her report establishes that, to a certain extent, the increase in the number of signals carried since 1990 can be traced to some of the CRTC’s decisions and regulations. Therefore, to that extent, she argues that the increase in the average number of signals carried is not solely an indication of demand or value.

[217] Second, the BDUs argue that her report shows that the average number of signals carried has peaked and has already begun to decline, indicating the beginning of a downward trend in the average number of signals carried.

[218] Third, Ms. Assheton-Smith notes that the Broadcasting Act treats standard definition (SD) and high definition (HD) versions of the same channel as one signal, so the average of 55 signals per subscriber reported in the Mediastats study [155] is overstated because it counts the two versions of the same signal as two separate signals.

[219] Further to the Board’s considerations that are relevant under subsection 2(b) of the Retransmission Regulations, the BDUs urge the Board to be careful not to disadvantage Canadians in rural or small communities who have fewer local over-the-air stations available and are therefore more dependent on distant signals to receive basic level of broadcasting services than other Canadians who are located in larger markets and who obtain more services from the locally available signals. [156]

c. Analysis

[220] We agree with the proposition that the average number, and trends variation thereof, of distant signals carried is not – in and of itself – a direct indication of value. We also take note and agree with the statement that an SD and HD version of the same channel should be counted as one signal, as the latter is perfect substitution for the former for a subscriber receiving both.

[221] However, as the Collectives pointed out, considerations relating to the Broadcasting Act are already embedded into the tariff structure they are proposing. The differential rate for the Francophone market reflects carriage rules in that marketplace and the special rate provided to small systems reflect priorities of the Broadcasting Act and remain an integral part of the tariff structure.

[222] As a whole, the evidence in this proceeding on the factor enunciated at subsection 2(b) of the Retransmission Regulations does not lead us to conclude that consideration of this factor has an effect on the royalty rates.

[223] Consideration of the CRTC’s new policies with respect to the distribution of television services, which are discussed in response to Question 3 of Board Notice CB-CDA 2016-011, are also relevant to this provision of the Retransmission Regulations. We will address Broadcasting Regulatory Policy CRTC 2015-96 (the Broadcasting Policy) specifically in a dedicated section below.

viii. The Board’s Power to Vary and the Collectives’ Revised Tariff Proposal

[224] On May 8, 2015 – more than two years after the Proposed Tariff was filed – the Collectives proposed a set of revised rates [157] with the filing of their Statement of Case. On June 18, 2015, they sent a letter outlining the proposed revised rates, along with other information, to the Objectors and other BDUs. [158]

[225] The Collectives justify their late filing of the revised rates on the basis that they did not have the necessary information on which to base an assessment of the actual value of distant signals to the BDUs until that information was provided to them in August 2014, by way of confidential disclosure through the interrogatory process in these proceedings. They point out the last contested hearing on the retransmission rate took place in 1992. They say that, apart from some information provided to them as part of an interrogatory process prior to an agreement in 2004 – which they characterize as “long-obsolete” – they had no access to any source of detailed information about exactly how the BDUs distribute signals or services, or how much they charge, pay for, or profit from the distribution activities until the August 2014 disclosure. The confidential information was then analyzed by the Collectives’ expert economists, Professor Church and Dr. Wall, whose reports were provided in April 2015.

[226] The BDUs dispute this rationale. They argue there was other financial information available to the Collectives from various public sources, or on request outside of the interrogatory process, regarding the value to the BDUs of the program rights contained in distant signals. They submit the Collectives could have obtained this information and used it to establish the parameters of their initial proposed tariff.

[227] We need not resolve this debate. Even assuming there may be something to be said for the Collectives’ explanation, it is not the only considerations in play. On any analysis, the revised rates amount to a significant amendment to the initially Proposed Tariff. They must be examined – and if permitted to be claimed, given effect – in a manner that does not undermine the integrity of the tariff-setting process provided for in the Act. There are some concerns that need to be addressed in this regard.

[228] First, as noted above, the revised rates are substantially higher than the rates of the Proposed Tariff as published in the Canada Gazette – varying from an increase of approximately 89 per cent in 2014 to approximately 73 per cent in 2018. These are not marginal adjustments. For that reason alone, fairness concerns require that the proposed increases be carefully scrutinized.

[229] Second, the filing of revised rates in this fashion, more than two years after the initial filing of the Proposed Tariff, has implications with respect to the provisions in the Act for notice to interested and affected persons. The purpose of notice is to inform interested and affected persons that a proposed tariff has been filed and that they may object to it within 60 days of the time of publication.

[230] Finally – and importantly – the filing of revised rates in this fashion negatively impacts the ability of Retransmitters to protect themselves from the retroactive effect of the tariff ultimately approved by the Board.

[231] With these concerns in mind, we consider what effect, if any, to give to the revised rates.

[232] At the hearing, and in their written submissions, the parties framed much of their arguments regarding the Collectives’ revised rates around the principle of “non ultra petita”. That principle is a civil litigation concept generally understood to mean that a court will not make a ruling beyond what is requested by the parties. For instance, a judgment that grants a monetary award going beyond that sought by the claimant is said to be ultra petita or extra petita. In administrative law settings, the legislator is free to remove tribunals from the constraints of non ultra petita, particularly in situations in which a tribunal has been created to advance interests that go beyond the immediate interests of the parties before them. In this respect, the Federal Court of Appeal has confirmed that the Copyright Board is not rigidly bound by the principle. Ultimately, it is a matter of procedural fairness. [159]

[233] These proceedings do not attract the application of non ultra petita, however. All participants are well aware of the Collectives’ revised rates and what they are requesting. The entire hearing, and the expert and other evidence presented, revolved around the revised request. The issue is whether the Board should permit the revised claim to be considered and, if so, to what extent. That is a matter of procedural fairness and the Board’s power to amend or vary a proposed tariff.

[234] The Board has the authority under section 73 of the Act to modify or vary a proposed tariff and impose terms and conditions as it considers appropriate. That section empowers the Board to “establish […] such terms and conditions related to those royalties as the Board considers appropriate […and] vary the [proposed] tariffs accordingly.” While this is not an unlimited power, and must be exercised carefully, there are circumstances in which such a modification or variation – an “amendment” – will be called for. Indeed, in many instances the Board varies tariffs that are proposed in some fashion. In doing so, and in determining whether it should approve rates other than those originally proposed and published in the Canada Gazette, however, the Board must attempt to ensure that doing so would not unfairly prejudice interested or affected persons or give rise to some other procedural or substantive unfairness or violation of the principles of natural justice.

[235] With those considerations in mind, we have concluded, in the circumstances of these proceedings, that we ought not to approve a tariff in excess of the amounts claimed by the Collectives for the tariff period preceding the filing of the Collectives’ statement of case and the delivery of their letter dated June 18, 2015. We say that for the following reasons.

[236] As noted, the revised rates are significantly higher than those contained in the initial proposed tariff. Although a tariff can have a retroactive effect when it is approved by the Board, a proposed tariff always operates prospectively from the start of its effective date, which runs no earlier than from January 1^st following its proper filing. There is no mechanism in the Act whereby a tariff proposal may take effect prior to such effective date, and certainly not prior to its filing with the Board. Permitting the Collectives to file significantly “revised rates” in the fashion they have in the present case would enable them to benefit retroactively as if the amendment were made at the original filing date more than two years earlier, once approved – an advantage to which they would not otherwise be entitled. Generally speaking, we do not think the Board’s broad power to amend should be interpreted in a manner that would permit a party, in effect, to substitute for an initial proposed tariff another substantially different proposed tariff, in the guise of an “amendment”, after consideration of the initially proposed tariff is already substantially underway. The Board must have regard to the potential impact of such an initiative on the parties and other interested or affected persons.

[237] There is a related consideration as well. Under subsection 71(2) of the Act, a tariff proposal must be filed at least nine months before its effective date. In practice, this results in a notice period to potential users of a tariff of nine months from the proposed tariff’s filing, and approximately 7 months from the proposed tariff’s publication in the Canada Gazette. One of the functions of this notice period is to permit potential users (whether they submit objections or not) to arrange their affairs in order to protect themselves with respect to the accrued liabilities associated with potential retroactive tariffs. Generally they do so by collecting additional amounts from their customers or otherwise setting aside necessary funds, or by altering their business activities or pricing structures.

[238] In the circumstances of this case, we think it unlikely that Retransmitters – in the face of such a significant potential retroactive increase in royalties over such an extended period of time – would be able to find reasonable ways in which to retroactively collect from their customers and/or otherwise set aside funds necessary to pay the difference, along with interest, between the royalties originally proposed and published in the Canada Gazette on June 1, 2013, and the royalties sought in the revised proposed tariff. This may be particularly relevant for smaller, independent and less financially-robust Retransmitters.

[239] Finally, the publication of a proposed tariff in the Canada Gazette also serves to inform parties of the possibility to object to such a proposal, and the timelines within which these are to be made. In this case, the Collectives argue that their letter of June 18, 2015 served a similar purpose.

[240] The letter had its flaws. By its very nature it could not perform the same function as formal notice of a proposed tariff published in the Canadian Gazette. Although it is said to have been sent to all known Retransmitters regardless of size and regardless of whether the recipients had filed a notice of objection, it contained little meaningful information as to what an unrepresented Retransmitter could do in face of its receipt. In the end, however, we do not find it necessary to determine whether the letter could represent the functional equivalent of formal notice under the Act because, based on our analysis set out elsewhere in these reasons, we would not approve a tariff in excess of the amounts claimed in the initial proposed tariff for the post-letter tariff period in any event. In addition, the letter – whether flawed or not – could not resolve the difficulties referred to above regarding retroactivity and notice with respect to the lengthy prior period.

[241] The tariffs we fix reflect the foregoing analysis.

[242] We turn now to the final legal issue for consideration.

ix. The effect of Broadcasting Regulatory Policy CRTC 2015-96 on the current proceedings

[243] The CRTC’s Broadcasting Policy, sometimes referred to by the CRTC and the parties as “Let’s Talk TV”, was released on March 19, 2015.

[244] One of the questions that the parties were asked to address specifically in their respective Legal Briefs was “Whether the Board should account for Broadcasting Regulatory Policy CRTC 2015-96 set to come into force in March 2016 and if so, how?”

a. The Collectives

[245] The Collectives argued that given that the effects of the Broadcasting Policy on the value of the retransmission of distant signals would likely be minor and not be known for years, the Board should not take into account the Broadcasting Policy in determining the royalties for the years under consideration.

[246] Given that the changes to the Broadcasting Policy phased in beginning in March 2016, almost halfway through the proposed tariff term, the Collectives argued that experts for both sides agreed that the implications of the Broadcasting Policy, including on the extent of carriage and the value of distant signals, will not be known for many years.

[247] With respect to the predicted impact of the Broadcasting Policy and the introduction of “skinny basic” in 2016, Mr. Grant states that this change may result in fewer distant signals on average being delivered to Canadian subscribers over time. Mr. Grant also points out that all other packages offered by BDUs would still contain distant signals and concludes that it would still take a number of years before the changes can be implemented and any effects assessed. [160]

[248] While corporate witnesses for the Objectors contended that the effects of the Broadcasting Policy would be drastic, their own representations before the CRTC were that the removal of the key U.S. 4+1 signals from skinny basic “would harm the system perhaps irreparably” and would result in “significant churn”. [161] Their own studies before the CRTC indicated that subscribers greatly valued distant signals. At this juncture, it is simply premature to accord the BDUs’ contrary contentions any real weight.

b. The BDUs

[249] The BDUs argued that the Board should take into account the amended distribution and packaging rules contained in the Broadcasting Policy, which came into force on March 1, 2016. They contend that a likely effect of these changes is that Canadians will choose to subscribe to fewer distant signals. The overall value of distant signals to BDUs will decrease concomitantly. Canadians will have increased control over the programming services they can choose to receive from BDUs, and increased access to the same or similar programming on-demand through the VOD platforms. This may provide incentive to Canadians to reduce the number of overall linear television services to which they subscribe without having to sacrifice the amount of programming they can access. There is no reason to believe that distant signals will be immune from this trend. At the same time, the CRTC rules operate to require BDUs to carry some distant signals on the basic service in many cases. This means that the retransmission royalty will be owing for the entire or a large subscriber base despite the overall decline in value of distant signals to BDUs.

c. Analysis

[250] We agree that any effects of the Broadcasting Policy would likely be an eventual overall decrease in the number of distant signals retransmitted, as the evidence shows that a significant amount of retransmitted distant signals are never viewed. In a system allowing a subscriber to pay solely for the signals such subscriber picks à la carte, the number should inevitably go down, although the effect on the aggregate value of all distant signals remains uncertain.

[251] However, quantifying any effect of the Broadcasting Policy on the value of the retransmission of distant signals would be a challenge in the present proceeding with the data currently available. While it is reasonable to assume that the average number of distant signals received by BDU subscribers is likely to decrease, the correlation between the average number of distant signals and the value of such signals cannot be assumed to be linear. Better data for BDUs should be available from the CRTC when the Board considers the next tariff dealing with television retransmission. Hence, we decide to refrain from making any adjustment to the royalty rate in this case based on the effect of the Broadcasting Policy.

d. Technological neutrality

[252] We want briefly to address the notion of technological neutrality as it relates to the current proceeding.

[253] The principle of technological neutrality is a recognition that, absent parliamentary intent to the contrary, the Act should not be interpreted or applied to favour or discriminate against any particular form of technology. It results from the balancing of user and right-holder interests discussed by the Supreme Court of Canada in Théberge, a “balance between promoting the public interest in the encouragement and dissemination of works of the arts and intellect and obtaining a just reward for the creator.” [162] Because this long-standing principle informs the Act as a whole, it must be maintained across all technological contexts: “The traditional balance between authors and users should be preserved in the digital environment.” [163]

[254] Given the approach we use and explain in details later in the present reasons, there are no issues of technological neutrality that necessitate a specific analysis or specific adjustment in the determination of the proper rates in this proceeding.

VII. ECONOMIC ANALYSIS

[255] Three methodologies have been presented in this proceeding to set tariffs for the retransmission of distant signals: the proxy approach, the trend-analysis approach, and the direct-market approach.

[256] Three variants of the proxy approach were proposed by Professor Church (the Collectives), Dr. Chipty (the BDUs), and Dr. Wall (the Collectives). Professor Church estimates a price of $2.76 per subscriber per month. His approach is based on a proxy composed of 24 U.S. specialty services and uses Numeris data. Dr. Chipty estimates a price range of $0.87 to $1.20 per subscriber per month. Her approach is based on a proxy composed of 18 U.S. specialty services and 47 Canadian category B specialty services and uses set-top box data. Dr. Wall calculates the price of distant signals by summing up the growth rates of the wholesale price of A&E and the average number of distant signals per subscriber in 2009-2013, and applying this growth rate to the $0.98 certified price of 2013 for 2014-2018. Dr. Wall obtains a rate of $1.95 for 2018.

[257] The trend-analysis approach suggested by the BDUs starts from $0.98 per subscriber per month in 2013. Further adjustments may be applied to the rate due to inflation and change in viewership.

[258] Dr. Wall’s direct market approach is based on the market valuation of distant signal packages. This approach yields a price of $4.97 per subscriber per month. The differences between Dr. Wall’s proxy approach and his market approach, and the prices they generate, are explained below.

[259] These methodologies are analysed below. Our preferred approach is explained subsequently.

A. Professor Church’s proxy approach

i. Description

[260] Professor Church presents a comparable-services or proxy-market analysis to estimate the value of distant signals. The proxy approach estimates the value of distant signals based on the observed value of comparable services in a competitive market. Professor Church’s method has three parts. First, he identifies a group of programming services (channels) which are qualitatively comparable to distant signals. Second, he verifies that the comparable set of channels is supplied competitively. Finally, he calculates payments to the comparable set of channels.

[261] Professor Church explains that, from an economic perspective, the per-subscriber price paid by BDUs for a group of U.S. specialty services is an acceptable proxy for the price to be paid by BDUs for distant signals. In so concluding, he relies on two assumptions. First, the same BDUs both purchase and retransmit U.S. specialty services and distant signals. The amounts paid by a single buyer for two different goods reflect that buyer’s relative preference for those goods. Second, the BDUs purchase specialty services in a competitive market.

[262] He further explains that U.S. specialty services are of comparable value to distant signals for the following reasons.

[263] First, the aggregate of U.S. specialty service programming and the aggregate of distant signals offer a wide range of similar programming such as drama, comedy, soaps, documentaries, news, sports.

[264] Second, viewing-minutes comparison using a variety of aggregated and disaggregated sources suggests similar (sometimes higher) viewership shares for distant signals in total compared to U.S. specialty channels. For example, based on Canada-wide Numeris 2009-2014 diary data, [164] viewing share of U.S. conventional stations substantially exceeds the viewing share of U.S. specialty services. Professor Church also performs a robustness check, given that some U.S. channels may be misclassified in the Numeris data. He assumes that only approximately 70 per cent of U.S. 4+1 viewing is distant viewing; even under that assumption, the viewing share of distant signals is still as high as U.S. specialty service viewing.

[265] Third, survey data and statements from BDUs confirm that the U.S. 4+1, mostly consists of distant-signal tuning, are a uniquely valuable part of their programming. [165]

[266] Based on his analysis, Professor Church suggests that U.S. specialty services might even be a conservative comparator for distant signals, meaning that customers may value them less than they value distant signals, especially the U.S. 4+1 signals. Professor Church also argues that the prices of U.S. specialty services in Canada are set in a competitive market, and that the evidence is not consistent with U.S. specialty services having market power in the supply of programming to BDUs in Canada.

[267] Professor Church argues that reviewing the evidence filed by BDUs in recent “Let’s Talk TV” [166] proceeding before the CRTC shows the importance of distant signals to BDUs and their subscribers in Canada. According to Professor Church, the evidence demonstrates the significant and comparable value of the bundle of U.S. specialty services and distant signals, particularly the U.S. 4+1 signals retransmitted by all BDUs and viewed primarily as distant signals in Canada.

[268] Accordingly, Professor Church calculates total payments made by four English-Canadian BDUs (Bell, Rogers, Shaw, and TELUS) to 24 U.S. specialty services. [167] To obtain the total payments made by the four English-Canadian BDUs to the bundle of U.S. specialty channels, Professor Church multiplies the monthly rates of the U.S. specialty services by the number of subscribers to the service averaged over twelve months. By dividing this total payment for U.S. specialty services by the total number of subscribers to those BDUs, Professor Church obtains a monthly per-subscriber price of $2.76 for the set of U.S. specialty services. [168] Under the assumption that U.S. specialty services have similar values to distant signals, this price is also the rate payable for distant signals.

[269] Professor Church argues that since he divides the total payments by the total number of subscribers to the BDUs, not just the subscribers to the included services, this controls for the lower penetration rate of the U.S. specialty services compared to the penetration rate of distant signal. [169]

[270] Professor Church explains that this wholesale estimated price of $2.76 per month per subscriber may be conservative (i.e. low) for a number of reasons. First, even for the four BDUs he selected, all payments made to the U.S. specialty services are not provided. As a result, the total payment he calculated is an underestimate of the true total.

[271] Second, to account for differences in the penetration rates of U.S. specialty services and distant signals, Professor Church adopted the conservative assumption that specialty services would achieve 100 per cent penetration, without any additional wholesale payment being made. As a result, the calculated monthly per-subscriber fee paid by the BDUs for the bundle of U.S. specialty services (proxy) has been significantly reduced from its actual market price.

[272] Third, Professor Church suggests that BDUs may be able to exercise market power when buying U.S. specialty services, which could drive down payments for U.S. specialty services. This market distortion suggests that the value of U.S. specialty services to the BDUs and to their subscribers is greater than the wholesale price would indicate.

[273] Professor Church applies no discount for simultaneous substitution, as he argues that the Numeris data he uses for his analysis has already taken care of simultaneous substitution in viewing minutes. [170]

[274] The BDUs and their experts made several comments on Professor Church’s analysis and argue that his approach results in a significant overstatement of the value of distant signal programming. [171]

ii. BDU’s comments

[275] Dr. Chipty’s main comments to Professor Church’s report are as follows.

[276] First, Professor Church assumes that the amount of viewing of distant-signal programming is the same as the viewing to his benchmark U.S. specialty services, although the evidence on which he relies shows that there is actually substantially less viewing of distant-signal programming.

[277] Second, Professor Church makes no adjustment to account for simultaneous substitution, even though when substitution occurs, the programming on the distant signal is replaced by the same programming from a local signal for which no royalties are payable.

[278] Third, Professor Church includes all payments made to specialty services in the benchmark rate rather than just the portion of the payments that reflect the value of the programming, which would result in an over-compensation of the rights holders in distant signal programming.

[279] Fourth, Professor Church includes viewing of local U.S. 4+1 signals as viewing of distant signal programming, which significantly overstates the value of distant-signal programming.

[280] Fifth, Professor Church ignores other important factors that require further adjustments to the benchmark rate. For example, he does not consider the characteristic of substitutability and the fact that distant signal programming is often duplicated in other signals and services provided by the BDU; he does not take into account the fact that the U.S. specialty services included in his benchmark group are not able to sell advertising in the Canadian market; he does not consider whether it would have been appropriate to include Canadian specialty services in the benchmark group, even though these services have more in common with distant signals than the U.S. specialty services.

[281] The BDUs highlighted how the Collectives’ witness Mr. Grant explained how a number of U.S. OTA television stations located near the US-Canadian border are actually local signals in parts of Canada, because of the reach of the stations’ Noise Limiting Bounding Contour (NLBC) plus 32 kilometres. [172] For example, as the BDUs point out, Mr. Grant explained how the Buffalo station WKBW, the ABC affiliate, is local to most of the Niagara Peninsula, [173] and WIVB, the CBS affiliate, is local to almost all of Toronto.

[282] The other point the BDUs raise is that as a result of Professor Church’s misunderstanding of simultaneous substitution, when he analyzes the Rogers set-top data to determine the relative viewing of distant signals, he counts tuning to local stations that have been substituted for distant signals as distant-signal tuning. [174] In other words, he counts distant-signal tuning where no distant-signal tuning is occurring.

[283] During the cross-examination, the BDUs showed that Professor Church was not familiar with the programming content of many of the U.S. specialty services in his proxy. [175] Also, the BDUs raised the point that some of the specialty services Professor Church included in his proxy target very specific audiences in “niche” markets and as such, their prices tend to be higher than the prices of more general distant signals. [176]

[284] According to the BDUs, in constructing his benchmark group of U.S. specialty services, Professor Church said he chose services that offered programming of a similar quality and variety as the distant signals he was trying to value. [177] However, he admitted that nowhere in his report did he analyze or describe the programming carried out by each of the U.S. specialty services. [178] Of the 21 unique channels [179] included in his benchmark group of U.S. specialty services, Professor Church admitted that he did not know about the programming offered on eight of them. [180]

[285] Moreover, the BDUs state that Professor Church agreed that one of the 21 services included in his benchmark, Peachtree TV, is actually a retransmitted over-the-air signal and not a specialty service. [181] Removing Peachtree TV leaves 20 services for analysis.

[286] The BDUs further point out that Professor Church accepted that seven of these 20 U.S. specialty services are news services, [182] and of the remaining 13 services, six are sports services. [183] The BDUs argue that, as Professor Church also agreed, specialized services such as the ones included in the benchmark are highly valued by people who have interest in that genre of programming, which could result in higher prices. [184]

[287] The BDUs also point out that Professor Church has admitted that where U.S. specialty services offer original programming not available on any other channel, a BDU would risk losing subscribers if it did not offer that U.S. specialty service while its competitor did offer it. [185]

[288] With respect to Professor Church’s comment that BDUs may exercise market power, Ms. Blackwell, the BDUs’ other expert, rejects this assertion by stating that the evidence she provided in her report is consistent with a highly competitive market for BDUs. [186] According to Ms. Blackwell, Professor Church refers to market shares of different BDU platforms, and in particular, the national, market share of cable companies which he calculated at 66 per cent of subscribers, as an indication of market concentration at the local level that could allow BDUs to exercise market power. However, Ms. Blackwell argues that Professor Church has provided expert evidence in other fora concluding that market shares are not a reliable indication of market power.

iii. Analysis

[289] We accept that Professor Church’s method is sensible. It is conceptually similar to the Board’s original approach to set royalties in its 1990 Decision and to the BDUs’ expert report by Dr. Chipty. We conclude, however, that, as indicated by the BDUs, there are important shortcomings with Professor Church’s proxy approach, and that certain adjustments are necessary if such a proxy approach is to be used.

[290] First, there are some important issues with Professor Church’s choice of proxy.

a. Representativeness of his selected 24 U.S. specialty services

[291] Although Professor Church provided some explanation that a proxy composed of his selected 24 U.S. specialty services has similar programming content to Canadian distant signals, we are not satisfied that the 24 U.S. specialty services on which he relies are sufficiently representative to provide a valid proxy.

[292] For a proxy to be useful, it must be similar to the target in relevant dimensions. In this case, we seek similarity in both the statistical distribution of programming types (e.g., genres) and the quality of the content. For example, if distant signals have a given distribution of news, drama, comedy, etc., taken in the aggregate, the proxy should also have a similar genre distribution. Moreover, the quality of the news, drama, comedy, and other genres on the proxy should be similar to those of distant signals, in aggregate.

[293] By referring to consumer surveys and the list of top 100 shows in the United States (by audience), Professor Church claims but does not show that the satisfaction of consumers from distant signals, and the quality of their programs, is comparable to that of the proxy at the aggregate level. [187]

[294] However, Professor Church presents no analysis of the programming content (for example by genre) of distant signals and specialty services. [188] In fact, Professor Church testified that he was not familiar with the programming content of many of the 24 U.S. specialty services he chose for his proxy. [189] Professor Church claims that the viewing minutes of distant signals, in the aggregate, are at least as much as the viewing of specialty services; [190] from the equality of viewing, he infers an equality of value. However, there are two issues with this argument.

[295] First, there is currently no consensus among the parties on the relative shares of distant-signal viewing and specialty-services viewing in total TV viewing. [191] This disagreement comes both from the different data sources used by the parties and how the parties use the data they have. [192]

[296] Second, the parties also disagree on the nature of the relationship between viewing and value. [193] Although we agree that there is a positive relationship between viewing and value, we do not know the nature of this relationship.

b. Presence of niche services

[297] As raised by the BDUs, [194] some of the specialty services Professor Church included in his proxy target very specific audiences in “niche” markets. Examples of these services include Black Entertainment Television Network, Golf Channel, Military Channel, NFL Network, Speed Network, and Playboy TV. [195] We agree with the BDUs’ submission that the prices of niche services generally tend to be higher than the more general channels because they can exercise market power in these niche markets. As a result, the total payment for these services may be higher than the payment for more general channels like distant signals. This issue leads to a higher price for the proxy. [196] This issue was also discussed for the price of A&E in the 1990 Decision. [197] This potential market power of the proxy warrants a downward adjustment on its price.

c. Are prices of the U.S. services competitive?

[298] There is not a satisfactory economic analysis by Professor Church to prove that the prices of the selected U.S. specialty services are offered competitively. Reviewing the payments of the specialty services by the BDUs does not confirm that there is a competitive market for the U.S. specialty services in Canada. For example, there are disparities among the BDUs on payments for the same service. [198]

[299] Second, Professor Church does not correct his benchmark proxy for the “non-copyrighted” portion of the costs embedded in the price of the proxy. We agree with Dr. Chipty that the tariff should apply only to the programming portion of the payments. The cost of non-copyrighted content should be removed from the payment.

[300] Third, Professor Church does not take into consideration the effect of program substitutability. We agree with Dr. Chipty that it is possible to substitute distant-signal programs with other opportunities to view the same programs. [199] As such, it is reasonable to accept that these opportunities put downward pressure on the value of distant signals more than the value of specialty services because, at least, the time-shifting feature of distant signals will not be as important to the subscribers. Substitution opportunities may become available through the use of the PVR, or though VOD and OTT services. [200]

B. Dr. Chipty’s proxy approach

i. Description

[301] Dr. Chipty considers the set of possible benchmark signals or services that might be used to derive a value for distant signal programming. She concludes that a mix of U.S. specialty services and Category B Canadian specialty services could be used as an appropriate starting point, since the mix of programming among these services is similar to the programming on distant signals and the rates between these services and BDUs are freely negotiated.

[302] Dr. Chipty explains that an appropriate methodology to set royalty rates constitutes the following steps:

Step 1 – Identifying appropriate benchmark services and the payments they receive;

Step 2 – Identifying the share of payments earned by benchmark services attributable to the programming, as opposed to other inputs;

Step 3 – Adjusting for simultaneous substitution; and

Step 4 – Adjusting for the difference in relative value of programming.

[303] Dr. Chipty also conducts a sensitivity analysis to adjust for other opportunities to substitute programming with identical programming available elsewhere on the BDUs’ channel lineup. Her approach is detailed below.

[304] In Step 1, Dr. Chipty’s approach to choose a benchmark is based on comparing the value of distant signal programming with a set of benchmark programming. According to her, the simple count of distant signals is misleading because of the flat fee rate structure designed by the Board. [201] Dr. Chipty argues that a benchmark should embody as best as possible the two following conditions: (i) comparable appeal of programming; and (ii) comparable competitive conditions.

[305] Dr. Chipty chooses the set of U.S. specialty services and Canadian Category B specialty services as a benchmark for distant signals. Her justification for this benchmark is as follows. First, like the U.S. specialty services, Canadian Category B specialty services are not “must carry” and they negotiate fees for carriage with BDUs.

[306] Second, the business models of Canadian Category B specialty services and Canadian distant signals are more similar to each other than to the U.S. specialty services. For example, both Category B specialty services and Canadian distant signals earn Canadian advertising revenues, while U.S. specialty services and U.S. distant signals do not.

[307] Third, both Category B specialty services and Canadian distant signals maintain some degree of vertical integration with content partners, while no U.S. specialty services and U.S. distant signals are vertically integrated.

[308] Finally, since Canadian Category B specialty services carry Canadian programming, U.S. specialty services and Canadian Category B specialty services together better resemble the mix of programming available on distant signals and maintain a more comparable programming mix and range of genres compared to U.S. specialty services alone, which are more “specialized” relative to distant signals.

[309] Based on these factors, Dr. Chipty concludes that a combination of the U.S. specialty services and Canadian Category B specialty services is a better benchmark than U.S. specialty services alone. [202]

[310] Seven BDUs provided information on the number of subscribers and price per subscriber as of 2013 for 25 unique U.S. specialty services and 105 unique Category B specialty services. Among these services, Dr. Chipty selected a benchmark group consisting of 18 U.S. specialty services and 47 Category B specialty services, as offered by six BDUs. [203] Dr. Chipty excluded certain services, since either information on their price or number of subscribers was not available or their relevant information did not exist in the set-top box data she received from the BDUs.

[311] To obtain a starting benchmark rate, she calculates total available payments by the six English BDUs for the benchmark service and divides this total payment by the number of subscribers to the six BDUs for her benchmark programs. This yields a benchmark price of $3.29 per subscriber per month for 2013. After adjusting for inflation from 2013 to 2015, she obtains a benchmark price of $3.38 per subscriber per month for 2015. [204]

[312] To isolate the cost of programming in Step 2, Dr. Chipty argues that when she calculates the total available payments by the six English BDUs for the benchmark service, the payments include more than just payments for programming rights. She specifically distinguishes among three different components of a payment as follows: (i) cost of programming, including returns to rights-holders; (ii) costs of other inputs, such as assembling or aggregating the content into a compilation of programming, branding, marketing, and distributing the content; and (iii) profit earned by the service.

[313] Dr. Chipty argues that an appropriate royalty rate should reflect payments only for programming rights (i.e. item (i)). However, because of data limitations, she can only isolate the profit from the payment, and thus, her benchmark price includes both the cost of programming and the cost of other inputs. To isolate the profit from total payments, Dr. Chipty uses the CRTC’s report, [205] where a 25 per cent average profit is calculated for all Canadian Category B specialty services and a 10 per cent average profit is calculated for non-vertically integrated Canadian Category B specialty services. Since there is no profit margin available for the U.S. specialty services, Dr. Chipty assumes at least 10 per cent profit for the U.S. specialty services. This is under the assumption that since the U.S. specialty services are not vertically integrated with BDUs, they have a lower profit, like the non-vertically integrated Category B specialty services.

[314] These profit margins imply that the cost of programming is 75 per cent of payments for Category B specialty services and 90 per cent of payments for U.S. specialty services. Dr. Chipty reduces her benchmark rate accordingly.

[315] To support Dr. Chipty’s argument, the BDUs state that reducing the benchmark price so that it only reflects the value of the programming is the only approach that is consistent with the Act. To support this argument, the BDUs point out that the retransmission statutory licence is set out at Section 31 of the Act, and that the Supreme Court of Canada reviewed the legislative history and intent of section 31 in Cogeco v. Bell. [206] The BDUs argue that it is clear from the Supreme Court’s analysis that the provision is only intended to apply to “works” (i.e. programming) and intentionally excludes “signals”. Accordingly, asking BDUs to pay royalties to the owners of programming that reflect both the value of the works (programming) and also non-programming value attributable to the signal would overcompensate the owners of the programming.

[316] In Steps 3 and 4, Dr. Chipty adjusts for the relative value of programming and simultaneous opportunities using the set-top box data the BDUs collected and provided to her.

[317] Bell, Rogers and Shaw collected set-top box data in the cities of Toronto, Montreal, and Vancouver during the two-week period beginning May 4, 2015. This sample covers all Rogers’ digital TV customers, only Bell’s Fibe customers, and only Shaw’s Gateway set-top box customers. The set-top box data included a schedule of programs aired on the signals and services distributed by BDUs as well as the hours tuned to this programming by subscribers. Using these data, the Mediastats Report, and CRTC data, [207] Dr. Chipty calculates some adjustments.

[318] Dr. Chipty explains that, given the way these set-top-box data is collected, she focuses on the most popular packages in the three largest cities. Moreover, she considers only the services for which she had data available from the set-top box. Dr. Chipty weights her samples so as to make them representative of all BDU subscribers in the three cities. She argues that because she is focused on the viewing behavior in large cable systems (i.e. more than 6,000 subscribers), and that most large cable systems serve the largest cities of Canada, this sample better serves her analysis than nationally representative Numeris data. The reason is that it is the viewing behavior in those cities, not viewing behavior nationally, that is most relevant for her study. Moreover, the set-top box data is sufficiently detailed to allow for adjustments to her analysis.

[319] Dr. Chipty applies the following adjustments to her benchmark price.

a. Adjusting for the difference in popularity

[320] Dr. Chipty notes that distant signals are viewed less than the benchmark services. She argues that this difference in viewership requires adjustments to the benchmark price. Dr. Chipty uses the ratio of the viewership minutes of distant signal to the viewership minutes of the benchmark services as the adjustment factor. This adjustment is imperfect, as Dr. Chipty notes. She explains that, because of the geographic proximity of Toronto and Montreal to the U.S., the data misclassify some U.S. signals as local signals rather than distant signals. Dr. Chipty corrects the bias in her adjustment factor using the local-distant split in the Numeris data. [208]

b. Substitution opportunities

[321] Dr. Chipty discusses various substitution availabilities for distant signal programs: (a) episode available simultaneously on another signal or service. This includes simultaneous substitution of distant signals by local signals and other simultaneous substitution opportunities such as programs available on video-on-demand (VOD) and over-the-air (OTA); (b) episode available at other times on another signal or service; and (c) program available elsewhere, including on VOD, OTA, and over-the-top (OTT) services like Netflix. According to Dr. Chipty, these opportunities for choice available to consumers create competition among programmers, which should place downward pressure on prices.

[322] After reassigning distant signal minutes to correct for the share of local and distant signals in Toronto and Montreal according to Numeris data and taking into account the simultaneous substitution of distant-signals programming, Dr. Chipty calculates an adjustment factor as the ratio of distant-signal viewing to her benchmark-services viewing to be used as a measure of the relative popularity of programming. Applying this adjustment factor to her benchmark price, after isolating the cost of programming, yields a royalty rate of $1.20 per subscriber per month.

[323] Moreover, as part of her sensitivity analysis, Dr. Chipty uses the 57.6 per cent personal video recorder (PVR) penetration rate as an extra measure to account for episodes available at different times. If the use of the PVR is included in the calculation, the adjustment factor drops further, yielding a royalty rate of $0.87 per subscriber per month.

[324] Dr. Chipty’s final calculation provides a rate range of between $0.87 and $1.20. She then recommends that the appropriate, distant-signal royalty rate should be approximately $1 per subscriber per month based on this range.

[325] Dr. Chipty argues that this rate is a conservative estimate for royalty rates since her calculation ignores differences between distant signals and the benchmark services in other opportunities to substitute programming; ignores the increasing availability and use of other platforms (e.g., VOD and OTT) to time-shift; uses the Numeris split of distant / local for the U.S. 4+1, even though more of U.S. 4+1 viewing may be local in locations with large cable systems; and assumes that U.S. specialty margins are only 10 per cent, even though U.S. data suggests this number may be substantially higher. [209] Moreover, Dr. Chipty notes that many distant signals attract small audiences, and that households view only a few distant signals.

[326] Dr. Chipty also considers other factors that might affect the valuation of distant-signal programming. Examples she considers include the mandatory carriage of some distant signals, the Canadian revenue sources available to U.S. specialty services, the requirement of BDUs to carry unaffiliated Category B specialty services, and the specialization of specialty-service programming. Dr. Chipty argues that these factors may indicate that her calculated rate of $1 is overstated.

ii. Collectives’ comments

[327] In response, the Collectives argue that Dr. Chipty’s analysis includes several factual and methodological errors. [210] The Collectives’ comments are presented below.

[328] First, the Collectives’ experts state that it is inappropriate to add Category B specialty services to the proxy group. According to the Collectives, Category B specialty services are small, lightly-watched, low-cost, and in the aggregate, less valuable to the BDUs than are U.S. specialty services. As a result, including Category B services in the analysis drives down the average fees paid to the services in the benchmark group. [211]

[329] Furthermore, most of these Category B specialty services are owned by the vertically-integrated BDUs and their respective affiliated programming undertakings, and thus their fees are not negotiated at arms’ length. To expand on this comment, the Collectives and their experts [212] argue that approximately 70 per cent of the Category B specialty services used in Dr. Chipty’s benchmark are owned by a vertically-integrated BDU.

[330] As a result of vertical integration, those Category B specialty services are traded in one of three ways: (i) traded internally between the vertically-integrated BDU and its broadcasting arm, (ii) traded and cross-licensed between vertically-integrated BDUs, who are effectively required by regulation to carry each other’s Category B specialty services (as discussed below), or (iii) traded between a vertically integrated BDU and a non-vertically-integrated BDU. The Collectives explain that, in at least the first two transactions, the prices are not set by arm’s length negotiations.

[331] The Collectives point out that Dr. Chipty acknowledged that vertical integration may have an impact on the price of a Category B specialty service, but she claims that the impact is ambiguous. [213] However, the Collectives state that Professor Church’s analysis shows that, on average, a non-vertically integrated BDU pays substantially more than a vertically integrated BDU for the same Category B service. [214] The Collectives also indicate that the existence of a distortion of the price of Category B specialty services has been repeatedly stated by non-vertically integrated BDUs to the CRTC. [215] The Collectives argue that even if the impact were ambiguous, as the objective is to use a price that is free from such distortions, it is inappropriate to rely on such prices.

[332] Moreover, the Collectives explain that Canadian BDUs are required to distribute three unrelated Category B specialty services for each related Category B service. This requirement results in creating artificial demand for Category B specialty services. The Collectives also state that Category B specialty services are subject to content restrictions, as they are prohibited from infringing on the genre exclusivity granted to other services. [216]

[333] Additionally, the Collectives argue that while Dr. Chipty claims that the business model of some distant signals and some Category B specialty services may be similar, she provides no analysis to demonstrate any such similarity, its extent or impact on the comparative value. [217]

[334] In response, the BDUs draw [218] on Dr. Wall’s testimony which states while the vertically-integrated Category B specialty services account for 70 per cent of the 47 services in Dr. Chipty’s benchmark group, they generate 83 per cent of the revenues associated with that group, meaning that the Category B specialty services are earning more than the average amount of revenue for the group. [219] The BDUs also argue that while Professor Church says that Canadian Category B specialty services cannot be used in a benchmark group for distant-signal programming, Dr. Wall includes those and other Canadian specialty services in his analysis. Moreover, Dr. Chipty states that Professor Church assumes, without basis, that his set of U.S. specialty services is a perfect benchmark that requires absolutely no adjustment, which, in her opinion, is inappropriate. [220]

[335] Second, the Collectives’ experts testified that Dr. Chipty makes a series of analytical errors that lead her to make unwarranted downward adjustments to the prices paid for her benchmark group. For example, the Collectives’ experts reject the methodology of Dr. Chipty to adjust the benchmark price based on the viewing minutes of the benchmark services and distant signals. As Dr. Wall [221] and Professor Church [222] testified, this assumption of linearity between value and viewing is not tested or justified empirically by Dr. Chipty and is incorrect.

[336] Professor Church also disagrees with Dr. Chipty on isolating the cost of programming in specialty services because he argues it models the wrong hypothetical competitive negotiation between BDUs and the right holders. As Professor Church explains, in a competitive market, the negotiations would not occur between the BDUs and individual programming rights holders, but between the BDUs and channels, and that the BDUs would not separate the costs that go to programming versus the costs that go to cover other inputs in order to only pay for the costs of programming. [223]

[337] Third, the Collectives’ experts state that Dr. Chipty’s calculations are based on set-top box tuning data, which is incorrectly described as viewing data, owned by three BDUs and developed by them for her to use in this case.

[338] The Collectives argue that while Numeris provides nationally representative viewing data, which is industry accepted data and used by BDUs, the Collectives, as well as the CRTC, Dr. Chipty used set-top box tuning data which is not publicly available.

[339] To address this comment, the BDUs respond that, as Ms. Cooper testified, [224] the Collectives rely on viewing studies as a measure of the value of programming to allocate the retransmission royalties among the rights owners. The BDUs argue that changes in viewing of distant signal programming is similarly a reasonable method for determining changes in the value of that programming that might justify changes in the royalty rate.

[340] The BDUs also point out that in his report, Professor Church stated that viewing is a measure of the quality of a channel, that the CRTC has identified the importance of set-top box tuning minutes for measuring audiences, and that “inferences regarding channel popularity based on tuning minutes are consistent with the inference from viewing data collected in the Diary evidence complied by Numeris.” [225]

[341] Furthermore, the BDUs argue that according to Professor Church’s testimony, he used the tuning minutes from the set-top box data to determine the relative value of U.S. specialty services to distant signals. [226] The BDUs also point out that Professor Church agreed that if the deviation between tuning minutes and viewing minutes is the same for distant signals as it is for specialty services, then the same relative indicator of value exists between the two and it is not a mistake to use the tuning minutes from the set-top boxes. [227]

[342] Next, the Collectives argue that set-top box data is not reliable for at least the following reasons. [228] First, set-top boxes do not record viewing of a signal. Set-top boxes only record what the set-top box is tuned to. Second, set-top boxes do not record if anybody is watching the tuned signal, or how many people may be watching the tuned signal. Third, set-top boxes continue to record tuning to a signal even after the television set has been turned off. Finally, set-top boxes cannot capture the tuning habits of subscribers to satellite services, which represents a very significant deficiency in the data set.

[343] The Collectives also point out that Bell has cautioned against using set-top box data as it is “not a complete audience metric such as those provided by BBM” and “consequently it cannot be relied upon with complete accuracy”, [229] and that Vidéotron submitted to the CRTC that because of inherent limitations of set-top box data “we are convinced that BBM [Numeris] remains the best tool for providing appropriate audience measurements.” [230]

[344] Moreover, the Collectives argue that even if set-top box data could be used in theory, the particular set-top-box data that Dr. Chipty collected is neither representative nor appropriate to use to set a national tariff. In particular, the Collectives state that the set-top box data Dr. Chipty uses is limited to three cable BDUs, in three unrepresentative markets, for an unrepresentative time period, and from an unrepresentative set of subscribers to unrepresentative cable packages. [231]

[345] According to the Collectives, while the vast majority of Canadian subscribers (88.5 per cent) are served by large systems, large systems are located across Canada not just in Toronto, Montreal, and Vancouver, and hence, to accurately survey the viewing behaviour of subscribers to large systems across Canada (i.e. about 90 per cent of Canadian subscribers), it is necessary to survey national viewing habits, not just the viewing habits of subscribers in three large markets. [232]

[346] To support their argument, the Collectives draw on Mr. Kiefl’s report [233] which states that the set-top-box data used by Dr. Chipty is completely unrepresentative of actual viewing shares and habits; these data are entirely at odds with the viewing data collected and published by Numeris on which all industry participants routinely rely.

[347] According to Mr. Kiefl, Dr. Chipty’s set-top data raises four issues.

[348] First, the amount of tuning measured in each of the three markets is materially different from what PPM data disclose for those markets; set-top data measures physical “tuning” of TV sets, while Numeris measures “viewing”.

[349] Second, the amount of tuning to Canadian conventional signals in the set top-measured markets is materially lower than what PPM national data disclose – and since most distant signal viewing is to Canadian conventional signals, this effectively depresses the amount of distant-signal tuning measured.

[350] Third, the two-week period chosen is not representative of the entire broadcast year.

[351] Fourth, Dr. Chipty’s calculation of the share of all viewing accounted for by distant-signal viewing is a small fraction of the distant signal viewing share calculated based on PPM analysis and an even smaller fraction of what the Strategic Report claims.

[352] Mr. Kiefl concludes that participants in Dr. Chipty’s study have unrepresentative viewing patterns when compared to Numeris PPM data. PPM data indicate an audience share for Canadian conventional stations is more than 10 percentage points greater than in Dr. Chipty’s data.

[353] Mr. Kiefl also points out that Dr. Chipty’s estimate of the audience share of Canadian conventional stations differs significantly from the estimate Strategic Inc. reported. While Dr. Chipty’s share of tuning/viewing of distant signals is 4.3 per cent for her two-week sample in 2015, it is 18 per cent in the fall of 2014, according to the estimate of Strategic Inc.

[354] The Collectives argue that, even though Dr. Chipty corrects for the proportion of distant viewing of U.S. conventional signals to make it consistent with Numeris data, she fails to correct for the fact that BDUs’ tuning data also underestimate the amount of distant viewing of Canadian distant signals and overestimate the proportion of viewing of Category B specialty services.

[355] In response, while Dr. Chipty acknowledges that her set-top box data is not nationally representative, she argues that this is not problematic since she is not interested in national viewing habits, but only in the viewing habits of subscribers to large systems (i.e. more than 6,000 subscribers). [234] Moreover, the BDUs argue that not only does Professor Church rely on set-top box data for his analysis, [235] he only uses data from four neighbourhoods in the service area of a single BDU, [236] while Dr. Chipty uses data from the three largest BDUs, serving the three largest cities. According to the BDUs, any concerns with the representativeness of the sample used Dr. Chipty are insignificant compared to the sample used by Professor Church.

[356] The Collectives disagree with Dr. Chipty’s sensitivity analysis using PVRs. They argue that even though there is an increase in the penetration of PVRs, usage of PVRs remains very low. [237]

[357] The Collectives claim that, after correcting Dr. Chipty’s analysis just to reflect the nationally representative share of distant viewing, her proposed rates would rise from $1.20 to $2.28 per subscriber per month. [238]

iii. Analysis

[358] We accept that Dr. Chipty’s method is also sensible. It is conceptually similar to the Board’s original approach to set royalties in 1990 and to the approach by Professor Church. Dr. Chipty applies more adjustments to her benchmark price than Professor Church did. These adjustments are more in line with the adjustments the Board applied in its 1990 Decision. However, we conclude that Dr. Chipty’s analysis suffers from important drawbacks as indicated by the Collectives.

[359] First, her data source, the set-top box data, is inconsistent with other broadcasting industry data sources. Numeris data is currently used much more widely in the industry than set-top box data. However, based on the evidence presented by the parties, we are unsure that either Numeris and set-top box data are capturing TV viewership correctly.

[360] Second, and we believe this to be the biggest issue, the selection of her sample is problematic. For her analysis, Dr. Chipty focuses only on the subscribers of more popular services of three BDUs that provide set-top box data in the three largest cities in two weeks in May 2015, excluding satellite viewers. Dr. Chipty does not provide any evidence to confirm that her particular selection of her sample from the entire set-top box data available to her does not incur any bias in her analysis. On the other hand, the Collectives explain that these selection criteria are very likely to make her data unrepresentative of nation-wide viewership and to introduce bias in her analysis. We agree.

[361] Third, Dr. Chipty’s use of PVR penetration for the substitution-opportunities adjustment does not have enough empirical support. The evidence provided by the Collectives shows that only a small portion of subscribers who have a PVR use it.

[362] Fourth, the inclusion of Canadian Category B specialty services in the benchmark may not be appropriate since their prices are not necessarily set by arm’s-length negotiations in a free market.

[363] Fifth, even though Dr. Chipty attempts to correct for the share of U.S. distant-signal viewing in the set-top box data, she does not correct the share of Canadian distant-signal viewing to reflect the national share. As a result, her share of distant-signals viewership may be underestimated.

[364] To conclude, we believe that while the adjustments Dr. Chipty proposed are necessary to value distant signals, her adjustment factors need to be modified to correct for the foregoing factors.

C. Dr. Wall’s approaches

i. Description

[365] Dr. Wall uses three methods to value distant signals. His main methods (1 and 2) compute directly the market price of distant signals based on the retail prices charged by BDUs to the subscribers. Method 3 is an update to the Board’s original A&E approach.

[366] In Method 1, Dr. Wall selects packages that consist mostly or entirely of time-shifted distant signals. All BDUs party to this proceeding offer such packages. The prices BDUs charge for these packages range from $3 to $10 per month. This translates to a monthly retail price of $0.04 to $0.75 per signal, or $0.30 on average.

[367] Dr. Wall claims that the underlying cost to BDUs for the distant signals included in the time-shifted service packages is primarily the retransmission tariff rate. Based on the tariff rate in effect in 2013 [239] (i.e. $0.98) and given the average number of distant signals received by subscribers in 2013 (i.e. 54), the effective BDU acquisition cost per distant signal is roughly $0.018 ($0.98 ÷ 54). Dr. Wall argues that this rate is significantly lower than the average market wholesale price to the BDUs for a distant signal, and hence, the BDUs profit greatly by selling distant signals. [240]

[368] Using the data from the interrogatory process, Dr. Wall estimates the average price, the average cost, and the average mark-up for both TELUS and Rogers. Then, by applying the implicit cost ratio he calculates to the weighted average per distant signal retail rate, Dr. Wall obtains the implicit value or wholesale cost of distant signals for TELUS and Rogers. In his analysis, Dr. Wall further assumes that TELUS and Rogers are representative of the industry, and as such, his estimates can be generalized to the entire industry. By multiplying these wholesale cost of distant signals by the average number of distant signals received in 2013, Dr. Wall estimates a value of $4.97 per subscriber per month for distant signals received by an average subscriber.

[369] Method 2 considers the implicit prices of distant signals included in basic and extended-basic packages the BDUs offer. These packages typically also include a large number of non-distant signals. Therefore, Dr. Wall asserts that the estimates of distant signal values using this method are not as precise as Method 1.

[370] For Method 2, Dr. Wall uses the basic and extended-basic packages offered by Bell, Rogers, Shaw, and TELUS in 2013 to estimate the value of distant signals. According to Dr. Wall’s calculation, prices for the covered, basic services range from $34 to $43, and the number of included specialty services and over-the-air signals ranges from 57 to 120. This calculation leads to a weighted-average price per basic-service package of $39.31. With an average of 85 specialty services and over-the-air signals included in these packages, the implicit, weighted-average, retail price per service/signal is $0.51. Also, the weighted-average price of extended-basic services is $76.46. The average number of included specialty services and OTA signals in extended basic service packages is 190. This leads to the implicit, weighted-average retail price per service/signal of $0.43 for the extended-basic packages.

[371] By using the wholesale costs of the specialty services included in the packages under consideration (provided by the BDUs) and the mark-ups for the specialty services (estimated as part of Method 1), Dr. Wall applies his estimated mark-up to the aggregate wholesale costs to approximate the retail price or value of the specialty services included in each package. If this calculation results in the price of the specialty services exceeding the package price, he uses a lower mark-up. The residual price attributable to the OTA signals was calculated as the difference between the total package price and the approximated specialty services price. Finally, the implicit, retail price per signal was calculated by dividing the residual price by the number of OTA signals in the corresponding basic or extend basic service package.

[372] This calculation yields a price range for distant signals with the lower-end stemming from the price of the basic service packages and the higher end stemming from the price of the extended basic packages. By multiplying these prices to the cost ratios estimated in Method 1, Dr. Wall calculates the range of wholesale costs of a distant signal to the BDUs. By multiplying these wholesale costs to the number of distant signals subscribers received in 2013 (i.e. 54), Dr. Wall estimates a monthly price range for distant signals for an average subscriber. Dr. Wall shows that his estimated value of $4.97 in Method 1’s falls roughly in the middle of the range he estimates in Method 2.

[373] In Method 3, Dr. Wall provides an alternative estimate of the valuation of distant signals by updating the Board’s 1990 Decision, which was based on the price of the A&E specialty service. Dr. Wall, however, states that this calculation is not a basis on which he recommends that the Board value distant signals today.

[374] The Board’s 1990 proxy analysis used two variables. The first was the wholesale price of A&E, which was $0.25 in 1990. The Board adjusted this rate to $0.15 to account for differences between the adopted proxy service and a typical distant signal. [241] The second was the quantity of distant signals an average subscriber received, which was 4.56 in 1990. Since that time, the price of A&E has increased and the average number of distant signals received by subscribers has also increased.

[375] First, based on the information provided by BDUs, Dr. Wall calculates the average wholesale rate for A&E subscriber for 2009 to 2013. Using this information, he then estimates the annual growth rate in the average wholesale price for A&E from 1990 to 2013 and also for the most recent period of 2009-2013. Dr. Wall also calculates the annual increase in the average number of distant signals received by subscribers in 2009-2013. By summing up the annual growth of the average price of the wholesale rate for A&E in 2009-2013 and the annual growth of the average number of distant signals received by subscribers in the same period, Dr. Wall estimates the expected average annual growth rate of the Board’s 1990 Decision proxy-based tariff model based on recent growth trends of the key elements of that Decision. Dr. Wall then applies this updated proxy growth rates to the most recent tariff in 2013 to obtain the projected retransmission tariff rate for the period 2014 to 2018.

ii. BDU’s comments

[376] The BDUs and their experts made several comments on Dr. Wall’s methods. Dr. Chipty states there are significant problems in Dr. Wall’s methods that make his approaches “fundamentally incapable of recovering a reasonable royalty rate” [242] because of “unwarranted extrapolations and assumptions” [243] he made.

[377] According to Dr. Chipty, in Method 1, Dr. Wall assumes that all BDU subscribers, including those that do not take the theme packs, value distant signal theme packs as much as the subscribers that have chosen to take them. Dr. Chipty further states that Dr. Wall also assumes that all distant signals, even those that are network duplicates, are as valuable as the subset of distant signals that are included in the theme packs. By doing so, Dr. Chipty explains, Dr. Wall assigns the value that some subscribers place on some distant signals to the value that all subscribers place on all distant signals and by doing so substantially overstates the rate. [244] In Dr. Chipty’s opinion, neither of these assumptions is correct.

[378] Another issue Dr. Chipty raises with Dr. Wall’s first two methods is with respect to the structure of the royalties for the retransmission of distant signals and its implication on the number of signals retransmitted and associated fees. As Dr. Chipty explains, the Board’s original decision structured the royalty rate as one flat fee per subscriber receiving at least one distant signal, regardless of how many distant signals the subscriber actually receives. The effect of this rate structure is that BDUs are encouraged to carry as many distant signals as makes sense regardless of where the signal comes from or the marginal value of the signals carried. [245] Because of this flat fee or “all-you-can-eat” rate structure, Dr. Chipty states that calculating a per-signal value and extrapolating that value to all signals is not a sensible approach because it is very difficult to estimate how much of the carriage of distant signals is because of the value of distant signals and how much is a result of the rate structure where adding additional signals does not carry any additional cost. [246]

[379] With respect to Method 2, Dr. Chipty testified that the most serious issue of this approach is that the packages Dr. Wall uses include both local and distant signals and that he incorrectly assigns the same value to both types of signals by assuming that distant signals are as valuable to subscribers as local signals. [247] According to Dr. Chipty, this is inappropriate because while local signal programming is among the most watched of the programming carried by BDUs, distant signal programming is the least watched programming. As a result, Dr. Wall is allocating the much higher value of local signal programming to the much lower value of distant signal programming.

[380] Dr. Chipty also states that Dr. Wall ignores the effect of simultaneous substitution on the value of distant signal programming in his calculation. Dr. Chipty states that merely adjusting for the fact that only a minority of subscribers chose to subscribe to distant signal theme packs brings Dr. Wall’s rate down close to $1 per subscriber per month.

[381] Moreover, the BDUs submit [248] that, while Dr. Wall admitted there was a wide range in the implicit prices per distant signal charged by BDUs for the packages that he examined, and that the differences in prices could be due to a number of factors including different ownership, different business models, different interests, or that the package may be related to something else a BDU offers, [249] he did not examine any of these factors in his report when calculating an average price per signal for the purpose of calculating a tariff rate. He also did not consider whether any of the factors that result in a wide range in the implicit prices require some adjustment before being used to determine the average.

[382] The BDUs further contend that Dr. Wall agreed that subscribers only take theme packs which they believe would be at least as valuable as the price they pay for them, [250] that a majority of subscribers to a large BDU have chosen not to take the distant signal package at a price of $9, [251] and that more customers would be interested in taking the distant signal package if the price were reduced. [252] The BDUs argue that, yet, Dr. Wall did not take into account these factors in his calculations of a derived, average wholesale rate which he applied to the entire subscriber base.

[383] The BDUs also note that Dr. Wall implicitly assumes that subscribers will value a second, third and fourth version of CTV or Global as much as they value the first version. The BDUs point out that this is in contrast with the Board’s 1990 Decision in which, the Board determined that duplicate programming could reduce the value that viewers place on distant signals. [253]

[384] Ms. Blackwell, another expert from the BDUs, testified that the mark-up Dr. Wall uses does not include an amount to cover network costs. According to Ms. Blackwell, Dr. Wall’s methodology is based on the retail price for basic and extended basic packages for some of BDUs, which is attributed to specialty services and OTA signals. That means 100 per cent of the retail price is assumed to be associated with programming services of either specialty or OTA services, and that none of the retail price was attributed to the value of the distribution network. Ms. Blackwell explains that the capital investment in the distribution network is reflected in BDUs’ expenses through the annual depreciation expense and the cost of servicing debt to finance the investment, and that the largest BDUs provided information to the CRTC indicating that these two cost components represent 19.1 per cent and 7 per cent of basic-service revenue. This means that based on these figures, approximately 26 per cent of the total retail price of basic service should be attributed to the distribution network. In Ms. Blackwell’s opinion, revising Dr. Wall’s calculations accordingly suggests a valuation per distant signal of approximately one-third of those estimated by Dr. Wall when the valuation is based on the basic-service package, and that the valuation per distant signal based on the extended-basic package is even lower.

iii. Analysis

[385] We accept that Dr. Wall’s direct measure of the value of distant signals is sensible in principle and could be used if alternative methods did not exist. The main advantage of Dr. Wall’s method is that it is based on revealed preferences. This approach would be superior to a proxy analysis if full information about the market value of distant signals and the cost structure of the BDUs existed. However, we agree with the BDUs that Dr. Wall’s estimate suffers from shortcomings that make it less attractive when compared to the proxy method. Dr. Wall’s estimate suffers from at least the following important difficulties.

[386] First, in Dr. Wall’s first method, there is no reason to believe that the market price of the distant signals obtained from the packages that consist mostly or entirely of time-shifted distant signals represents the average value of all distant signals. For example, Dr. Wall identifies and selects only 12 packages which consist mostly or entirely of time-shifted distant signals. However, distant signals exist in a wide variety of packages including basic packages and extended-basic packages. The distant signals selected by Dr. Wall constitute only a small portion of all distant signals offered by the BDUs.

[387] Also, these packages are add-on packages, where their prices reflect a wide variety of factors set by the business model of the BDUs. This means that the prices of these packages do not necessarily reflect their competitive market value. For example, some BDUs charge the same price for their packages of distant signals even though they include different number of distant signals. [254] The result is that the price per signal varies significantly among these packages. This result shows that for the BDUs and subscribers, the number of available signals is not necessarily directly linked with the value, certainly not with the value of individual signals. Hence, any methodology using the number of distant signals as a simple multiplying factor is problematic.

[388] Second, the price mark-ups Dr. Wall estimates may not show the correct relationship between the cost of production (i.e. the wholesale value of distant signals) and the market value of the final product (i.e. the retail value of distant signals) because there is not always a simple relationship between the price of the output and the cost of inputs. As the BDUs state, many factors affect the price of final goods and services (retail price of distant signals in this case), and the cost of one input (wholesale price of distant signals) is just one of these factors.

[389] This issue is more problematic for intangible assets such as copyrighted works (e.g., distant signals) since by their nature, their market value may not have much relationship with their cost of production: in the case of intangible assets, on the one hand, there are many examples where substantial investment and expenditures were spent to create a product which did not find any market value, and on the other hand, there are many examples where little investment is done to create a new product or service, and because of consumer demand, etc., that product is highly valued in the market. That being said, Dr. Wall did not provide enough justification for the price mark-ups he estimated.

[390] Third, by applying the cost ratios obtained from TELUS and Rogers to all other BDUs, Dr. Wall assumes that all BDUs have similar cost and pricing structures to these two large BDUs. This extrapolation may not hold for all BDUs.

[391] Fourth, similar cautions apply to Dr. Wall’s Method 2, since Method 2’s analysis partially relies on the prices obtained in Method 1.

[392] Fifth, the fact that Method 1 and Method 2 support each other does not increase the validity of either.

[393] Sixth, there are also other assumptions that Dr. Wall made which he did not explain. For example, he did not explain why his estimated mark-up in Method 1 sometimes yields a negative price for distant signals, and what is the basis for changing this mark-up to a lower number in these cases.

[394] Dr. Wall’s proxy approach (Method 3) is an update of the Board’s 1990 Decision. Even though Dr. Wall’s approach is informative, it consists of a narrow analysis based on the price of just one channel. We reject this method for the following reasons.

[395] First, even though A&E was a good proxy for distant signals in 1990, the content and viewership of A&E may be different from those of the distant signals today, and hence, it may not serve as a good proxy anymore. Also, the availability of other services and the diversity of their programming content justify using more than a single service as a proxy.

[396] Second, comparing the current, distant-signal market with that of 1990 reveals that the average number of distant signals per subscribers grew from 4.56 in 1990 to 54 in 2014. The price of A&E increased substantially in the same period. [255] Applying the 1990 methodology yields a price of about $18.5 per subscriber per month for distant signals. This price is not plausible because even though the number of distant signals exploded, there is no reason to believe that subscribers value all distant signals identically. For example, many of these distant signals have very limited viewership.

[397] Third, there is no reason to believe that the value and number of distant signals will increase in 2014-2018 the same way that the value of A&E and the number of distant signals increased in 2009-2013. Dr. Wall did not provide any analysis why the trends of 2009-2013 will continue for 2014-2018.

[398] For the foregoing reasons, we are not prepared to adopt the methodologies proposed by Dr. Wall in the circumstances of this case.

D. The trend-analysis approach

i. Description

[399] The approach was mentioned in the BDUs’ Statement of Case [256] and addressed in the Collectives’ Reply Statement of Case. [257] It was also discussed during final arguments by the BDUs and the Collectives. [258] It was not, however, developed in any expert reports.

[400] This approach starts from the last certified price of $0.98 for 2013 and is based on the BDUs discussion of the history of distant signal prices since 1990. [259] From 1990 to 1994, rates were set by the Board after hearings. From 1995 to 2003, rates were proposed by the Collectives and accepted by the BDUs. From 2004 to 2008, rates were set by settlement after the exchange of interrogatories. The exchange of the interrogatories occurred in 2003 and the settlement occurred in 2005. From 2009 to 2013, rates were set based on a settlement in 2010.

[401] The BDUs’ argument is that, at least up until 2008, the prices were either set by the Board after a hearing, were proposed by the Collectives, or were the result of a settlement after the exchange of (interrogatory) information. The prices of 2009-2013 were certified by the Board after a settlement between the parties. As the BDUs argue, these prices function as market prices.

[402] The BDUs submit that the rate for 2014 should remain unchanged at $0.98, because there was no significant change in the retransmission marketplace since the last certified rate. The BDUs also submit that, as their evidence suggests distant signals will continue losing value in the forthcoming years, the rate should decrease by 2 cents per year for 2015-2018.

ii. Collectives’ comments

[403] In response, [260] the Collectives reject the BDUs’ starting-point-and-marginal-changes (SPMC) approach. The Collectives argue that this approach totally ignores the economic analysis of both the Collectives and the BDUs. The Collectives indicate that Dr. Chipty attempted to determine the value of distant signals, not the change in the value of distant signals since 2013. The Collectives also point out that the parties’ settlement on an amount to compromise their dispute in 2013 is irrelevant for the valuation in this proceeding, particularly that the BDUs have only recently disclosed their internal financial information on which much of the Collectives’ analysis is based.

[404] Moreover, the Collectives argue that BDUs’ proposed approach incorrectly values distant signals. According to the Collectives, looking only at incremental change since the 2013 settlement tariff, which was completed only two years ago, means that broader trends and changes in the value of distant signals are not correctly assessed. To support this argument, the Collectives point to the Board’s Decision in SOCAN Tariff 2.A, 1998, where the Board wrote:

The Board respectfully disagrees with this previous position. It finds this interpretation restricted and limiting. A valuation of this sort should be set in relation to the whole time during which the tariff has existed. Some account should be taken of changes that occurred incrementally over that whole period in the use of music and, as a result, in the value of SOCAN’s licence to broadcasters. The Board also finds that, to date, the account taken of these changes has been insufficient. The reduction in the rate manifests this broader approach. [261]

[405] The Collectives further state that the settlement of a dispute over the price of a good or service cannot be equated with the competitive price of a good or service since the parties may compromise on a price to avoid the risks and costs of adjudication. This is particularly true if the parties do not have equal access to relevant information, as is the case for the Collectives who had no access to any of the BDUs’ highly-confidential business information since 2004.

[406] The Collectives emphasize that it is the first time in 20 years that the royalties paid by Retransmitters for retransmitting distant signals are before the Board for adjudication, and it is also the first time ever that the value of distant signals has been the focus of expert economic testimony informed by so much detailed financial information obtained from the BDUs.

iii. Analysis

[407] The main argument for this approach is that, in the absence of reliable information about the determinants of the value, past values are the best indicators of future values. This argument assumes that self-motivated market players have the best incentive and information to set prices for their own benefit, and that under certain conditions such as symmetric information between the parties and the absence of transaction costs, a settlement price is “close enough” to a market efficient price. This is a typical assumption of efficient markets.

[408] We agree with the BDUs that, in a perfectly-competitive market (for example, no market power for the economic agents, free entry and exit to the market, full information, no externalities, etc.) and in the absence of transaction costs, an agreement between economic agents may be considered a reasonable market price. However, in the absence of these regularity conditions, or in the presence of transaction costs, the market may not exist, or if it exists, it may not generate a fair-and-equitable price. In other words, an agreement may not represent a fair-and-equitable price at all times.

[409] However, the question remains to assess how far the agreement is from a fair-and-equitable price. Based on the degree of the deviation from a fair-and-equitable price, we may accept the agreement as a price or reject it. In this proceeding, if we accept that past agreements have not deviated too far from the market efficient price, this approach provides the best possible market price.

[410] It is not easy to justify this argument based on the evidence in the file. Specifically, this approach suffers from significant shortcomings as follows.

[411] First, this option has a weaker theoretical foundation than the proxy approach put forward by both parties.

[412] Second, this approach relies mainly on previous settlement tariffs, which may not be good indicators of the market price if perfect competition conditions are violated because (i) there could exist large deviations between prices resulting from settlement and market efficient prices for various reasons; and (ii) the likelihood of settlement and its resulting price can be influenced by factors other than market forces. [262]

[413] Third, neither party presented any analysis for this method.

[414] Fourth, there may be an undervaluation of distant signals from 1994 to 2003: while both parties agree that the number, viewership, and value of distant signals to subscribers increased in this period, [263] the distant-signal tariff remained constant at $0.70 per subscriber per month from 1994 to 2003.

[415] In particular, if one party possesses information that the other party does not, the asymmetric information between the parties could lead to an unfair-and-inequitable price. One of the Collectives’ arguments to revise their rates is that they lacked the information they needed to determine with accuracy the value of distant signals until the interrogatory process of this proceeding, and that only the BDUs were aware of the high value of distant signals to subscribers. [264] The Collectives’ position is that due to asymmetric information between the parties and litigation costs, all settlement prices have been below their true values. [265] We agree with the Collectives that due to the lack of information and asymmetric information between the parties, the existence of transaction costs such as litigation costs, etc. previous settlement prices may have been undervalued.

E. The Board’s Approach

[416] Before turning to the approach adopted by the Board, we note that the parties submitted a multitude of ancillary arguments as to how the Board should price distant signals. We have considered these other arguments as well but we find that there is insufficient evidence to properly evaluate the merit of these arguments and quantify the effect their application would have on the royalty rate in this proceeding.

[417] We now turn to the approach we adopt. As do the parties, we begin by explaining our methodology for setting the rate for the year 2014 for Retransmitters that have more than 6,000 subscribers. We will then set the rate for the years 2015 to 2018 as well as for Retransmitters that have 6,000 subscribers or less.

i. Rate for 2014

[418] Our approach is based on an amalgam of Professor Church’s and Dr. Chipty’s proxy approach but makes adjustments to their methodologies and assumptions, based on the record and to take into account certain of the weaknesses in each analysis referred to above. When these adjustments are applied, this option yields the fair and equitable price in our view. This approach follows the steps below to estimate the price of distant signals.

[419] In the first step, a proxy for distant signals is constructed. The proxy has to satisfy two conditions: (i) its price should be set in a competitive market; and (ii) its program content should be similar to that of distant signals.

[420] To adapt the proxies proposed by Professor Church and Dr. Chipty, we construct a proxy which, on the record, adequately resembles the content of Canadian distant signals, and at the same time, ensures that the price of the proxy is the result of a competitive market. We start with the set of the 24 U.S. specialty services proposed by Professor Church and the 47 Canadian category B specialty services proposed by Dr. Chipty.

[421] To ensure that the prices of the proxy are the result of a competitive market, we exclude vertically-integrated Canadian category B specialty services. About 70 per cent (33 out of 47) of the Canadian category B specialty services Dr. Chipty used in her proxy are vertically integrated. [266] We agree with the Collectives that the price of a proxy including these services may not resemble a competitive market price. [267] Certain U.S. specialty services are also excluded because their content is very different from distant signal content. [268] For example, we exclude Playboy TV, although Professor Church included it in his proxy. This leaves 23 U.S. specialty services and 14 Canadian category B specialty services as candidates for inclusion in the proxy.

[422] Using the genre distribution of the specialty services Dr. Chipty provided in her response to the Board Notices 2016-088 and 2016-094, we then construct a proxy that most closely resembles the genre distribution of distant signals. This genre distribution data provided by Dr. Chipty is the disaggregated Table 5 of Dr. Chipty’s report which was already submitted to the Board as part of her expert report. [269] This data presents the distribution of the 12 genres [270] in each of the specialty services she used for her proxy.

[423] We are aware that there are some issues with these data. First, Dr. Chipty did not provide sufficient explanation on how she constructed this distribution from the raw data she received from the BDUs. We cannot replicate her table without further explanation. Second, while the data include the genre distribution of most of the services Dr. Chipty used for the construction of her proxy, the genre distribution of some services she and Professor Church used for their proxies is not available. Third, there are slight differences in the genre distribution of the SD and HD of the same service in Dr. Chipty’s data while they basically provide the same programming content. Fourth, the genre distribution is based on Dr. Chipty’s selected set-top-box data, which is not representative of the viewership in Canada. The fourth issue may not be as critical for this exercise since we assume that the distribution of genres in the specialty services is correct even though the set-top-box data themselves may not be representative. The first and second issues undermine the reliability and usability of the data, but since these data provide the best available information about the programming content of the services, we use it.

[424] This newly constructed proxy includes 20 U.S. specialty services and three Canadian category B specialty services. The difference between the new proxy and Professor Church’s proxy is as follows. First, four U.S. specialty services are removed from Professor Church’s proxy: Big Ten, CBS Sports Network, CNN – International, and Playboy TV. Big Ten, CBS Sports Network, and Playboy TV are not included in Dr. Chipty’s proxy or in the disaggregated Table 5 used for genre distribution. CNN – International is almost a duplication of CNN – Cable News, which is already in the proxy. Removing Big Ten, CBS Sports Network, and CNN – International reduces the share of sports and news genres in the proxy to make the programming content of the proxy more similar to that of distant signals being retransmitted. Second, three non-vertically integrated Canadian category B specialty services are added to the proxy: Bite TV, AUX TV, and BBC Kids. These services are the only non-vertically integrated Canadian category B specialty services which have both payment information and a genre distribution in the file.

[425] In the second step, total payment made by four English-Canadian BDUs to the bundle of these specialty services is calculated. The four BDUs are Bell, Rogers, Shaw, and TELUS. The selection of these four BDUs is based on Professor Church’s approach. [271] This selection is reasonable since these four BDUs provided more complete payment information than other BDUs. Total payment for the proxy amounts to $21,187,297. [272]

[426] In the third step, the total payment for the U.S. specialty services is divided by the total number of subscribers to those four BDUs, which is equivalent to 8,078,000 subscribers. [273] This yields an average price of $2.62 per subscriber per month for the proxy.

[427] In the fourth step, further adjustments are applied to the price of the proxy in order to reflect the price of distant signals. Based on the evidence filed by the parties, we find the following adjustments reasonable.

a. Cost of programming

[428] First, we isolate the cost of programming from the proxy. We agree with Dr. Chipty that the tariff should apply only to the programming portion of the payments. [274] However, we apply a 35 per cent, downward adjustment to the price of the proxy. The 35 per cent includes 25 per cent to exclude the profit and 10 per cent to exclude input and overhead costs.

[429] In her approach, Dr. Chipty proposes a 25 per cent adjustment on the payments of Canadian category B specialty services. This is the average profit margin of Canadian category B specialty services. Due to the lack of a better measure, she applied a 10 per cent adjustment on the payments of U.S. specialty services. [275]

[430] In our opinion, a 10 per cent adjustment on the payments of U.S. specialty services is too low since there is no reason to believe that the profit margin of the U.S. specialty services is lower than that of the Canadian category B specialty services, nor has any party led any evidence to this effect. Therefore, we apply a 25 per cent discount on all services in the proxy to exclude the profit portion of the payments.

[431] Dr. Chipty does not make any adjustment for input and overhead costs due to the lack of information. [276] While we do not have figures for input and overhead costs, we know that these costs exist, and even a conservative estimate would place these at 10 per cent. Therefore, we find that it is reasonable to apply a 10 per cent deduction on the price of the proxy to exclude input and overhead costs. This adjustment reduces the price to $1.70.

b. Market power of specialty services

[432] Second, we adjust for the potential market power of specialty services. As discussed by the BDUs, some of the specialty services included in the proxy target very specific audiences in “niche” markets. Examples of these services include Black Entertainment Television Network (BET), Golf Channel, Military Channel, NFL Network, and Speed Network. [277]

[433] We agree with the BDUs that the prices of these services generally tend to be higher than the more general channels because they can exercise market power in these niche markets. [278] As a result, the total payment for these services may be higher than a more general channel, which leads to a higher price for the proxy. This issue was also raised for the price of A&E in the 1990 Decision. [279]

[434] This potential market power of the proxy requires a downward adjustment on its price. In its 1990 Decision, the Board used a 25 per cent adjustment on the price of A&E for its market power. [280] Similar to the Board’s 1990 Decision, we also apply a 25 per cent downward adjustment to the price of the proxy in the current proceeding.

[435] We note that since the 1990 Decision, two changes may have affected the 25 per cent adjustment that was made to account for the market power. First, there are now many more specialty services in the market compared to 1990. Potential pressure from the competition could put push downward the price of specialty services. This, in itself, could indicate an adjustment lower than 25 per cent. Second however, many of the specialty services included in the proxy are more specialized than A&E. Sports channels are a good example. A higher degree of program specialisation means that these services could target more specific niche markets and could successfully use their market power to push prices upward. In itself, this would lead to an adjustment higher than 25 per cent. Since we do not have evidence in this proceeding to quantify the effect of these two factors on the price of the specialty services, we will assume that they both cancel out, and use the same 25 per cent adjustment factor considered by the Board in its 1990 Decision. This adjustment reduces the price from $1.70 to $1.28.

[436] The research and information presented in Board Notice 2016-088 for the parties’ respective comments and submissions, although not used to determine this 25 per cent adjustment, provides a comfort level as to the reasonableness of the adjustment, which was used previously by the Board in a similar situation.

c. Program substitutability

[437] Third, we adjust for program substitutability. As Dr. Chipty explains, it is possible to substitute distant signal programs with other opportunities to view the same programs. Substitution opportunities may become available through the use of the PVR, or through VOD, OTT services, and new features that the BDUs offer to their subscribers. [281] Dr. Chipty shows that subscribers use these new features to substitute distant signal programs more than specialty services programs. [282]

[438] Moreover, the time-shifting feature of distant signals will not be as valuable for the subscribers due to these new features. For example, if the user has the ability to record the original broadcast of the local signal using a PVR and listens to it at any time, this viewer will not find any value in having access to the time-shifted distant signals (even though the value of the original local signal is unaltered). The copy made with the PVR will play that time-shifting role. Therefore, it is reasonable to accept that these opportunities decrease the value of distant signals more than that of the value of specialty services.

[439] It should be noted that the Collectives argued that a distant signal recorded by a PVR remains distant when viewed from a PVR. [283] This may be true, but is beside the point. PVRs can be used to record any types of programs, including local signals, distant signals, and specialty services. The adjustment we do is only in respect of the recording of a local signal using a PVR and the effect on the time-shifting value of distant signals.

[440] Although Dr. Chipty recommends these adjustments, the magnitude of the effects of VOD, OTT services, and new features that the BDUs offer cannot be quantified based on the evidence filed initially by the parties. Moreover, while Dr. Chipty uses PVR penetration of 57.6 per cent for this adjustment, [284] we believe that actual PVR use is a better measure for this adjustment than the PVR penetration. Dr. Chipty presents PVR use of 14.5 per cent [285] and Mr. Kiefl presents PVR use of 7.7 per cent. [286]

[441] For the purpose of this adjustment we use the new data which were presented to the parties for their comments in Board Notices 2016-088 and 2016-094. We believe that it is reasonable to apply a downward adjustment on the price of the proxy based on PVR use and viewing of OTT services using the data presented to the parties and on which they were afforded an opportunity to provide comments. [287] The new data suggest PVR use of 7.5 per cent and OTT services viewing of 9 per cent. The sum of these two amounts is 16.5 per cent.

[442] However, we believe that a 16.5 per cent adjustment would overestimate the true impact of alternative viewing opportunities for two reasons. First, local signals are only a proportion of all programming being recorded on the PVR, and a reduction should be done to the adjustment to take this into account. Second, OTT services are not perfect substitutes for distant signals. Therefore, it would be incorrect to assume that all usage of such services substitutes for distant-signal viewing.

[443] While we do not have evidence on the level of alternative viewing opportunities that are actually taking place, we find it reasonable to estimate the total effect of all such services by decreasing the 16.5 per cent figure by one half, resulting in an adjustment factor of 8.25 per cent. In our view, this figure better reflects the fact that the substitution rate is lower than the usage rate of these services. This adjustment reduces the price from $1.28 to $1.17.

[444] We also examined a potential adjustment for the CRTC’s “Let’s talk TV” effect. However, as discussed by the parties, [288] although this CRTC policy has some possible negative effects on the viewership and value of distant signals, the magnitude of the effects is unclear at this time and the effects will manifest themselves only in the longer run. As a result, we do not consider such effects in the calculation of the price at this time.

[445] In our view, this approach yields a fair-and-equitable-price for distant signals. First, the content of the proxy is more similar to the content of distant signals. Second, more precise rates are used to isolate the cost of copyrighted content. Third, the price of the proxy is adjusted for potential market power. Fourth, more precise rates are used to quantify the impact of substitution opportunities.

[446] Other adjustments may have been appropriate as well, but we were unable to obtain an estimate for these, due to the lack of reliable evidence. We note them for the sake of completeness.

[447] First, due to the lack of data, no adjustment on the price of the proxy can be done based on the relative viewing of distant signals to the specialty channels contained in the proxy. As explained by Dr. Chipty, the price of the proxy should be adjusted by the ratio of viewing attributed to distant signals relative to the viewing attributed to the selected proxy. Since the parties did not provide reliable viewing data for each specialty service, it is not possible to make that adjustment.

[448] Second, one remaining issue with respect to the programming content is whether similar genres on the specialty services and distant signals have similar values to subscribers. One may argue that, for example, a comedy program on a channel specialized on comedy has a higher value to subscribers than a comedy program on distant signals because of the supposedly better quality of the programs on the specialty channels; or vice versa, news on distant signals may have a higher value to subscribers than specialty services such as CNN and FOX News since they provide local news. However, we do not have any evidence to quantify this difference and its impact on the value. In addition, and contrary to the adjustment we have made for the input and overhead costs, we do not even have evidence on the whether the adjustment should be positive or negative. Hence, we do not take into account this effect.

[449] Third, although both Professor Church and Dr. Chipty testified that the price of their proxy is a competitive price, there is no viable economic analysis by the parties to prove that the prices of their selected specialty services are offered competitively. Reviewing the payments of the specialty services by the BDUs does not suggest that there is a competitive market for the U.S. specialty services in Canada. For example, there are large disparities among the BDUs on payments for the same service. [289] Assuming a competitive market scenario, payments for the same service from different BDUs should roughly be the same.

[450] Regarding the potential market power of the BDUs and the Collectives raised by Professor Church and Dr. Chipty, even though we do not rule out the possibility of exercising bargaining power by either party based on the evidence presented in this proceeding, we do not know which party has greater bargaining power. As such, similar to Professor Church and Dr. Chipty, we do not consider this possibility in the calculation of the rate. It should be noted that this possibility of exercising bargaining power by the parties should not be confused with the earlier discussion of market power of specialty services when they target niche markets.

[451] Hence, we believe that a rate of $1.17 per subscriber per month would be reasonable for distant signals in 2014. However, as was decided above, we are not prepared to approve a tariff in excess of the amounts initially proposed by the Collectives as published in the Canada Gazette. This implies that for the year 2014, we approve a rate of $1.06.

ii. Rates for 2015-2018

[452] To calculate the price of distant signals for 2015-2018, we consider two factors: inflation and expected change in viewership. On the one hand, inflation is expected to put a slight upward pressure on the prices from 2014 to 2018. On the other hand, based on the evidence presented in this hearing, it is expected that decrease in viewership put a slight downward pressure on the prices. We agree with the BDUs that the viewership of distant signals will decrease as a result of more widespread use of alternative viewing opportunities, and possibly, CRTC’s “Let’s talk TV” effect. Although we are not sure of the magnitude of the decrease in distant signal viewership, it is reasonable to assume that the increase in inflation and decrease in viewership will possibly be in the same order of magnitude and will cancel out.

[453] As a result, we believe that an annual price of $1.17 per subscriber per month would be reasonable for distant signals for the years 2015-2018. However, as the rate initially proposed by the Collectives for 2015 is $1.14, this is the rate we approve. For the years 2016 to 2018, for which the rates proposed initially were higher than $1.17, we approve the latter.

iii. Rates for Retransmitters with no more than 6,000 subscribers

[454] Retransmitters are classified into small systems and large systems. The Act requires the establishment of a preferential rate for small-retransmission systems. [290] A small-retransmission system is one retransmitting a signal to no more than 2,000 premises in the same community and that meets other conditions as per the “Small Retransmission System” Regulations. [291] As such, in its 1990 Decision, the Board set a flat royalty of $100 per annum for small Retransmitters. [292] The rate for small Retransmitters is unchanged since 1990 and will remain so in this matter, on consent of the parties.

[455] A large retransmission system is a Retransmitter that is not small. In the 1990 Decision, the Board set a rate of $0.70 per subscriber per month for Retransmitters with more than 6,000 subscribers (premises). [293] However, the Board indicated that it “does not believe that the provisions of the Act prevent it from setting different rates within the class of large systems, provided that such rates are fair and equitable.” [294] Accordingly, the Board created subclasses of Retransmitters between 1,001 and 6,000 subscribers. To explain this decision, the Board wrote: “a fair and equitable tariff may treat Retransmitters in different circumstances differently […] As systems become smaller, they tend to have higher average fixed and operating costs and to charge higher monthly fees.” [295] Subsequently, the Board set rates from $0.20 for Retransmitters with up to 1,500 to $0.65 for Retransmitters with 5,501-6,000 subscribers. [296]

[456] In the most recent Retransmission Decision (2013 Decision), the Board certified again varying rates based on the number of subscribers. However, in that decision, the rates for 2013 decreased from the top $0.98 for the Retransmitters with more than 6,000 subscribers to $0.41 for the Retransmitters with up to 1,500 subscribers. The new rates were the result of a settlement among the parties. [297]

[457] The Collectives’ proposed rates for 2014-2018 (based on their Statement of Case) are discounted by $0.05 from the top $2 for Retransmitters with more than 6,000 subscribers for 2014 ($2.38 for 2018) to $1.50 for Retransmitters with up to 1,500 subscribers for 2014 ($1.88 for 2018). [298] The Collectives’ proposed rates for 2014-2018 impose a bigger disconnect between small Retransmitters that pay a flat fee of $100 per year and slightly larger Retransmitters that must pay considerably higher fees under their proposed rates than that of the 2013 Decision. As such, we believe that the Collectives’ rates are inconsistent with both the 1990 and 2013 decisions that certified a more gradual increase in rates from small Retransmitters to large Retransmitters. To correct this issue, we use the same $0.05 or $0.06 per each subclass increments of the 2013 Decision for 2014-2018.

[458] The Collectives claim that they had inadvertently omitted the rates for the Retransmitters with no more than 2,000 subscribers in their proposed rates as published in the Canada Gazette, but then added the rates for these Retransmitters in their Statement of Case. [299] The Board has always approved these rates in the past. We do the same in the present instance as some Retransmitters with no more than 2,000 subscribers may nevertheless not qualify as a small retransmission system as per the “Small Retransmission Systems” Regulations. [300]

iv. Discounts

[459] As mentioned above, no changes have been proposed in respect of certain long-standing discounts such as for the BDU’s size, for institutional consumers and for Francophone markets. We approve these discounts as initially proposed by the Collectives.

F. Rates approved and total royalties generated

[460] The following table shows the rates that we are approving for all the years under examination and all sizes of Retransmitters.

Table 3: Monthly Rate for each premises receiving one or more distant signals (in dollars), 2014-2018

Table 3: Monthly Rate for each premises receiving one or more distant signals (in dollars), 2014-2018

Number of premises	2014	2015	2016-2018
Up to 1,500	0.49	0.57	0.60
1,501 - 2,000	0.54	0.62	0.65
2,001 - 2,500	0.60	0.68	0.71
2,501 - 3,000	0.66	0.74	0.77
3,001 - 3,500	0.71	0.79	0.82
3,501 - 4,000	0.77	0.85	0.88
4,001 - 4,500	0.83	0.91	0.94
4,501 - 5,000	0.89	0.97	1.00
5,001 - 5,500	0.94	1.02	1.05
5,501 - 6,000	1.00	1.08	1.11
6,000+	1.06	1.14	1.17

[461] We estimate annual royalties at about $123 million for 2014, $127 million for 2015, and $130 million for 2016-2018. These estimates are based on the estimate of $110 million the Board did in its 2013 Decision for the year 2013, to which is applied the percentage increase of the rates we approve for the years 2014 to 2018.

VIII. ALLOCATION

[462] As mentioned earlier in this decision, the Collectives confirmed to the Board on January 31, 2019, that an agreement on allocation among all of them had been reached. Accordingly, the Board approves the following portion of the royalties that is to be paid to each Collective Societies:

For the years 2014-2015

For the years 2014-2015

Collectives	Allocation (%)
Border Broadcasters, Inc. (BBI)	0.96
Canadian Broadcasters Rights Agency (CBRA)	13.50
Copyright Collective of Canada (CCC)	53.38
Canadian Retransmission Collective (CRC)	14.85
Canadian Retransmission Right Association (CRRA)	9.76
Direct Response Television Collective Inc. (DRTVC)	0.70
FWS Joint Sports Claimants Inc. (FWS)	3.25
Major League Baseball Collective of Canada, Inc. (MLB)	0.80
Society of Composers, Authors and Music Publishers of Canada (SOCAN)	2.80

For the years 2016-2018

For the years 2016-2018

Collectives	Allocation (%)
Border Broadcasters, Inc. (BBI)	1.13
Canadian Broadcasters Rights Agency (CBRA)	10.72
Copyright Collective of Canada (CCC)	54.13
Canadian Retransmission Collective (CRC)	16.10
Canadian Retransmission Right Association (CRRA)	10.65
Direct Response Television Collective Inc. (DRTVC)	0.64
FWS Joint Sports Claimants Inc. (FWS)	3.68
Major League Baseball Collective of Canada, Inc. (MLB)	0.15
Society of Composers, Authors and Music Publishers of Canada (SOCAN)	2.80

IX. TARIFF WORDING

[463] On May 9, 2019, in Notice CB-CDA 2019-026, the Board consulted the parties on a Draft Tariff. Responses were provided on May 24, 2019, and replies were received on May 31, 2019.

[464] Section 6 of the Draft Tariff would have required a DTH to pay royalties for “each premises” rather than each premises that receives a distant signal. The BDUs submit that it is incorrect to require these systems to pay royalties for “each premises” since it is technically possible for a DTH BDU to only deliver local signals to a subscriber. Both the BDUs and the Collectives agree that section 6 should be deleted and the royalty provisions set out in section 7 should apply to both cable systems and DTH. We agree. The approved tariff is adjusted accordingly.

[465] Both parties also commented on the transitional provisions of the Draft Tariff. They submit that interest should only be payable on “additional royalties”, not on “all payments made by retransmitters” as expressed in the Draft Tariff, and that interest should only be payable for the period that commences with the date the additional royalties are payable pursuant to the tariff and ends on the date that the additional royalties are actually paid.

[466] We agree. The approved tariff reflects this submission. Having interest accrue on additional royalties rather than all payments makes sense: Retransmitters have an ongoing relationship with the Retransmission Collectives. They have been paying continuously since January 2014 based on the tariff which expired in December 2013.

[467] The Collectives also reached an agreement with respect to the inter-collective re-allocation of previously paid royalties for the 2016-2018 tariff years, as well as the inter-collective re-allocation of addition royalties for the 2016-2018 tariff years, as well as the interest to be associated with these re-allocation payments. The approved tariff reflects this agreement.

[468] The Collectives further propose to include a table of interest rates as opposed to interest factors. According to the Collectives, the proposed amendments are required in order for the parties to be able to accurately calculate the interest that will be owed by different Retransmitters. This approach is necessary because some of the Retransmitters have already paid the additional royalties owing in order to “stop the clock” on interest, whereas others have chosen not to do so and interest is, therefore, continuing to accrue on those unpaid additional royalties. In reply, the BDUs agree with this approach.

[469] Historically, the Board used interest factors rather than interest rates. In 2011 decision with regard to CBC Radio, [301] the Board stated that the practice of using interest factors should be generalized. The Board also decided to continue to derive the interest factors using month-end Bank Rates. The Board reiterated this principle in a more recent decision in respect of Access Copyright. [302]

[470] We believe that while the use of interest factors can simplify the calculation of interest for the parties, both interest factors and interest rates should in theory produce the same results. This means, no party will be affected as a result of switching from interest factors to interest rates. The tariff is approved according to this proposal.

[471] Other minor changes to the Draft Tariff (including the forms) are made in the approved tariff, as per the parties’ comments.

Gilles McDougall

Secretary General