August 13, 2020
On this page:
- Executive Summary
- The Bureau's Approach for this Submission
- Contextual Considerations for this Proceeding
- Economic Framework and Competition Considerations for this Proceeding
- Assessment of Rate-Setting Methodologies
- Appendix A: International Comparators
- Appendix B: Summary of Bureau Advice
- The Commissioner of Competition (Commissioner) wishes to intervene in the proceeding initiated by Telecom Notice of Consultation CRTC 2020‑131 (Notice of Consultation), published by the Canadian Radio‑television and Telecommunications Commission (CRTC) on 24 April 2020.
- The Commissioner submits these comments of the Competition Bureau (Bureau) in response to the Notice of Consultation.Footnote 1 The Bureau makes these comments further to the Commissioner's mandate to act as an advocate for the benefits of a competitive marketplace.Footnote 2
- In the Notice of Consultation, the CRTC poses a series of questions aimed at "… identify[ing] issues with the approach to rate setting for wholesale telecommunications services …" in Canada. In this submission, the Bureau highlights the potential effects that particular wholesale rate-setting methodologies may have on competition.
- Accordingly, the Bureau directs its comments principally to the questions posed in paragraphs 18 and 19 of the Notice of Consultation. Some of the Bureau's comments may also be relevant to other questions specified in the Notice of Consultation. In other stages of this proceeding, the Bureau may comment on the responses of other parties to any question posed in the Notice of Consultation insofar as those responses relate to, or may have an effect on, competition.
II. Executive Summary
- Wholesale access promotes competition in respect of vital telecommunications services across Canada.Footnote 3 Wholesale access supports competition by enabling network owners to access support structures and civil infrastructure in order to expand and modernize their networks, and by allowing wholesale-based competitors to compete with network owners.Footnote 4 Greater competition means lower prices, increased consumer choice, and higher levels of innovation across the Canadian telecommunications industry.
- Administering a wholesale access regime is complex. Regulators must create, manage, and update a multi-faceted regulatory approach that enables service providers to compete vigorously for customers while, at the same time, minimizing the associated costs of such access on the health and quality of Canada's telecommunications networks.
- This proceeding is a unique opportunity to perfect that balance. While rate setting is not the only factor governing the success of wholesale access, the strong financial incentives that it creates necessarily places it at the heart of the regime. By appropriately tailoring a rate-setting approach, the CRTC can position both Canada's wholesale access regimes and Canada's telecommunications networks for continued success in the decades to come.
- This submission analyzes the effect that the five rate-setting methodologies identified in the Notice of Consultation have on competition in the Canadian telecommunications industry. In performing this analysis, the Bureau is informed by economic and competition perspectives, complemented by diverse viewpoints gleaned from conversations with industry participants, foreign regulators, and outside advisors.
- At this point in the proceeding, the Bureau's main substantive position is that rate setting should consciously support the very competition that wholesale access is intended to foster. No particular methodology identified at this point in the proceeding is uniquely qualified to achieve that goal, as each must be considered in terms of the particular product or service to which it will be applied.
- However, methodologies best placed to effectively promote competition are those that:
- maximize economic efficiency,
- recognize and account for the economic relationship between wholesale and related retail rates,
- acknowledge and minimize information asymmetries, and
- avoid undue rate differences between network types.
- Accordingly, in addition to presenting its initial analysis in this submission, the Bureau is ready to support the design and development of more detailed regulatory proposals in future stages of this proceeding, and in follow-on proceedings as appropriate.
III. The Bureau's Approach for this Submission
- In the recent past, the Bureau has analyzed wholesale rate setting in both the Canadian broadband internet and mobile wireless industries. To supplement those perspectives in preparing this submission, the Bureau has further consulted with domestic industry participants, foreign communications regulators, and outside advisors with experience in the area of wholesale rate setting. This section of the Bureau's submission discusses these efforts in order to set out the Bureau's perspective in this proceeding.
Recent Bureau Work in Respect of Wholesale Rate Setting
- In August 2019, the Bureau concluded its Broadband Market Study (Broadband Study).Footnote 5 The Broadband Study was a year-long effort undertaken by the Bureau to evaluate the state of competition in Canada's broadband industry. In conducting the Broadband Study, the Bureau surveyed consumers and industry participants with the goal of better understanding how internet service providers in Canada compete for consumers' business.
- While the Broadband Study did not analyze wholesale rate-setting methodologies, it did summarize a wide range of views on the practice of wholesale rate setting in Canada. In particular, the Broadband Study noted:
"There is wide debate in the industry regarding whether or not wholesale rates are set at appropriate levels. Facilities-based competitors claim that wholesale-based competitors gain access to networks at rates that are below the actual costs of the facilities-based competitor, which has significant negative effects on investment incentives. At the same time, some wholesale-based competitors point to examples where a facilities-based competitor has set retail prices at levels that are less than the regulated fees that a wholesale-based competitor would have to pay in order to offer those same services to that customer. This is a 'Goldilocks' problem—set rates too low, and facilities-based competitors are less likely to invest; set rates too high, and wholesale-based competitors are not able to bring pricing discipline to the marketplace."Footnote 6 (internal citations omitted)
- From this perspective, the Bureau agrees with the CRTC's characterization of the wholesale rate-setting problem as expressed in the Notice of Consultation.Footnote 7 That is, setting rates that both promote vigorous competition and maintain investment in Canada's world-class telecommunications networks is challenging but critically important to our digital future.Footnote 8
- The Bureau more explicitly analyzed wholesale rate-setting methodologies in its Further Comments submission in the proceeding initiated by Telecom Notice of Consultation 2019-57, Review of mobile wireless services.Footnote 9 In that submission, the Bureau recommended that MVNO wholesale access rates be set using commercial negotiation backstopped by CRTC arbitration. That recommendation was made in the context of the facts of that proceeding, and formed part of a broader proposal aimed at encouraging greater competition in the Canadian wireless industry.
- In the present proceeding, the Bureau analyzes rate-setting methodologies more generally, recognizing that the Notice of Consultation seeks comment on wholesale rate setting in respect of a broader range of telecommunications services.Footnote 10 Given that wholesale rate-setting methodologies engage both passive infrastructure and active network access, the Bureau builds on its past work by focusing on:
- active network access, as opposed to passive infrastructure access;Footnote 11 and on
- fixed network access, which offers a far greater range of international wholesale rate-setting comparators than does mobile network access.
Additional Work Undertaken by the Bureau to Inform this Submission
- The Bureau's past efforts to understand wholesale rate setting have come in the context of analyzing competition in specific industry settings. Accordingly, in preparing this submission, the Bureau has performed a broader review of academic literature and international practices in wholesale rate setting. In this connection, the Bureau has sought expertise and perspective by consulting with a range of market participants within the Canadian telecommunications industry, conferring with a number of foreign telecommunications regulators, and engaging outside advisors with experience in Canadian and international wholesale rate setting.
IV. Contextual Considerations for this Proceeding
- Prior to commenting on specific rate-setting methodologies, the Bureau wishes to address two key considerations that provide important context for the CRTC's approach in this proceeding:
- the degree to which international comparisons are meaningful; and
- the role of wholesale rate setting as a single component within a broader wholesale access regime.
International Comparisons Must Be Identified Selectively
- In the Appendix to the Notice of Consultation, the CRTC discusses rate-setting approaches in six foreign jurisdictions: Australia and the United States; France, Germany, and the United Kingdom, all of which were then members of the European Union; and the European Union (EU) itself. The Bureau agrees that practices adopted in other jurisdictions, including these six, can offer insight for the present proceeding.
- However, comparing such practices to the Canadian context can be difficult. There are relatively few close foreign counterparts to the Canadian telecommunications industry. Of particular importance, the Bureau notes that, unlike many foreign jurisdictions, Canada has a two-platform telecommunications industry, featuring both significant coverage and take-up of both telephone and cable-basedFootnote 12 networks.Footnote 13 This factor does not vitiate international comparisons, but does narrow the range of jurisdictions that can serve as apt comparators to Canada.Footnote 14
- Therefore, as an alternative to the jurisdictions included in Appendix to the Notice of Consultation, the Bureau has identified six OECD countries that feature robust two-platform telephone and cable environments. These Two-PlatformFootnote 15 Comparators, identified based on an approach set out in Appendix A to this submission, are Belgium, Denmark, Hungary, the Netherlands, Portugal, and the United States.
- No two countries are identical. Canada features a vast geography with high costs to deploy networks to rural and remote communities. This factor creates a different context for the relationship between wholesale rate setting and investment incentives than may be present in jurisdictions where households are more densely located.
- However, the Two-Platform Comparators feature telecommunications sectors with a significant presence of both telephone- and cable-based networks, as well as high national broadband penetration and well-developed cable-based networks (as measured by market share). The Bureau notes that these conditions are not always satisfied in respect of the six jurisdictions included in the Appendix to the Notice of Consultation.
Wholesale Services Regulation in Foreign Jurisdictions
- Having identified the Two-Platform Comparators, the Bureau next analyzed whether and how these countries applied wholesale access regulation. This analysis helps identify the extent to which the Two-Platform Comparators offer useful comparisons for reviewing wholesale rate-setting approaches.
- At the same time, the Bureau investigated the degree to which the Two-Platform Comparators differed from well-developed telecommunications sectors not characterized by pervasive take-up of both telephone and cable company broadband. Establishing comparisons between different products and services subject to wholesale regulation across jurisdictions is a complex exercise that requires significant country-specific knowledge. For this reason, the Bureau selected an existing, publicly available, and comprehensive classification that included a large number of countries and that can be independently verified.Footnote 16 This classification, provided by the Body of European Regulators for Electronic Communications (BEREC),Footnote 17 allowed the Bureau to select a broad group of 29 countries, referred to as One-Platform Comparators,Footnote 18 alongside the smaller group of six Two-Platform Comparators.
- Table 1 examines how Two-Platform Comparators and One-Platform Comparators apply wholesale broadband access regulation. This table presents the proportion of jurisdictions that apply regulation to telephone and cable companies separately, and how this regulatory treatment varies between twisted pair copper, fibre-to-the-node, and hybrid fibre-coaxial networks (Legacy Networks) and all-fibre networks (NGA Networks).Footnote 19
Table 1 : How Foreign Jurisdictions Regulate Wholesale Fixed Access, by PlatformFootnote 20 Two-Platform Comparators One-Platform Comparators Legacy Networks NGA Networks Legacy Networks NGA Networks Telephone companies 83% 67% 90% 72% Cable companies 67% 17% 10% 7%
- Table 1 shows two significant differences between Two-Platform Comparators and One-Platform Comparators:
- Wholesale regulation of cable companies in the One-Platform Comparator jurisdictions is rare as compared to the situation in Two-Platform Comparator jurisdictions.
- There is generally more wholesale regulation in respect of Legacy Networks than NGA Networks in both Two-Platform Comparator and One-Platform Comparator jurisdictions.
- Throughout this submission, the Bureau will focus its international comparisons on the Two-Platform Comparator jurisdictions, with reference to how those jurisdictions differ from the One-Platform Comparators. This side-by-side comparison helps identify the extent to which regulatory choices vary with industry structure.
Wholesale Regulation is More than Just Rate-Setting
- Introducing, maintaining, or removing wholesale access is a broader policy consideration that is beyond the scope of this proceeding. Such a decision should only result from a staged approach, wherein the CRTC should consider:
- whether firms that provide a service possess market power; and
- which remedies are most appropriate to constrain such market power when it is identified.Footnote 21
- Accordingly, for the purposes of this proceeding, the Bureau assumes that the scenario on which the Notice of Consultation is predicated is the outcome of such a staged process. Critically, rate regulation is rarely the sole outcome of such a staged process. Rather, rate-setting methodologies are typically part of a larger package of remedies aimed at constraining market power. Table 2 demonstrates that the vast majority of foreign jurisdictions studied by the Bureau impose additional "non-price" remedies alongside rate regulation.
Table 2 : Ex Ante Remedy Choice in Jurisdictions that Regulate Wholesale Fixed AccessFootnote 22 Two-Platform Comparators One-Platform Comparators Legacy Networks NGA Networks Legacy Networks NGA Networks TransparencyFootnote 23 100% 100% 96% 94% Non-discriminationFootnote 24 100% 100% 100% 100% Mandated accessFootnote 25 100% 100% 100% 100% Cost accountingFootnote 26 83% 100% 96% 88% Price controlsFootnote 27 100% 100% 90% 91% Accounting separationFootnote 28 53% 67% 81% 85%
- While this proceeding focuses only on price-control-type remedies—i.e., wholesale rate-setting methodologies—international experience shows that these are typically combined with other regulatory tools to ensure that a wholesale access regime meets its policy objectives. In this connection, the Bureau views the next wholesale wireline review as an appropriate forum to assess the efficacy of non-price remedies.
V. Economic Framework and Competition Considerations for this Proceeding
- In this section of its submission, the Bureau sets out key competition-related concepts that stem from its review of economic theory and competition policy perspectives regarding rate setting for wholesale access in telecommunications. These concepts should guide the CRTC in determining a wholesale rate-setting methodology.Footnote 29
- The objective of economic regulation is to mimic competitive marketplace outcomes when an industry fails to obtain those outcomes through market forces alone.Footnote 30 As a network industry, the cost structures and investment profiles of the telecommunications sector tend toward concentrated infrastructure markets with high barriers to entry.Footnote 31 These factors create an industry structure of high fixed costs and low variable costs, such that the textbook optimal competitive result of price tending towards marginal cost is not necessarily a sustainable market solution or a desirable regulatory option.
- The search for alternative cost bases, the trade-offs between them, and the challenges of efficient and transparent implementation of specific costing methodologies have been approached in different ways. Some countries have responded by imposing mandated wholesale network access to promote "service-based" competition by wholesale-based competitors. However, other countries, particularly the US, have instead moved increasingly to limit such regulation in order promote "facilities-based" competition, including entry of new network owners.Footnote 32
- To be clear: the Bureau recognizes the important role of both facilities-based and wholesale-based network competition to determine marketplace outcomes.Footnote 33 The competitive process – i.e., the rivalry between businesses that makes them work harder to win customers – is more important to marketplace outcomes than the identity or technology of any one competitor. Competition itself is the end goal, as it brings about low prices, increased consumer choice, and higher levels of innovation that are critical to ensuring the prosperity of all Canadians.
Economic Efficiency is the Appropriate Benchmark for Assessing Wholesale Rate Setting
- Competitive marketplaces maximize economic efficiency.Footnote 34 Therefore, in determining the appropriateness of rate-setting methodologies in this proceeding, the CRTC should consider how each methodology would affect economic efficiency.
- Economic efficiency can be broken into three component forms: allocative, productive, and dynamic efficiency.
- A market achieves allocative efficiency when it apportions a product to those who value it the highest.Footnote 35 To achieve allocative efficiency, wholesale regulation should reduce entry barriers to the greatest extent possible, thereby allowing the competitive process to deliver the lowest possible prices to the widest range of buyers.
- A market achieves productive efficiency when it yields a given mix of economic output at the lowest cost possible.Footnote 36 To achieve productive efficiency, wholesale regulation should incentivize all suppliers to minimize their costs in providing relevant products and services.
- A market achieves dynamic efficiency when it minimizes the cost of service over the long-term by introducing new processes and technologies efficiently.Footnote 37 To achieve dynamic efficiency, wholesale regulation should provide appropriate incentives for suppliers to improve their products and processes, including network upgrades, capacity expansion, and innovative means for maximizing their use.
- Trade-offs exist between allocative, productive, and dynamic efficiency: no single approach can advance all three simultaneously.Footnote 38 For example, a policy that maximizes allocative efficiency, by pushing wholesale access rates down towards short-run marginal cost, can decrease dynamic efficiency by reducing investment incentives.Footnote 39 Similarly, very high wholesale rates might improve dynamic efficiency by maximizing investment incentives but, at the same time, can lower allocative efficiency by reducing the ability of competitors to price at sufficiently low levels in relation to their costs.
- Accordingly, from an economic efficiency perspective, there is no perfect rate-setting methodology. The CRTC is unlikely, in selecting these methodologies, to precisely replicate the trade-offs between different kinds of efficiency achieved by a perfectly competitive marketplace. However, this should not deter the CRTC from seeking to maximize overall economic efficiency by explicitly considering how each rate setting methodology will tend to affect economic efficiency.
Vertical Integration Can Influence Pricing Incentives and Competitive Outcomes
- In many aspects of the Canadian telecommunications industry, network owners are vertically integrated, such that they own upstream facilities (e.g., access networks) but also provide services in downstream retail markets in competition with wholesale-based competitors that do not own parallel upstream facilities. This type of market structure can significantly influence wholesale and retail pricing incentives and, consequently, impact competitive outcomes in both the wholesale and retail markets.
- In particular, incentives may exist to:
- engage in exclusionary conduct in the retail market; and
- raise prices in the retail market as a result of rate levels in the wholesale market.
- In determining a wholesale rate-setting approach, these incentives should be kept front of mind, so that the CRTC's approach does not inadvertently create or strengthen any incentives that may act contrary to the low prices, increased consumer choice, and higher levels of innovation that are delivered by the competitive process.
Incentives Leading to Exclusionary Conduct
- In the typical case where there are few sellers at the wholesale level, there may be an incentive for those sellers to engage in exclusionary conductFootnote 40 known as "margin squeezing" by:
- raising wholesale rates to levels at which wholesale-based competitors could no longer sustain a profit downstream;
- lowering retail prices in the downstream market to levels below wholesale rates; or
- a combination of both strategies—i.e., simultaneously raising wholesale rates and lowering retail prices to create a margin between them at which a wholesale-based business model would not be profitable.Footnote 41
- In each of these circumstances, the profits of wholesale-based competitors are "squeezed" such that their competitive influence may be reduced or eliminated. In some cases, these competitors may be the preferred choice for some group of end consumers. Therefore, such conduct can fundamentally frustrate the intent underlying mandated wholesale access—that is, providing greater competition and delivering more economically efficient outcomes in the marketplace.
- Each wholesale rate-setting methodology provides different scope for margin squeezing behaviour. For example, as discussed further in this submission, a margin squeeze incentive is less likely in respect of retail-minus methodologies, whereas it can be more prevalent in methodologies that set wholesale rates based on cost or commercial negotiation.
- Margin squeezing can also be explicitly addressed on an ex ante basis by regulators. Such a regulatory approach can take the form of some specification of an Economic Replicability Test (ERT).Footnote 42
High Wholesale Rates can Incentivize Higher Retail Prices
- Under some rate-setting methodologies, it is possible for wholesale rates to exceed the costs incurred by a network owner to provide wholesale services. When this occurs, the network owner earns a positive margin on every sale that a wholesale-based competitor makes using that network owner's infrastructure.
- Generally speaking, the existence of this wholesale margin provides an incentive to increase retail prices. High wholesale rates can lead to high retail prices by two mechanisms. The first is that, by raising the network access costs of wholesale-based competitors, those competitors are likely to pass through at least some proportion of that increase to their customers in the form of higher prices. This softens retail competition and provides scope for network owners to raise prices. The second is that, by creating (or increasing) a wholesale margin, network owners are incentivized to raise retail prices. The logic underpinning this second effect is based on the intuition outlined in the economic literature on Upward Pricing Pressure as applied to vertically integrated markets.Footnote 43
- The key premise of this concept is that the existence of a wholesale margin "cushions" a network owner against the downside risk of raising prices. In a circumstance where, following a price increase, a customer of a network owner switches their service to a wholesale-based competitor that uses that network owner's facilities, the network owner will continue to earn some positive margin (i.e., the wholesale margin) from that customer's consumption. This is different from the alternative situation where the network owner does not earn a wholesale margin, such that the network owner would simply lose all of the revenue that flows from that customer's consumption. The fact that the network owner, bolstered by a wholesale margin, continues to earn some amount of revenue from that customer makes raising prices relatively more attractive.Footnote 44
- To counteract these incentives, the CRTC should be mindful of approaches that increase the difference between what it understands to be a network owner's costs of providing a wholesale service, and the associated wholesale rate.Footnote 45 All other things being equal, when network owners earn a greater margin in the provision of wholesale services, this will tend to lead to higher retail prices and a reduction in retail competition both from network owners, for the reasons set out above; and from wholesale-based competitors, as a result of higher input costs.
Information Asymmetries Can Hinder the Rate-Setting Process
- Setting wholesale rates is inherently complex and data-intensive. This can lead to situations of information asymmetry, in which one party in the wholesale rate-setting process has more or better information (e.g., a network owner in respect of its actual wholesale costs) than another (e.g., the regulator who is charged with discovering those costs and setting wholesale rates). Information asymmetries challenge administrative processes, and can result in rates that deviate from intended methodological outcomes.
- Phase II was designed, based in part on long-standing tariff regulations, to be implemented through an adversarial process. In this process, network owners prepare studies specific to their network architecture and submit it for public comment, review, and approval. However, it can be difficult for third parties (who do not own or operate the network in question) to evaluate these studies.Footnote 46 When this is true, an asymmetry of information exists between the network owner and other parties, including both the stakeholders who are key to an adversarial process and the regulatory agencyFootnote 47 which must act based on the information put onto the record of a proceeding.Footnote 48
- Addressing these asymmetries requires two strategies. First, the rate-setting methodologies that the CRTC selects must address the operational considerations that surround what is necessarily an information-heavy and data-intensive process. Second, at the same time, recourse to benchmarking can help identify situations where rates fall outside a reasonable range.
Rate Differences between Network Types can Create Competitive Advantages
- When wholesale rates are set separately for cable- and telephone-based networks, any resultant rate differences can affect the competitive position of each of these classes of providers in two ways:
- Rates that differ significantly between cable and telephone companies in each geography could incentivize wholesale-based competitors to sign up consumers only to the cheaper technology.Footnote 49 Depending on the level of wholesale rates, this outcome could frustrate competition between network owners at the wholesale level, and result in one class of competitors carrying a disproportionate share of traffic from subscribers to wholesale-based competitor services.Footnote 50
- Rates that correctly compensate one class of network owner but fail to compensate (or overcompensate) the other could affect the relative profitability of network investments for these classes of carriers. Such a differential would necessarily have follow-on effects on the ability of network owners to compete for retail customers.
- It is therefore important that any consideration of rate-setting methodologies ensure their fitness for both platforms, along with the potential competitive effects of imbalance in wholesale rates.
VI. Assessment of Rate-Setting Methodologies
- In the Notice of Consultation, the CRTC seeks comment on five wholesale rate-setting methodologies. In this section of its submission, the Bureau assesses the advantages and disadvantages that these wholesale rate-setting methodologies may have for competition.
- Appendix B presents a summary of the Bureau's assessment of each of the five rate-setting methodologies identified by the CRTC.
Cost-Based Approaches Share Significant Similarities
- Three of the five methodologies proposed in the Notice of Consultation are cost-based approaches (i.e., continued use of Phase II costing, costing using an Efficient Operator approach, and costing using an Accounting-Based Costing approach). In fact, cost-based approaches predominate internationally. Table 3 illustrates the frequency, as of 2018, with which regulators applied a number of approaches to wholesale rate setting, including the five methodologies identified by the CRTC.
- As Table 1 suggests, cost-based approaches are the most widely-used in both comparator groups, most markedly in the Two-Platform Comparator grouping. Retail-minus and commercial negotiation approaches are not used within the Two-Platform Comparator jurisdictions, and are used only sparingly in other jurisdictions.
Table 3 : Access Price Control Approaches in Jurisdictions that Mandate Wholesale Fixed AccessFootnote 51 Two-Platform Comparators One-Platform Comparators Legacy Networks NGA Networks Legacy Networks NGA Networks Long-run incremental costing (including methodologies like Phase II and efficient operator)Footnote 52 56% 50% 39% 45% Retail-minusFootnote 53 0% 0% 7% 8% Accounting-based costingFootnote 54 36% 33% 33% 29% Commercial negotiationFootnote 55 0% 0% 10% 9% Others and combinations 8% 17% 11% 8% TOTAL 100% 100% 100% 100%
- Cost-based methodologies share a common feature in that they seek to set wholesale rates at some level relative to a properly specified cost base. Because of this common feature, there are associated commonalities in respect of the advantages and disadvantages of the three cost-based methodologies. Accordingly, before analyzing these methodologies in greater depth, this submission first discusses the broader commonalities among cost-based approaches.
- From an economic perspective, when wholesale services are supplied at an appropriate measure of cost:
- network owners are compensated for the costs that they incur in providing the wholesale service, which aids productive and dynamic efficiency;Footnote 56 and
- wholesale-based competitors obtain network access at a price that is similar to the costs of the relevant network owner, which provides sufficient scope to aid the expansion of allocative efficiency (i.e., lower prices and a broader number of customers being served).Footnote 57
- There are four negative aspects to cost-based approaches:
- Economic theory is not settled as to which cost base should be reflected in wholesale pricing (e.g., marginal cost, incremental cost, etc.).Footnote 58 This is because each different cost base brings about different effects on economic efficiency. Ultimately, the choice of one cost base over another will tend to require value judgements involving the relative value of each of the three component forms of economic efficiency.
- Even when an appropriate cost base can be agreed upon, its estimation is complex. Properly implementing a cost-based methodology requires advanced knowledge of accounting, engineering, economics, and regulation.Footnote 59 This means that the outcomes of a resulting rate-setting process can be opaque, complicated, and subject to significant debate as to their correctness.
- Given the significant effort and time associated with developing and implementing specific costing methodologies, cost-based approaches can be rigid.Footnote 60 In times of technological change, where costs and demand can change in unanticipated ways, quickly and easily updating cost-based wholesale rates may be impossible. This can cause significant departures between wholesale rates and actual underlying costs during the interim period where rates are adjusted to meet a new reality.
- Cost-based approaches can provide an economic incentive for network owners to engage in exclusionary marketplace conduct.Footnote 61 Such conduct can frustrate the underlying intent of wholesale access – that is, providing greater competition and delivering more economically efficient outcomes in the marketplace.
Conclusion on Cost-Based Approaches
- Most likely to maximize economic efficiency.
- Choice of cost base involves subjectivity.
- Costs are difficult to accurately estimate.
- Resulting rates cannot fluidly adapt to marketplace changes.
- Creates incentive to engage in exclusionary behaviour.
Phase II Costing
- Phase II costing (Phase II) was developed in Canada as an incremental costing method for rate setting. Phase II is a long-run modelling approach that incorporates, on an ongoing basis, the actual costs incurred by the network owner in respect of their equipment choices.Footnote 62
- Phase II relies on a network owner's "historic" costs and marries them with projections as to indicators such as the expected life of equipment, its capacity relative to expected growth in usage, engineering factors (e.g., the point at which equipment needs to be upgraded), and financial factors (e.g., the cost of capital).
- In a Phase II rate setting proceeding, each network owner files information in relation to a series of specified factors. Interveners then propose alternative values for these factors, and the CRTC sets the ultimate wholesale rate for that network owner.
- There are two main advantages to Phase II:
- Applied correctly, Phase II allows network owners to recover their costs of providing wholesale services, which incentivizes productive and dynamic efficiency.
- The Phase II process creates a discovery method for ensuring that rate-setting will be based on up-to-date information.
- There are three negative aspects to retaining a Phase II approach:
- Because Phase II relies heavily on detailed filings by both network owners and other market participants, its discovery method is subject to significant information asymmetry challenges.Footnote 63
- Phase II has proven to be difficult and time-consuming to implement.
- The results of the Phase II process have frequently been contentious, with some network owners alleging a "serious disconnect" between Phase II methodologies and the cost of providing wholesale services.Footnote 64
Conclusion on Continued Use of Phase II Costing
- Canada-specific methodology that has been developed through years of use.
- Combines cost-based approach with certain actual costs incurred, representing a middle ground between promoting competition and compensating network owners.
- Significant information asymmetry issues; difficult for regulator or other interveners to know true costs of operating the relevant network.
- Difficult and time-consuming to implement.
Retail-Minus Rate Setting
- The retail-minus approach is a price- rather than cost-oriented approach. It involves setting wholesale rates at a discount to some observed retail price for one or more closely-related downstream services. This methodology is associated with an economic model called the Efficient Component Pricing Rule.Footnote 65
- There are three main advantages to retail-minus:
- Once the regulator determines an appropriate discount, a retail-minus approach based on an easily observable measure of price (e.g., average revenue per user) can be easy to calculate, and can be updated frequently in response to changing market conditions.Footnote 66
- Under a retail-minus approach, there is little ability for network owners to margin squeeze, because the margin for the upstream wholesale input is set by the regulator.Footnote 67,Footnote 68
- Properly-specified retail-minus methodologies tend to ensure that network owners' investment incentives are preserved.Footnote 69
- There are four negative aspects of a retail-minus approach:
- In circumstances where retail-minus sets wholesale rates above an appropriate cost-based level, network owners will capture some margin on retail sales made by wholesale-based competitors, which will provide those network owners with an incentive to increase prices.Footnote 70
- When applied in marketplaces that are not vigorously competitive, the retail-minus benchmark price may be based on a supra-competitive price that maintains a market power-related margin – which is exactly what wholesale regulation is designed to eliminate.Footnote 71
- It can be difficult to determine a proper downstream benchmark price when there is significant discounting and bundling activity, as is common in the telecommunications sector.Footnote 72
- Retail-minus provides a transparent mechanism for industry participants to influence prices upward through coordination, as any increase in the retail price of network owners' services will translate into an associated wholesale rate increase.Footnote 73
- From these perspectives, retail-minus is likely of greatest use in markets for transitional or immature offerings.Footnote 74 In such circumstances, where there is a high degree of uncertainty about costs and demand, retail-minus may avoid speculative or ultimately inaccurate cost predictions. However, for more established markets with greater stability, retail-minus approaches may tend to fare poorly in achieving allocative efficiency.
Conclusion on Retail-Minus
- Retail-minus rates can be easy to calculate, and can quickly adapt to marketplace changes.
- Little ability to margin squeeze.
- Network owners' investment incentives are often preserved.
- If network owners earn a margin on their wholesale costs, they will be incentivized to increase prices.
- If the marketplace is not vigorously competitive, retail-minus benchmark price may be based on supra-competitive price.
- Determining benchmark price can be difficult in the face of significant discounting and bundling activity.
- Provides a transparent mechanism for pricing coordination.
Efficient Operator Costing
- As a method of LRIC costing, Phase II is unusual among Canada's foreign counterparts. Most of these instead use "efficient operator"Footnote 75 LRIC approaches that rely more heavily on modelling a hypothetical network that uses efficient technologies to serve wholesale customers.Footnote 76
- There are three main advantages to efficient operator modelling:
- Because determinations are less dependent on information from network owners, this reduces issues of asymmetric information.
- An efficient operator approach encourages both productive and dynamic efficiency on the part of network owners, since a network owner whose costs exceed the modelled costs will have a financial incentive to improve efficiency.
- Once an efficient operator model is established in respect of a service, updating that model regularly against demonstrated results can be relatively simple compared to approaches based on (historical) costs that have been actually incurred.
- There are two negative aspects of efficient operator modeling:
- Like other economic modelling approaches, efficient operator methodologies are not simple to implement. Efficient operator modeling requires significant engineering and economic knowledge to specify the network to be modeled. Experience in other jurisdictions indicates an initial learning curve accompanying first-time implementation.Footnote 77
- In the Canadian context, an efficient operator approach would require specific policy and other modelling choices to reflect Canada's two-platform competitive environment (and the fact that Canada has a number of regional operators).
- Nevertheless, the Bureau recognizes the international trend toward efficient operator methodologies over the last two decades.Footnote 78 This trend may reflect a level of comfort with these operational challenges, which regulators have overcome by, for instance, delegating the efficient operator study to a third-party consultancy with relevant expertise.Footnote 79
Conclusion on Efficient Operator Costing
- Fewer issues of information asymmetry compared to other cost-based approaches.
- Provides financial incentive for inefficient operators to achieve productive and dynamic efficiency.
- Once efficient operator model is established, it is reasonably easy to update.
- First time implementation of efficient operator modeling requires significant effort.
- Efficient operator approach would need to be adapted to marketplace realities in Canada (e.g., two-platform environment, regional competitors, etc.)
- Accounting-based costing (ABC) is designed to calculate the relevant costs of a specific operator, based on that operator's historical accounting data. Allocation methodologies, including Activity-Based Costing, are used to allocate direct, indirect and overhead costs to the products and services offered by that operator.Footnote 80 Because ABC methods start from overall costs, and allocates each to one cost category or another, they are often referred to either Fully-Distributed Costing (FDC) or Fully Allocated Costs (FAC).
- There are two major advantages to ABC:
- This is the traditional form of costing methodology establishing in the telecommunications industry through the 1970s and 1980s, including in Canada, where it was known as Phase III costing.Footnote 81 It is well-established and understood.
- It fully reflects the actually-incurred historical cost accounting data of the operator in question. It therefore guarantees that the operator will be completely compensated for its costs.
- With the introduction of competition in the sector, regulators in Canada and in foreign jurisdictions identified a number of disadvantages to ABC, and increasingly replaced it with Phase II,Footnote 82 efficient operator, and other forward-looking methodologies:
- ABC is less likely to result in allocative efficiency. Economic theory typically prefers current cost accounting, as opposed to examining historical costs,Footnote 83 because current costs more closely reflect the underlying economic costs of providing wholesale access. Therefore, relying on ABC imports a significant risk that wholesale rates could deviate significantly from current economic costs, with associated negative effects on economic efficiency.
- There is a built-in information asymmetry challenge associated with ABC that is likely to result in further deviations from economic costs. Because ABC relies on the specific costs of a specific operator, that operator will know its own costs much better than the regulator and other market participants and, therefore, will have an incentive to allocate costs in a manner that it finds most advantageous.
- Economic theory typically prefers current cost accounting, as opposed to an historical approach, for rate setting.Footnote 84 The reason for this is that economic modelling more directly addresses the heart of the rate-setting problem by assessing the underlying economic costs of providing wholesale access, rather than the actual costs as allocated and determined by an accounting system. Therefore, relying on ABC imports a significant risk that wholesale rates could deviate significantly from current economic costs, with associated negative effects on economic efficiency.
Conclusion on Accounting-Based Costing
- Completely compensates network owner for costs.
- Well-established and well-understood.
- Because rates are based on historical cost information, ABC can cause significant deviations from current economic costs, with associated negative effects on economic efficiency.
- A commercial negotiation approach to rate setting requires each network owner and each wholesale-based competitor to determine a wholesale rate. There are different models of negotiation. Some may require industry participants to arrive at a wholesale rate on their own. Others may involve some form of mediation, arbitration, or other regulatory oversight or involvement in the process.
- The two main advantages to a commercial negotiation approach:
- Unlike cost-based approaches, commercial negotiation does not oblige parties to produce and analyze significant amounts of information.Footnote 85
- Commercial negotiations can structure wholesale rates in a way that is more responsive to underlying marketplace changes – for example, by specifying rates that react to a variety of factors.
- There are two negative aspects of commercial negotiation:
- There is no guarantee that a resultant wholesale rate would be efficient from an economic perspective. Rather, the price could reflect the difference in bargaining power between the network operator and competitor.
- An approach based only on commercial negotiation does not guarantee that network owners and wholesale-based competitors will be able to find mutually agreeable terms upon which to base a contract. Network owners have few obvious incentives to enter into any contract that lowers their profits below the amount they would earn from not signing the contract. This could mean that network owners would refuse to enter into an agreement with wholesale-based competitors if they believed that such a refusal would allow them to earn higher profits from competing solely against their facilities-based rival or rivals. This would, of course, frustrate the goal of wholesale access regulation – i.e., to facilitate entry in order to increase competition.
- Therefore, any rate-setting approach based on commercial negotiations requires some form of backstop. Indeed, it is difficult to imagine how commercial negotiation could function as a rate-setting mechanism for mandated wholesale network access supply without a backstop.
Final Offer Arbitration as a Backstop to Commercial Negotiation
- In the CRTC's current telecommunications wholesale rate-setting approaches, tariffed rates act as such a backstop, and commercially-negotiated rates are permissible (e.g., off-tariff agreements). In the broadcasting context, by contrast, Final Offer Arbitration (FOA) is used as a backstop to commercial negotiations.Footnote 86
- There are two main advantages to relying on FOA as the backstop that enables commercial negotiation to be used for rate-setting under mandated wholesale network access:
- Each party has a strong incentive to provide their best offer to the arbitrator, lest they risk the other party's proposal appearing more reasonable to the arbitrator who will ultimately determine wholesale rates. In this way, FOA can produce an optimal outcome from an economic perspective.Footnote 87
- Each party retains flexibility to make their case in the way they deem most persuasive. This allows different methodological approaches to co-exist side by side.
- There are three negative aspects of applying FOA in these circumstances:
- FOA determinations require significant expertise on the part of the arbitrator. In its original setting (Major League Baseball salary arbitration), performance and compensation of comparable players are observable such that the arbitrator can assess how reasonable the competing offers are.Footnote 88 In the current regulatory context, even with sufficient expertise in economics, accounting, engineering and law, information asymmetries could challenge the arbitrator's ability to evaluate complex offer packages.Footnote 89
- Without a clearly-established methodology, the preparation of FOA offers can involve significant uncertainty and, for participants who are not network owners, significant information asymmetry disadvantages.
- Because FOA is run separately for each bilateral commercial negotiation, it has the potential to be extremely time-consuming. This is particularly the case in view of the many network owner/wholesale-based competitor pairings that exist for each mandated wholesale offering.
- Particularly where there are a small number of network owners and wholesale-based competitors, and where an agreed-upon rate-setting methodology has not been settled, commercial negotiation backed by FOA can prove a helpful way to solve a rate-setting dispute.
Conclusion on Commercial Negotiation
- Potentially less information-intensive vis-à-vis costing methods.
- Contracts can be structured to be flexible in implementation.
- Incentivizes reasonability in negotiations and final offers.
- Negotiated rates are not guaranteed to be economically efficient, and may simply represent the relative bargaining power of the parties.
- No guarantee that network owner and wholesale-based competitor will conclude a contract without an effective backstop, shifting much of the weight to the backstop process.
- May be difficult to find effective mediators and arbitrators to support negotiation process; significant knowledge required and information asymmetries exist.
- The Bureau's review concludes that there is no one supreme wholesale rate-setting methodology. For example, while cost-based methodologies have a strong foundation in economic theory, they can be rigid and difficult to implement, which increases the risk that they could result in rates that are divorced from the intent of a well-specified methodology.
- Equally, there is no methodology proposed by the CRTC in the Notice of Consultation that is wholly inadequate. For example, while it is less certain that retail-minus or commercial negotiation approaches can achieve results that maximize economic efficiency, it may be possible to supplement these methodologies through safeguards and benchmarking to improve confidence in their outcomes.
- These conclusions underscore the fact that, perhaps more important than the methodological framework is the detailed rules on how a given methodology would be implemented. The Bureau is ready to support the design and development of more detailed proposals in future stages of this proceeding, or in follow-on proceedings as appropriate.
- The designated representative of the Commissioner for this proceeding is:
Major Case Director and Strategic Policy Advisor
Policy, Planning, and Advocacy Directorate
15th Floor, 50 Victoria Street
Gatineau, Quebec K1A 0C9
Appendix A: International Comparators
- The Appendix to the Notice of Consultation includes analyses of the rate-setting approaches used for selected wholesale services in Australia, the European Union, France, Germany, the United Kingdom, and the United States.Footnote 90
- Fixed access in Canada is a two-platform environment characterized by relatively high penetration, and use, of access networks owned and operated both historic telephone and cable carriers.Footnote 91 Canada ranks tenth (tied) of the 37 OECD member states for fixed broadband penetration, and fourth for the cable platform's broadband subscription share.
- Not all countries are similarly structured. Many foreign jurisdictions feature significantly lower penetration and usage rates on cable-based networks.Footnote 92 Given this significant difference, these countries may not serve as apt comparators to Canada.
- Accordingly, the Bureau has identified a set of comparator two-platform jurisdictions which, like Canada:
- are members of the Organization for Economic Cooperation and Development (OECD);
- have a relatively high level of telecommunications sector development;Footnote 93 and
- evidence high rates of participation by both telephone- and cable-based platforms in broadband service, as indicated by the proportion of fixed broadband subscriptions provided over cable-based networks.
- In particular, and in order to generate a manageable grouping, the Bureau identified those member states falling into both the top half of OECD broadband penetration percentages, and top third of cable broadband shares.
- The results of this analysis, set out in Table A-1, show that in addition to Canada, the OECD member states that meet these criteria are: Belgium, Denmark, Hungary, Netherlands, Portugal, and the United States. While the specific threshold of comparability among these countries may be debated, the Bureau considers comparisons with these well-developed, two-platform telecommunications environments to be informative.
Table A-1: Two-Platform Broadband Environments in the OECDFootnote 94 OECD Member State Fixed Broadband Penetration Cable Share of Fixed Broadband Rank
(top half shaded)
(population, not households)
(top third shaded)
Share Switzerland 1 46% 15 28% France 2 44% 27 15% Netherlands 3 43% 7 47% Denmark 3 43% 11 34% Norway 5 42% 15 28% Germany 5 42% 21 23% South Korea 5 42% 27 15% Belgium 8 40% 5 52% United Kingdom 8 40% 23 20% Canada 10 39% 4 55% Sweden 10 39% 24 17% Iceland 10 39% 35 0% Portugal 13 38% 12 33% Greece 13 38% 35 0% Luxembourg 15 37% 30 10% Australia 16 35% 22 21% United States 17 34% 1 65% New Zealand 17 34% 32 4% Hungary 19 32% 6 50% Estonia 19 32% 19 25% Finland 19 32% 19 25% Japan 19 32% 24 17% Spain 19 32% 26 16% Czech Republic 24 31% 15 28% Slovenia 25 30% 14 29% Ireland 25 30% 18 26% Austria 27 28% 10 35% Israel 27 28% 13 30% Italy 27 28% 35 0% Lithuania 27 28% 33 3% Slovakia 27 28% 27 15% Latvia 32 27% 33 3% Poland 33 20% 8 41% Chile 34 18% 3 55% Turkey 35 17% 31 7% Mexico 36 15% 9 38% Colombia 37 14% 2 61%
Relevant Comparator Services
- In Canada, wholesale fixed broadband access is mandated based on two combinations of architecture and service characteristics:
- Fixed wholesale access is mandated on an aggregated basis to what is referred to in this submission as Legacy Networks. For large incumbent telephone company platforms, "legacy" means twisted-pair copper DSL and fiber-to-the-node (FTTN) DSL. For large incumbent cable company platforms, "legacy" means hybrid fibre-coaxial cable. The "aggregated" basis on which legacy fixed access is mandated refers to hand-offs at national or regional points of interconnection.
- Fixed wholesale access is mandated on a disaggregated basis to what is referred to in this submission as NGA Networks, or all-fibre networks. The "disaggregated" basis refers to hand-offs at regional and local points of interconnection.
- Wholesale fixed broadband access is mandated in many other jurisdictions, including most of the two-platform and most of the one-platform environments described above. However, the specific product or products mandated in this manner differ from country to country, each with unique geographic and technical characteristics. In order to establish broad comparisons across countries to identify which approaches to wholesale rate setting are prevalent, and which supporting remedies accompany them, it is necessary to establish which mandated wholesale fixed broadband access products are comparable.
- BEREC reports based on regulatory accounting databases it maintains for this purpose.Footnote 95 These reports map regulated wholesale products provided by BEREC participants to product groupings within the wholesale markets designated by the European Union, in 2014, as ones that "may have such characteristics as to justify ex ante regulation",Footnote 96 i.e.:
- Market 1: Wholesale call termination on individual public telephone networks provided at a fixed location;
- Market 2: Wholesale voice call termination on individual mobile networks;
- Market 3(a): Wholesale local access provided at a fixed location;
- Market 3(b): Wholesale central access provided at a fixed location for mass-market products;
- Market 4: Wholesale high-quality access provided at a fixed location.
- Wholesale fixed broadband access products within BEREC-participating jurisdictions, accordingly, are found in the product groupings established for Markets 3(a), 3(b), and 4. Table A-2 identifies these product groupings, their descriptions, and whether they were included as comparable products in Tables 1-3 on that basis:
Table A-2: Comparator services in BEREC regulatory accounting databases Product nameFootnote 97 DescriptionFootnote 98 Closest Canadian tariffed equivalent Include as comparable for wholesale high-speed access? Mapping (Tables 1-3)? Market 3(a) Wholesale local access provided at a fixed location Disaggregated wholesale high-speed access M3a_ULL Local loop unbundling service on copper network Telco: legacy DSL.
No. M3a_SLU Sub loop unbundling on copper network - No. M3a_SA Shared Access service on copper network Telco: legacy DSL.
No. M3a_fiberLLU Fibre local loop unbundling - No. M3a_VULA(FTTC) VULA on Fiber to the Cabinet Network Telco: wholesale HSA (FTTN).
Cable: wholesale HSA (TPIA).
Yes. Legacy. M3a_VULA(FTTH) VULA on Fiber to the Home Network Telco: FTTP (xPON).
Cable: FTTP (RFoG/xPON).
Yes. NGA. M3a_DF Dark fibre in access network - No. M3a_DA Duct access on access network Telco: support structure conduit tariffs.
No. Market 3(b) Wholesale central access provided at a fixed location for mass-market products Aggregated/disaggregatedFootnote 99 wholesale high-speed access M3b_Acc_Leg Access component of bitstream service on copper access network (from the central office until the CPE) Telco: wholesale HSA (DSL/FTTN).
Cable: wholesale HSA (TPIA).
Yes. Legacy. M3b_backhaul Backhaul bandwidth component of bitstream service - No. Market 4 Wholesale high-quality access provided at a fixed location M4_Active_Leg Terminating segment on legacy copper network Telco: wholesale HSA (FTTN).
Cable: wholesale HSA (TPIA).
Yes. Legacy. M4_Active_NGA Terminating segment on FTTx network Telco: FTTP (xPON).
Cable: FTTP (RFoG/xPON)).
Yes. NGA. M4_Passive Access to passive infrastructure (dark fiber) - No.
- Table A-2 illustrates how, for European jurisdictions, eight of 13 submarket groups within Markets 3(a), 3(b), and 4 do not have a close tariffed equivalent in Canada (e.g., sub-loop unbundling), or were not wholesale high-speed access products at the data and networking layers, as opposed to layer-1 physical networking. These products were not included as comparable products in Tables 1-3.
- Of the remaining five submarkets:
- Three (disaggregated VULA FTTC, aggregated legacy access, and active legacy access) mapped onto legacy wholesale high-speed access and were included as "Legacy Networks" in Tables 1-3, and
- Two (disaggregated VULA FTTH, active NGA) mapped onto next-generation, all-fibre wholesale high-speed access and were included as "NGA Networks" in Tables 1-3.
- Geographic characteristics can differ significantly between carriers and between jurisdictions in ways that are not directly comparable.
- Finally, we note that no wholesale broadband fixed access regulation was assumed to exist in the United States for the above-noted "Legacy Networks" and "NGA Networks" products. Although the U.S. has mandated unbundled local loop and other regulatory remedies in respect of legacy PSTN infrastructure, it has not generally carried these over to fixed network access capable of reaching the CRTC's basic service targets.Footnote 100
Appendix B: Summary of Bureau Advice
|Qualities Attributable to each Cost-Based Methodology|| || |
|Phase II Costing|| || |
|Retail-minus|| || |
|Efficient Operator Costing|| || |
|Accounting-Based Costing|| || |
|Commercial Negotiation|| || |
*** End of Document ***