Reply of the Commissioner of Competition before the Canadian Radio-television and Telecommunications Commission

Telecom Notice of Consultation CRTC 2013-685
Wholesale mobile wireless roaming in Canada — Unjust discrimination/undue preference

February 10, 2014

I. Overview

  1. On January 29, 2014, the Commissioner of Competition (the "Commissioner") intervened in this proceeding and the Competition Bureau (the "Bureau") made submissions to the Canadian Radio‑television and Telecommunications Commission (the "CRTC") in response to its Call for Comments (the "Bureau Submission").Footnote 1 The Commissioner now files this reply by the Bureau pursuant to section 125 of the Competition Act (the "Reply Submission").Footnote 2
  2. The Bureau, as an independent law enforcement agency, ensures that Canadian businesses and consumers prosper in a competitive and innovative marketplace. Headed by the Commissioner, the Bureau is responsible for administering and enforcing legislation, including the Competition Act and related statutes, throughout the Canadian economy. The purpose of the Competition Act is to maintain and encourage competition in Canada to, among other things, provide consumers with competitive prices and product choices.Footnote 3
  3. The Bureau has reviewed and considered the submissions filed by the parties and interveners in this proceeding and has found no basis to change or modify any of the statements or recommendations made in the Bureau Submission. In particular, the Bureau maintains its assessment that the following factors are present in Canadian mobile wireless markets:
    1. incumbent mobile wireless service providers ("incumbents") have retail market power;
    2. incumbents have an incentive to protect their retail market power by imposing supra‑competitive terms in roaming agreements with entrant service providers ("entrants"); and
    3. entrants currently face significant disadvantages as a result of the supra‑competitive terms in their roaming agreements with incumbents.
  4. These factors render the incumbents’ discrimination against entrants in respect of roaming rates unjust. Therefore, it remains both necessary and appropriate for the CRTC to intervene and remove the incentive for incumbents to use roaming agreements as a strategic tool to eliminate or reduce the competitive pressure that would otherwise be associated with entry into and expansion in Canadian mobile wireless markets. If left unchecked, these roaming relationships will continue to adversely affect economic welfare, the Canadian economy and the sustainability of competition in Canadian mobile wireless markets, and will deprive consumers of the benefits associated with effective competition, including lower prices, higher quality service and greater innovation.
  5. This Reply Submission identifies evidence filed by entrants that supports the Bureau’s assessment, addresses claims made by incumbent service providers that Canadian mobile wireless markets are competitive, and provides additional perspective on appropriate remedies.

II. Wireless roaming agreements are adversely affecting entrant service providers

  1. As set out in the Bureau Submission, incumbents are incented to implement strategies that eliminate (either partially or fully) competitive pressure that would otherwise be created by the establishment of entrants.Footnote 4 These strategic actions by incumbents have likely resulted in, or will likely result in, entrants charging higher prices at the retail level, or providing less attractive non‑price elements of their services at the retail level, than they would if their roaming agreements were not affected by the strategic actions of the incumbents.
  2. The evidence filed by entrants, including Globalive Wireless Management Corp. ("WIND"),Footnote 5 Québecor Média Inc. ("Québecor")Footnote 6 and Bragg Communications Inc. ("Eastlink"),Footnote 7 provide evidence from the marketplace that roaming agreements are adversely affecting entrants.
  3. WIND and Eastlink provide detailed evidence on how high roaming rates negatively affect their current product offerings.
  4. WIND’s evidence is that its domestic roaming arrangements have caused it to establish "pricing zones" where its customers pay lower rates for service in geographic areas where WIND operates its own network and higher rates in geographic areas where service is only obtainable through a roaming arrangement.Footnote 8 The existence of high price zones attributable to roaming agreements with an incumbent very likely indicates that WIND is charging higher prices to its customers in those geographic areas where service is only obtainable through a roaming arrangement than it would otherwise charge if it were able to access the incumbent’s network on competitive prices and terms.
  5. Eastlink’s evidence is that it also priced its services using zones, but that these pricing zones resulted in poor customer attraction. Eastlink’s subscriber growth improved when it abandoned the zone approach. Eastlink chose to abandon the zone approach despite incurring a financial loss on every minute of voice service or byte of data service in roaming areas.Footnote 9 The available evidence suggests that the roaming agreements have placed Eastlink in a position where its only option for meaningful subscriber attraction is to incur financial losses whenever its customers travel outside of its own network coverage area.Footnote 10
  6. Eastlink’s evidence demonstrates that this impact is likely to become more significant in the future. The Eastlink Submission indicates that high domestic roaming rates have caused it to reduce the quality of the services it offers to its consumers. In particular, Eastlink notes that it has not enabled LTE roaming for its customers, even though it is able to do so, due to the significant risk that those customers will incur very high roaming charges from the enhanced data usage that LTE enables, and that these charges could have significant negative consequences for Eastlink’s profitability. Footnote 11 Wireless customers in the areas served by Eastlink, therefore, are faced with a choice between an incumbent, which can offer LTE access across substantially all of Canada, and Eastlink, which can only offer LTE coverage in its own network area.Footnote 12
  7. As customers become ever hungrier for data‑intensive services, the absence of LTE roaming is likely to represent a significant difference between Eastlink and incumbents — a difference that will diminish Eastlink’s ability to compete effectively against incumbents not affected by high wholesale roaming costs.
  8. In the Bureau’s experience, and consistent with economic theory,Footnote 13 non‑price factors such as these can be significant drivers of market outcomes. Here, if the effect is sufficient, these factors could result in entrants either becoming "niche" players with little competitive effect on incumbents or, in the worst case, simply exiting Canadian mobile wireless markets.
  9. In summary, entrants are caught between a rock and a hard place as a direct result of their roaming arrangements with incumbents. Entrants can either choose to absorb their high roaming costs, and incur significant financial loss, or they can pass along some portion of these costs, and have less success at attracting customers. Neither of these strategies appears to result in the creation and maintenance of sustainable, effective competition in Canadian mobile wireless markets. As such, these arrangements do not further the objectives of Canadian telecommunications policy set out in section 7 of the Telecommunications Act.Footnote 14

III. Analysis should focus on whether roaming rates are supra‑competitive

  1. Consistent with section 7 of the Telecommunications Act, the CRTC should focus its analysis on whether the domestic roaming rates incumbents charge to entrants are at supra‑competitive levels.
  2. Some participants in this proceeding assert that it is not appropriate to compare purely domestic wholesale roaming rates with international wholesale roaming rates or domestic retail prices. However, this argument misses the point.Footnote 15 The correct question from a competition perspective is not whether domestic wholesale roaming rates exceed international or retail rates, but rather whether domestic wholesale roaming rates exceed those that would be charged in a market where incumbents do not have the incentive and ability to forestall the growth of their rivals.
  3. The Bureau does not have access to the confidential record in this matter and, therefore, does not have the necessary information to provide the CRTC with detailed comments on the extent to which wholesale roaming prices charged to entrants are supra‑competitive. However, in the Bureau’s view, the evidence on the public record regarding the high roaming rates that entrants face is both compelling and consistent with the incumbents raising their rivals’ costs.
  4. The Bureau accordingly asks the CRTC to examine the confidential information filed by the parties in this proceeding for further evidence respecting this effect.

IV. Incumbents’ submissions do not affect Bureau’s market power assessment

  1. In their submissions, the incumbents have denied having market power. The Bureau does not find any of the incumbents’ submissions on market power persuasive. Rather, the Bureau remains of the view that the incumbents have market power in retail mobile wireless markets.Footnote 16
  2. The Bureau provides the following specific response to the submissions on market power made by the incumbents:
    1. Certain incumbents use concentration‑based measures to claim that Canadian mobile wireless markets are competitive or, in the very least, comparable to certain international jurisdictions.Footnote 17 While measures of market concentration can, in certain circumstances, be helpful indirect indicators of market power,Footnote 18 the presence or absence of high market shares (or a high or low number of "competitors" in a market) is of little relevance in this current proceeding. If one or more of these "competitors" is effectively marginalized as a result of incumbents’ strategic actions, then this firm is not, in fact, an effective competitor, and its presence or absence in a market will have little effect on market outcomes.Footnote 19 Including such firms in the calculation of market concentration indicators tends to overestimate the extent to which these markets are competitive.
    2. The incumbents also present marketplace statistics as evidence that Canadian mobile wireless markets are becoming more competitive. Incumbents put forward these statistics to demonstrate that average mobile wireless prices in Canada have fallen in recent yearsFootnote 20 and that entrants have been successful in attracting customers to their services.Footnote 21 However, even if these data are accurate and these conclusions are correct, the incumbents have not measured the entrants’ progress against the proper counterfactual. When considering the effect of the incumbents’ strategic behaviour, the analysis should be measured against a situation where no such strategic behaviour occurs (i.e., a competitive outcome).Footnote 22 What is relevant is not whether average prices have fallen period‑to‑period, or whether entrants have attracted new subscribers year‑over‑year, but rather how prices and consumer switching have evolved as measured against the counterfactual situation where entrants are able to access roaming services on competitive terms. In the Bureau’s view, it is likely that average mobile wireless prices in Canada would have fallen further, and entrants would have attracted more subscribers than they have in the current situation where entrants cannot access roaming services from incumbents on competitive terms.
    3. The Bureau also notes that several incumbents attempt to support their positions regarding market power by relying on a recent report prepared by Jeffrey Church and Andrew Wilkins (the "C—W Report").Footnote 23 As noted at paragraphs 11‑13 of the Bureau Submission, the limitations in the C‑W Report are sufficiently significant for the CRTC to disregard the conclusions that the C‑W Report reaches about the competitiveness of Canadian mobile wireless markets.
  3. The Bureau continues to be of the view that mobile wireless markets in Canada are characterized by high concentration and very high barriers to entry and expansion. Furthermore, Canadian mobile wireless markets are characterized by other factors that, when combined with high concentration and very high barriers to entry and expansion, create a risk of coordinated interaction in these markets.Footnote 24 Given these factors, the Bureau’s view is that incumbents have market power in Canadian retail mobile wireless markets.

V. Remedy

  1. As explained in the Bureau Submission, an effective remedy is one that would allow entrants to obtain roaming access on terms that do not reflect the strategic actions present in the current situation. In other words, the CRTC’s task in arriving at an appropriate remedy should be to identify domestic roaming rates that are not supra‑competitive, and prohibit incumbent service providers from imposing rates above that level.
  2. Roaming rates, in this context, will not necessarily coincide with either the roaming rates found in incumbents’ roaming agreements with international service providers or in incumbents’ retail rates.
  3. In determining this competitive level, the CRTC should consider the costs that incumbents must incur to provide roaming services to an entrant, noting that this rate may be different for some entrants than others. Furthermore, an appropriate remedy would not simply set rates at one particular level, but would allow some flexibility for incumbents and entrants to negotiate lower rates if circumstances warrant, so as to recognize that incumbents may be able to offer lower rates if they gain compensation through other legitimate (but perhaps yet unimagined) means.
  4. Several incumbents have made claims that so‑called "bilateral" roaming agreements with U.S. service providers, where both parties agree to roam on each others’ networks, create different pricing incentives than "one‑way" agreements, where only one party gains roaming access. In considering this issue, the CRTC should recognize WIND’s evidence that it was able to access considerably cheaper roaming rates with U.S. service providers compared to the rates it pays for domestic roaming with Canadian incumbents.Footnote 25 While the Bureau is not aware of the particulars of these arrangements, roaming agreements between WIND and a U.S. service provider may be effectively one‑way agreements, whereby WIND gains access to the U.S. provider’s network, but not vice‑versa.Footnote 26
  5. The Bureau submits that the CRTC should not consider the mere existence of an arbitration process to settle disputes between incumbents and entrants regarding the terms of their roaming agreements when assessing the need for remedies in this proceeding.Footnote 27 The mere existence of an arbitration process cannot be used as justification to preserve the ability of incumbents to use roaming agreements to diminish the competitive impact of entrants. From an economic perspective, there may be legitimate disincentives to using the arbitration system, such as the resources and time necessary to bring a dispute to an end, and the uncertainty regarding the outcomes of the process.Footnote 28 The failure of parties to resort to arbitration may rather be evidence that the mechanism is insufficient to address supra‑competitive roaming rates.Footnote 29
  6. Further, the availability of arbitration in respect of disputes that have arisen does not mean that the CRTC should not take steps to ensure a competitive market structure. In addition to being costly, arbitration only addresses a single dispute. Entrants disputing multiple arrangements could be faced with bringing multiple arbitrations and could be disadvantaged vis‑à‑vis well capitalized incumbents.
  7. The Bureau again states its position that an appropriate remedy for supra‑competitive roaming rates is one that is sufficiently effective to ensure that competitive rates are reached. If, in deciding upon an appropriate remedy, the CRTC is faced with a choice between a remedy that goes farther than is strictly necessary to eliminate the unfair discrimination or undue preference and a remedy that does not go far enough, the Bureau would prefer the former remedy.Footnote 30

VI. Conclusion

  1. The Bureau is pleased to provide this Reply Submission. As noted in the Bureau Submission, effectively encouraging the competitive provision of roaming access to entrants will enable Canadians to enjoy the benefits of greater competitive forces in mobile wireless markets, including lower prices, higher quality service and greater innovation.
Date modified: