ITU/BDT Arab Region al Workshop on

CoE/ARB ITU/BDT Arab Regional Workshop on "Developing Competition Policies and Strategies in Telecommunications" Rabat-Morocco, 19 - 21 December 2005 Markets Definition and Market Power – Legal and Economic Aspects : The EU case Presented by . Rochdi ZOUAKIA

CoE/ARB Disclaimer The eventual views expressed are the authors’ own and do not necessarily represent those of the NRA he is representing , its staff, or any of its experts.

I. MARKET DEFINITION : The EU PROCESS One of the focal points of the new EU electronic communications networks and services regulatory package adopted in February 2002 and was due to take effect by 25 July 2003, is to address relevant marketsthat are characterised by a lack of “effective competition” (defined in terms of the existence of a dominant market actor or actors with Significant Market Power (SMP)). The identification of such relevant markets requires an economic assessment, consistent with Community competition rules, of the range of services which act as competitive substitutes to certain other services, so as to act as a competitive restraint on any single or collective market actor (s) wishing to raise price and/or limit production for those services.

The Commission had further informed this process also by publishing SMP ‘Guidelines’ on market analysis and the assessment of the concept of SMP. NRAs must take the “utmost account” of the Commission’s Recommendation and SMP Guidelines when analysing the relevant markets appropriate to their national circumstances and identifying SMP operators therein. Member States can only define markets that differ from those defined in the Recommendation where they satisfy the criteria set forth in Articles 6 and 7 of the Framework Directive. To this end, an institutional mechanism requires that National Regulatory Authorities (NRAs) conduct their market analyses consistent with the Commission's forthcoming ‘Recommendation’ identifying those product and service markets within the electronic communications sector which may be the subject of ex ante regulation, and a Decision on trans-national markets.

II. ECONOMIC AND LEGAL CRITERIA FOR MARKET DEFINITION II.1 KEY ELEMENTS OF SUBSTITUTABILITY ANALYSIS The process of market definition is primarily an economic exercise, which is nevertheless echoed in case-law precedent and administrative practice developed in the application of competition rules. Relevant markets have two dimensions under EC competition rules, namely: the product dimension; and the geographic dimension. Both product and geographic markets are delineated through an analysis of the inter-changeability of products from both demand and supply side perspectives in terms of the characteristics of the service (s) concerned, their price and intended use. Limited inter-changeability with other services results in the exclusion of those services from the market definition exercise, but should be taken into account at the later stage of calculating SMP.

Demand substitutability : It focuses on the obstacles which might prevent a customer from switching easily to another product or a supplier located elsewhere. Examples of such obstacles might include high investment costs, long-term contracts and costly terminal equipment. An analysis of supply substitutabilityalso needs to account for the obstacles which might prevent a supplier from switching production or other resources to the relevant products. The usual short-term period of assessing market parameters under competition rulesshould be tempered by the acknowledgement in the draft SMP Guidelines that the timeframe will inevitably depend on the characteristics of each market. The temporal element for ex ante purposes should, however, reflect the dynamism of technology-driven markets and the forward-looking nature of the market analysis process. A time period more than 18 months to two years might therefore be appropriate, particularly in light of the availability of ongoing periodic review of the market.

It may be possible, in certain circumstances, to identify specific customer segmentations or submarkets to which such products or services are targeted, particularly when such a group is subject to price discrimination (the practice of selling the same product or service to different customers groups at different prices, and being able to sustain such price differentials). Chain substitution. This occurs where it can be demonstrated that, although products A and C are not directly substitutable, product B is substitutable for both product A and C and therefore products A and C may be in the same product market since their pricing might be constrained by the substitutability of product B (i.e., where there is clear price interdependence at the ends of the chain and where the degree of substitutability along the chain is sufficiently strong).

A relevant geographic marketcomprises the area in which the undertakings concerned are involved in the supply and demand of the relevant products or services, in which area the conditions of competition are similar or sufficiently homogeneous and which can be distinguished from neighbouring areas in which the prevailing conditions of competition are appreciably different. • As an example, the geographic scope of the relevant market in the fixed sector has traditionally been determined by reference to: • the geographic scope of the network concerned; and • the presence of legal and regulatory obstacles.

II.2 THE HYPOTHETICAL MONOPOLIST TEST (1) The classic market definition exercise is usually conducted through the empirical “Hypothetical Monopolist Test”, otherwise commonly referred to as the “SSNIP” test (which refers to a small but significant non-transitory increase in price). This test considers whether, starting from the competitive price, it would be profitable over a period of about one year for an entity or entities to implement a hypothetical small (in the range of 5 to 10%) increase in price. This will depend upon the degree of consumer switching to other products (demand substitution) and additional supply (supply substitution). If such a price increase is not profitable, the products to which consumers would switch, or additional output, are included within the relevant product or geographic market.

II.2 THE HYPOTHETICAL MONOPOLIST TEST ALGORITHM (2)

However, there are limits to the efficacy of the Hypothetical Monopolist Test, whether in terms of: • the quantitative data needed to estimate the effects noted above are often not available; • certain characteristics of the communications sector will not always permit the application of the test, (e.g., where peering arrangements are used as a substitute for paid connectivity relationships, or where quality of service, and not price, exerts the greater influence on customer choice); • the test assumes that the price from which the percentage price increase made is at the competitive level but, where this is not the case, the test gives rise to the “Cellophane Fallacy” (i.e., substitution inevitably results where the price test is applied to a market price that is already set at a monopoly level; in regulated industries, it is often difficult, if not impossible, to determine the level of the competitive price).

The Cellophane Fallacy is a very familiar topic in antitrust literature. The expression comes from the famous antitrust case of the 50's when the United States Supreme Court made an overly broad definition of the relevant market, thereby failing to detect the market power of Du Pont, which held a virtual monopoly on cellophane.

II.3 QUANTITATIVE TOOLS TO MEASURE MARKETS The traditional legal and economic standards used to define relevant markets are best implemented by quantitative market measurementtechniques, on which regulators can rely for a more accurate understanding of the parameters of the relevant markets that are the subject of their review. For example, a market definition exercise may be supplemented with an analysis of price elasticities and residual demand estimates. A type arguably most useful for our present purposes is the “cross-price elasticity of demand” technique. Event studies, client studies and consumer studies also provide further means of defining relevant product markets. Such studies will comprise historical data,however, which means that a regulator should interpret such information cautiously. In many cases, even these types of information will not be available

In economics, the cross elasticity of demand or cross price elasticity of demand measures the responsiveness of the quantity demanded of a good to a change in the price of another good. It is measured as the percentage change in demand for the first good that occurs in response to a percentage change in price of the second good. For example, if, in response to a 10% increase in the price of fuel, the quantity of new cars that are fuel inefficient demanded decreased by20%, the cross elasticity of demand would be -20%/10% = -2. In the example above, the two goods, fuel and cars, are complements - that is, one is used with the other. In these cases the cross elasticity of demand will be negative. In the case of perfect complements, the cross elasticity of demand is negative infinity. Where the two goods are substitutes the cross elasticity of demand will be positive, so that as the price of one goes up the quantity demanded of the other will increase. For example, in response to an increase in the price of fuel, the demand for new cars that are fuel efficient (hybrids for example) will also rise. In the case of perfect substitutes, the crosselasticity of demand is positive infinity.

III. ENTRY BARRIER ANALYSIS AS THE BASIS FOR EX ANTE REGULATION Market failure should be left to be addressed by the market itself, if the relevant markets in question are not characterised by insurmountable entry barriers, that is markets will be contestable. Accordingly, the nomination of relevant markets for ex ante regulation (“Candidate Markets”) should be based on a comprehensive understanding of whether entry barriers are of such significance as to render the particular market in question unlikely to be characterised by effective competition, in the absence of ex ante regulation.

III.A. The Key Element of Ex Ante Analysis: Barriers to Entry and Contestability (1)

III.A. The Key Element of Ex Ante Analysis: Barriers to Entry and Contestability (2) A barrier to entry can be defined as a restriction on entry into the market which has the effect of allowing firms already in the market to charge prices above their otherwise competitive levels while not attracting new competitors. In the absence of barriers to entry, even a dominant firm can in theory be effectively constrained by the threat of competition; the universally acknowledged lesson drawn from contestability theory is that, under certain conditions, the threat of entry will force even a dominant firm to price at levels which ensure both efficiency in production and competitive pricing.

III.B. Barriers to Entry Justifying Inclusion as a Candidate Market • These arise when there is an absolute barrier to entry in the industry. Such barriers can take one of two forms: • legal or technological; or • regulatory.

IV. Approach used to assess significant market power Under the new Directives and section 78 of the Act, SMP has been newly defined so that it is equivalent to the competition law concept of dominance. Article 14(2) of the Framework Directive states that: "An undertaking shall be deemed to have significant market power if, either individually or jointly with others, it enjoys a position equivalent to dominance, that is to say a position of economic strength affording it the power to behave to an appreciable extent independently of competitors, customers and ultimately consumers." Further, Article 14(3) of the Framework Directive states that: “Where an undertaking has significant market power on a specific market, it may also be deemed to have significant market power on a closely related market, where the links between the two markets are such as to allow the market power held in one market to be leveraged into the other market, thereby strengthening the market power of the undertaking”. If the NRA considers no firm has SMP by itself or collectively, the markets will be found to be effectively competitive.

In NRA’s view the most important criteria for the assessment of SMP in the relevant markets are: • Market growth and market shares; • Future potential market shares; • Barriers to entry and expansion ; • Economies of scale and scope; • Countervailing buyer power {when it leads to the existence of effective competitive constraints between suppliers in a relevant supply market}. ; • -Access to capital markets. • Note that: a higher share by revenue is not necessarily indicative of greater market power. For example, the supply of more costly services than competitors would also be consistent with a higher market share by revenue than volume.

V.Example of market definition : Ofcom analysis of broadband markets OFCOM did proceed as following: firstly, the Commission’s approach to market definition is set out. This is followed by a discussion of Ofcom’s general approach to market definition. Next, definitions of the relevant retail market are considered insofar as they are logically prior to and affect wholesale market definitions. Then the relevant wholesale markets themselves are defined.

The concept of the ‘hypothetical monopolist test’ (HMT) is a useful tool to identify close demand-side and supply-side substitutes. A product is considered to constitute a separate market if a hypothetical monopoly supplier could impose a small but significant, non-transitory price increase (SSNIP) above the competitive level without losing sales to such a degree as to make this unprofitable. If such a price rise would be unprofitable, because consumers would switch to other products, or because suppliers of other products would begin to compete with the monopolist, then the market definition should be expanded to include the substitute products.

When conducting the SSNIP test with respect to current claimed willingness to pay for broadband internet access services, Ofcom estimates the relevant critical loss of customers, i.e. the required percentage reduction in demand for the SSNIP to be unprofitable. As the price rises above the competitive level the hypothetical monopolist’s revenue experiences two conflicting effects. It gains more revenue from customers paying the increased prices and it loses revenue it used to receive from customers who substitute away from the service as a result of the price rise. In addition to the effect of changes in revenue on the hypothetical monopolist’s profitability, it is also necessary to take into account the marginal costs that it saves by not supplying service to the customers who substitute away from the service.

Supply-side substitution possibilities have then been assessed to consider whether they provide any additional constraints on the pricing behaviour of the hypothetical monopolist which have not been captured in the demand-side analysis. In this assessment, supply-side substitution will be considered as a low-cost form of entry which could take place within a relatively short period of time. That is, for supply side substitution to be relevant, there would need to be additional competitive constraints arising from entry into the supply of the service in question, from suppliers who are able to enter quickly and at low cost, by virtue of their existing position in the supply of other services.

Asymmetric broadband internet access provided via ADSL enabled ‘metallic’ telephone lines, or through cable, has three distinguishing features or functionalities which are not available, in practice, using narrowband internet access and that as a group distinguish it as a higher quality service than narrowband internet access: 1/ the service is always on, i.e. no dial up is required. This feature allows the user to maintain a permanent connection to the network so allowing real time delivery of services such as email; 2/ it is possible to use both voice and data services simultaneously, whether they are provided together, for example over the same access route, or separately, perhaps using more than one access route; and 3/ it has a faster downstream speed than a dial up connection.

Ofcom notes that the 3 features (quoted in the previous slide) of broadband provide for a clear distinction from narrowband and that those features are frequently highlighted in marketing of broadband by the major players. Moreover, there is an argument that broadband is an ‘experienced’ good and that it can be expected to become increasingly differentiated from narrowband over the period of this review.

Ofcom considers that during the time period of this forward looking review, demand side substitution between narrowband, other services and asymmetric broadband internet access is likely to be limited for both residential and business broadband customers. Ofcom considers that broadband internet access constitutes a distinct economic market on the demand side and will continue to do so during the time period of this review. Supply side substitution at the retail level is not relevant to this market review, because it would not provide a constraint on suppliers of wholesale broadband services. Therefore, Ofcom currently considers that broadband internet access services constitute a separate and distinct economic market from narrowband internet access and other services. This will remain the case for the time period relevant to this review.

For the period covered by this market review, Ofcom considerered that mobile internet access is in a separate market from fixed broadband internet access. Internet access over mobile networks is not an effective demand side substitute for broadband internet access on fixed networks. Internet access on a mobile phone currently offers considerably less functionality than a fixed broadband network. For example, only a fraction of the internet is accessible over a mobile telephone and only part of this fraction is deliverable because of the constraints of screen size on mobile telephones, and interactivity is constrained because of the lack of a full-size keyboard. However, Ofcom recognises that the extent of substitutability might need to be reviewed in future following the take-up of new mobile technologiesoffering packet switched services, such as UMTS. Ofcom might, therefore, reach a different conclusion in a future market review.

there is evidence to suggest that the market is indeed national in nature, with local characteristics. The main reasons for defining a national market are that: BT has priced and continues to price on a national basis at both the retail and the wholesale level; the cable operators price on a national basis, even though their cable franchise areas are in geographically distinct locations; ISPs design and price their products on a national basis; and available evidence on advertising practices suggests that all operators in the broadband Internet access market advertise on a national basis.

Annex: Supplementary Market Analysis The derivation of the formula for the critical loss calculation used in the SSNIP test is as follows: Critical Loss Derivation Outline of methodology The critical loss is the percentage reduction in demand for which the SSNIP leaves profits unaffected. So a larger loss of demand than the critical loss makes the SSNIP unprofitable. The change in profit equals the change in revenue less the marginal costs saved (assuming constant marginal cost, c): (1) R1 – R0 = p1q1 – p0q0 – c (q1 – q0) where R is profit, c is marginal cost p1 = p0 (1 + s) where s is the size of the SSNIP q1 = q0 (1 + L) where L < 0 is the percentage loss of demand

c = a x p0 i.e. a is the ratio of marginal cost to the initial price. Substituting these definitions in (1), specifying R1 – R0 < 0 and rearranging gives the critical loss: (2) –s/(1+s-a) The critical loss is calculated as –s/(1+s-a) where s is the SSNIP (which Ofcom assumes to be 10%) and a is the ratio of the marginal cost to the current price (i.e. the price before the SSNIP, here assumed to be £25 per month (including VAT). Examples Assume s = 10% (i) c = 0% x poimplies = -9.1% (ii) c = 50% x poimplies = -16.7% (iii) c = 60% x poimplies = -20% (iv) c = 70% x poimplies = -25%

Conclusion • It is clear that markets definition and SMP assessment is a long process, requiring data gathering and consumers surveys; • the ‘hypothetical monopolist test’ (HMT) is a useful tool to identify close demand-side and supply-side substitutes. • This process is to be conducted through clear guidelines to be set by NRAs; • Public consultations are of great help to setting guidelines as well to conducting the whole process

Recommended Documents : In www.ofcom.org.uk : Review of the Wholesale Broadband Access Markets; http://europa.eu.int/information_society/policy/ecomm/doc/info_centre/studies_ext_consult/mkt_def_reg_obl_ssd/executive_summary.pdf http://europa.eu.int/comm/competition/liberalization/legislation/#telecom_regulations

ITU/BDT Arab Region al Workshop on