تعریف بازار عمده فروشی در ارتباطات : مسئله دسترسی به پهنای باند عمده فروشی

The scope of a wholesale market may, in addition to demand- and supply-side substitution at the wholesale level, also be determined by substitution patterns at the retail level. Considering wholesale broadband access markets, it is argued that each of these forces can be strong enough to render a 5–10% price increase by a hypothetical monopolist at the wholesale level unprofitable and thus may lead to a wider wholesale market definition including, for example, cable networks in addition to DSL. Based on the theory of derived demand elasticities the paper discusses under which circumstances this could be the case. The position of the European Commission and the practice of national regulatory authorities are then reviewed in light of these arguments.

In the telecommunications sector, national regulatory authorities (NRAs) as well as competition authorities frequently have to deal with the issue of market definition at the wholesale level. The Recommendation of the European Commission on Electronic Communications Markets susceptible to ex ante regulation,2 for example, encompasses 18 markets, 11 of which are wholesale markets. Although wholesale market definition is, from a conceptual point of view, not much different from retail market definition, there is an additional issue which has to be taken into account and which this article attempts to highlight: the role of retail markets in wholesale market definition. The economic literature has, as far as the author is aware, not been very explicit on this issue to date, but nevertheless (or because of that), there have recently been intensive discussions between the European Commission and NRAs in particular in the context of wholesale broadband access markets. This article uses a simple framework applied to the definition of wholesale broadband access markets. Section 2 sketches the discussion between the European Commission and some NRAs on the definition of wholesale broadband access markets; Section 3 discusses the role of retail markets for wholesale market definition from a theoretical point of view; Section 4 outlines how competitive constraints via the retail level can be evaluated; Sections 5 reviews the position of the European Commission and the practice of the NRAs in light of these arguments; Section 6 summarizes and concludes.

It has been argued that wholesale market definition may depend, in addition to supply- and demand-side substitution at the wholesale level, on substitution patterns at the retail level. The impact of the constraints via the retail level will in general be stronger the larger the demand elasticity at the retail level is, the more of a wholesale price change is passed on to the retail level (i.e., the more competitive the retail level is), and the larger is the ratio of wholesale to retail price. The constraints via the retail level may be particularly important for the definition of wholesale broadband access markets, where DSL competes with alternative platforms at the retail level and the costs for bitstream access account for a large part of total costs. Retail competitive constraints, therefore, should be considered at some stage of the regulatory process. Some countries and the European Commission have followed “unconventional” approaches to this issue in the case of wholesale broadband access markets: while some disregarded the actual coverage and the actual resulting competitive constraint of alternative platforms at the stage of market definition, others appear to have followed the approach of the European Commission and looked at supply- and demand-side substitution at the wholesale level only. These approaches, however, do not seem to have lead to incorrect decisions regarding the SMP-assessment so far as: • those countries which disregarded the actual coverage of alternative platforms at the stage of market definition accounted for this factor at the stage of market analysis; • in those countries which excluded alternative platforms mainly with reference to supply- and demand-side substitution at the wholesale level a consideration of retail competitive constraints would probably not have led to a different conclusion; • the European Commission did not make use of its veto power, not even in case of the Netherlands, the only decision in which the inclusion of cable networks indeed is likely to have changed the result of the SMP assessment. In light of the estimates derived in Section 4, the relative coverage of cable networks, and the relatively low share of ULL-based DSL broadband services (see Table 1), the results of the market analyses appear reasonable. Only in the Netherlands does cable network coverage seems indeed large enough40 to provide a sufficient competitive constraint via the retail level. Nevertheless there seems to be need for a change in the Recommendation on Relevant Markets. In the spirit of technological and competitive neutrality, the restriction “if and when they offer facilities equivalent to bit-stream access” imposed on alternative platforms should be deleted from the Recommendation.41 Instead it should be recognised that competitive constraints via the retail level may exist and that these constraints may be strong enough to fulfil the HM-Test at the wholesale level. Of course, NRAs would need to provide evidence in a particular case that the extent of these constraints is indeed strong enough (with a standard of proof comparable to the one applied in past cases). In the meantime the European Commission should continue not to veto decisions where alternative platforms are included into the market and the competitive constraints via the retail level are indeed likely to be strong enough to constrain a hypothetical monopolist and, in the end, the DSL incumbent. If these competitive constraints are not taken into account, there remains the danger that DSL operators are regulated despite the existence of several alternative infrastructures delivering similar services (in large enough areas) at the retail level. Having discussed the issue of retail competitive constraints, the question remains how to deal with alternative platforms delivering broadband services at the retail level in cases where wholesale as well as retail competitive constraints are not strong enough to put them into the same market as DSL. Consider a case, for example, where the wholesale price only makes up a small part of total costs and retail demand elasticity is not large enough to compensate for that. If one strictly applied the HM-Test methodology, alternative platforms would not be part of the relevant wholesale market, even if they have full coverage. Still, undertakings operating on these platforms are fully fledged competitors at the retail level which do not depend on non-replicable infrastructure (or put differently, there is infrastructure-based competition at the retail level). Insofar as regulation at the wholesale level serves the promotion of downstream competition, it appears that finally even those platforms will have to be taken into account in regulatory decisions although they do not develop a sufficient competitive constraint on DSL at the wholesale level (neither at the wholesale level directly nor via the retail level). If there are several vertically integrated operators with large enough coverage, retail competition is unlikely to be impeded even if none of these operators is willing to grant access to third parties. Regulation at the wholesale level, therefore, is likely to be inappropriate even if a market for, say, DSL bitstream access can be defined by means of the HM-test and an SMP position is found.42 Whether the HM-Test is the appropriate or the only instrument for NRAs in such situations is certainly a question warranting future investigation.