Chapter 2. Sector specific or general competition legislation

Face - Contents - Bottom - Previous/Next

"Telecompetition"

Chapter 2. Sector specific or general competition legislation

Introduction

The telecommunications sector is covered by general competition regulation like any other sector. But certain matters are subject to sector specific competition regulation as well. The need for sector specific regulation is closely tied to incumbents' control over infrastructure of key importance to entrants'/competitors' ability to offer services on retail markets. In absence of regulated access to the incumbents' infrastructure entrants would only be able to enter markets by building new infrastructure, which would raise entry barriers to almost insurmountable levels.

This chapter addresses the need for and scope of necessary sector specific regulation in the Nordic countries and describes the different institutional setups. The chapter begins with an explanation for why regulatory intervention is needed in order to create competition on services in infrastructure industries. Then, the telecommunication sector's structure is compared to that of other infrastructure sectors. This is followed by a description of major similarities and differences in regulatory setup and regulation across the five countries in the working group. Then, the previous and present principles for regulation of the sector are described and their effect on the nature and extent of regulation is evaluated. This includes a description of the issue of access price regulation and its relationship with investment incentives. Finally, the most prominent cases handled by the Nordic competition authorities according to competition law are described.

The need for sector specific regulation

Large fixed (often sunk) costs and low marginal costs are common features for infrastructure sectors. Such cost structure creates a risk of market failures and thereby obstacles for successful introduction of competition. Competition based on separate infrastructures faces a powerful opponent in significant economies of scale giving incumbents a seemingly natural monopoly. Historically, this has been the reason for placing infrastructure companies under public ownership. Liberalisation and full or partial privatisation of infrastructure sectors has thus created new challenges.

So far, experience has taught regulators that privatised incumbents in a liberalised market often are unwilling to provide third party access (TPA) to their infrastructure to downstream competitors and will do so only at prices that effectively prevent entrants from competing on a level playing field. Hence, in absence of regulation that ensures entrants' access to incumbents' networks at reasonable terms, the only way to enter an infrastructure based market would be by constructing new independent infrastructure. But the magnitude of necessary investments and long depreciation periods are likely to result in a financial strain on entrants that ultimately will deter entry.

In addition, new infrastructure will in many cases just replicate existing infrastructure. Depending on the retail market structure, this may lead to monopoly pricing in geographical areas only covered by the incumbent and unsustainably low prices in areas with double coverage of infrastructure as every operator will prefer a modest contribution margin from serving a customer rather than no margin at all by leaving the customer to a competitor. Consequently, prices will fall to levels about marginal costs and entrants will be unable to recoup their investment costs.

In cases where prices cover variable costs (however miniscule) but are insufficient to cover depreciation and amortisation of sunk investments, it is difficult for competition authorities to determine whether the low prices are the result of fierce competition or a part of a strategy employed by the incumbents to drive entrants out of the market. Consequently, competition law and jurisprudence are in many cases inadequate to secure emerging infrastructure competition.

Hence, the special characteristics of the infrastructure sectors have created a need for sector specific regulation. Sector specific regulation is ex ante in nature - contrary to competition law¹⁵ - and has been introduced to open markets for competition. Regulation has e.g. been used to secure non-discriminatory access to the incumbents' infrastructure. Incumbents are thus required to legally or virtually split up their wholesale and retail operations and act as if their retail departments were entrants.

Even if sector specific regulation may be used to promote competition and efficiency on markets opened up to competition, there are a number of arguments against the use of special legislation in general and against legislation regulating prices in particular. Specific sector regulation leads to higher administrative costs as the legislation is implemented, markets are monitored and infringements are sanctioned. Sector regulation can also distort incentives on the market. It may become profitable for companies to focus on non-desirable activities. The converse can also apply in the sense that activities desirable in themselves are no longer profitable due to sector regulation. Regulated operators are also likely to allocate resources to circumvent regulation.

In the case of telecommunications, sector specific regulation has been used not only to address specific problems or prevent them from arising. Regulation has also been used to create new markets. One prominent example is the division of fixed line services and fixed line access. Originally, services and access were bought from the same supplier (as is still the case on markets for mobile services). Opening the market for competition would thus imply enabling competitors to supply a similar product. But to facilitate entry and to create competitive pressure on profits as quickly as possible, services and access were unbundled thereby creating separate markets.

Is the telecommunications sector different from other infrastructure sectors?
It has often been argued that telecommunication markets are particularly complex compared to other infrastructure markets such as railways, electricity, gas supply or postal services. Advocates of telecommunication sector's alleged unique complexity thus argue that the need for regulatory intervention in the telecommunication sectors is incomparable with the regulatory regimes in place to secure competition in the other sectors. The telecommunication sector is admittedly different in one important aspect. An impressive technological development during the past 10-15 years has reduced network construction costs significantly. Moreover, development of new wireless technologies has created better opportunities for network competition where natural monopolies previously prevailed. Similar opportunities are unlikely to arise in the markets for e.g. electricity, gas or railways. However, wireless technologies have not proven viable alternatives to existing fixed wire networks so far and network competition only appears financially feasible in geographically restricted areas.

So while the nature of services provided and the techniques used to provide them differ across the sectors; many characteristics are common seen from a regulatory point of view. Despite a unique technological development on telecommunication markets compared to other infrastructure sectors they all share the same feature: only one company has a distribution network that permits it to service the whole national market and competition is not feasible in absence of some kind of third party access.

Different approaches are possible to address a situation where a dominant infrastructureowner can prevent sustainable competition from arising through its control over essential infrastructure. The most drastic solution is to split up the incumbent into two legally and financially independent companies: an upstream network operator and a downstream service provider. The downstream service provider then has to buy access to the network on equal terms with entrants and downstream competition can emerge at a completely level playing field.

An alternative to actual separation of the incumbent is accounting separation, which – in theory – can yield similar results. The incumbents thus remain vertically integrated but "buy" access to their own infrastructure on equal terms with entrants. Public regulators then rely on balance sheets provided by the vertically integrated operator to verify that the company actually acts as if it had been vertically separated. But discrimination between a vertically integrated network operator`s own retail department and independent retail operators comes in many forms. Accounting separation therefore has to be supplemented by supervision of all other terms for access to the relevant network as well.

Accounting separation is the dominant paradigm for access regulation across the Nordic countries in the field of telecommunication. The vertically integrated incumbents are subject to a number of obligations including TPA, price control, service provision etc. This regulatory regime has to a large degree facilitated the emergence of new and vibrant competitive markets. The market setup for various infrastructure sectors is described in table 2.1:

Table 2.1: Sector overview of vertical separation and third party access

Postal services are only completely liberalised in Sweden and concerns about vertical separation and TPA are therefore purely academic for the remaining countries. In Sweden, third party access is given to a limited part of the network namely public mailboxes.

With regards to the other sectors, incumbent railway operators have generally been vertically separated as required in common EU legislation. So far, TPA has only been modestly successful in the railway sector. One has to be careful however not to judge vertical separation too harshly on these grounds since railways have experienced general decline in business and still encounter serious technical barriers to cross-border operations. Railway operators' ability to compete with e.g. trucks is thus seriously hampered on long distances where railways otherwise could enjoy a relative advantage. The liberalisation of the railway sector has so far led to a limited part of passenger traffic put out to tender.

In Norway and Finland, vertically integrated electricity suppliers are subject to requirements of accounting separation¹⁶ whereas in Denmark and Sweden, distribution networks for electricity are run by independent entities. Vertical separation is currently being considered in Finland for operators currently supplying 85 percent of the electricity consumed. In Norway, where liberalisation was completed already in the early nineties, consumers seem to have benefited from the market opening. Electricity markets in Sweden were liberalised in 1996 and consumers can now choose from a range of suppliers. In Denmark, retail markets for electricity have only been liberalised recently so it is still too early to conclude on the outcome of liberalisation. A pan-Nordic cross-border market for electricity has taken off in a modest scale. But transmission capacity between the countries is still too limited to really put local national suppliers to the test when consumption peaks during the day. It therefore appears to be too early to draw conclusions on advantages and drawbacks from the various structural setups chosen in the various countries.

Distribution networks for natural gas have been separated from downstream suppliers into a publicly owned company in Denmark. For offshore transmission, vertical separation has been implemented in Norway as well. For inland distribution vertically integrated gas companies in Norway are required to establish accounting separation of network and retail services. In either country, markets are too immature to evaluate the effects of liberalisation and choice of structural setup.

Experience from other infrastructure industries shows that several ways to promote TPA are practically applicable. Vertical separation is thus certainly a feasible alternative to accounting separation in some cases. But even though vertical separation provides the best prerequisites for retail market competition at equal terms for entrants and incumbents neither of the countries in the working group has opted for this solution.

One of the reasons why policy makers across Europe have refrained from vertically separating incumbents when markets were liberalised is that vertical separation can have serious drawbacks. One argument against vertical separation of incumbents is the presence of economies of scope. A vertically integrated company has much better information than a stand-alone network operator about the need for investments in infrastructure to satisfy enduser demand and can thus respond more quickly and appropriately to changes in market conditions. This is a particularly valid argument in the telecommunication sector because of the rapid technological development experienced during the past 10-15 years and the plurality of new products being introduced as a consequence thereof.

However, the importance of economies of scope depends on the level at which vertical separation is carried out. Economies of scope are modest if e.g. only collocation facilities and the local loop are administered by a vertically disintegrated operator. Investments in those two parts of the network only constitute an insignificant part of overall investments, and investment concerns could therefore only have carried little weight if regulators had opted for this solution.

Today, the discussion of possible vertical separation of incumbent telecommunication operators is in many cases merely of academic interest. As governments have sold part of or all their stocks in the incumbent operators, structural changes to promote public welfare to the detriment of stockholders will be tantamount to expropriation. Vertical separation of the incumbents is therefore difficult to implement even in cases when one or more governments hold a controlling interest. In none of the five countries do governments have complete ownership of the incumbent operator, cf. table 2.2.

Table 2.2: Government ownership of shares in incumbent operators as of June 2004

Governments in Norway and Iceland hold controlling interest in the incumbent operators. The Swedish and Finnish governments hold a controlling interest in TeliaSonera together. Only in Denmark does the government no longer hold any stocks in the incumbent operator.

All the Nordic countries have thus implemented third party access to fixed line telecommunication networks but have maintained vertical integration. As vertically integrated operators seldom are inclined to grant access to their networks at prices and terms that permit entrants to compete away the network providers' retail profits, government intervention is required to secure a level playing field at retail level. Designated government authorities have therefore been instituted across Europe to handle this task.

Vertical and horizontal regulators

Independent regulatory authorities
In all five countries, independent regulators are responsible for promoting competition on telecommunication markets through regulation of access to the incumbent operators' networks. As some governments still have significant financial interests, at times even controlling shares, in some incumbent operators, it is important that an institutional setup that not only works impartially but also is perceived as impartial by current and potential entrants is in place.

Keeping regulators isolated from political pressure is thus an obvious concern and establishing independent regulators to regulate and monitor the terms for access to the incumbents' networks is consequently an essential feature for successful liberalisation. The institutional setup in the five countries is shown in table 2.3.

Table 2.3: Regulatory institutions and ability to appeal

In all the five countries, regulators are independent of policy makers represented by government ministries. In Denmark and Iceland decisions by the NRA's can either be appealed to designated appeals bodies or directly to the courts of justice. In Sweden the first instance of appeal is the courts of law whereas in Norway and Finland government ministries have powers to overturn a decision taken by the regulatory body.

In Sweden, Norway, Iceland and Finland, political responsibility for the sector specific regulation is vested in ministries also responsible for other infrastructure related issues such as transport. Only in Denmark is the policy maker a ministry without similar cross-sector responsibilities.

Some operators have argued that government control over the incumbent operator leads to lessened regulatory pressure on the incumbent from the NRA. The comparisons of retail margins in chapter one contradict such assumptions however. A high gross margin implies better possibilities for entrants to profitably enter a market and undercut the incumbent's prices. In Sweden and Norway gross margins on PSTN offered on the basis of access to the local loop are among the highest in the region and Norway has the second highest gross margin on local calls. Claims of favouritism are thus not supported by price comparisons.

In neither of the countries in this working group have competition authorities and regulatory authorities been merged or regulatory authorities been given competition powers as e.g. in Great Britain. There are several arguments in favour of and against vesting sector specific regulation and general competition regulation in one public authority.

Telecommunication markets are in many aspects different from other markets regulated solely by competition authorities. Competition authorities usually address market failures or abusive behaviour on markets where basic prerequisites for competition are present. In the case of telecommunication markets, the task was (and to some degree remains) to dismantle public or private monopolies, i.e. to create a market rather than address a specific abusive behaviour.

The nature of interventions carried out by regulatory authorities and competition authorities is also a key issue. Sector specific regulation often implies detailed regulation of prices and other terms for access to essential infrastructure. Such types of regulation are not common within competition jurisprudence. General competition regulation and sector specific telecommunications regulation thus build on quite different paradigms. This may favour establishment of separate authorities.

But the institution of two authorities who both regulate competition in a given sector may give rise to difficulties. In small countries, the number of employees possessing the relevant skills is likely to be modest making authorities vulnerable to employee turnover. Such problems are diminished if one single authority is responsible for both sector specific and general competition regulation.

The new regulatory framework¹⁷ has accentuated this problem. The market analysis employed by NRAs to appoint SMP-operators are to be carried out according to principles normally employed when defining relevant markets and determining whether a supplier holds a dominant position according to competition law. This may favour a strengthening of the NRA's competencies in this field. Likewise, regulation of for instance access often requires technical knowledge. In this regard NRAs can benefit from their other technical tasks. But competition law and jurisprudence is in constant development as experience builds from all sectors covered by general competition regulation handled by general competition authorities. It is therefore first and foremost imperative that both authorities take part in the analysis benefiting from the authorities' complementary strengths.

Other types of relevant regulation and regulators Other kinds of regulation with other objectives than to promote competition also affect market development. Regulation of consumer rights and rules protecting intellectual property can have a bearing on the development of competition on e.g. distribution of digital content. Both consumer regulation and regulation of intellectual property can be vertical as well as horizontal.

The telecommunications sector is thus regulated by several authorities applying different sets of legislations. Besides regulatory authorities and competition authorities, agencies regulating e.g. consumer rights also influence policies in this field.

A comparison of the resources spent on regulation is more likely to illustrate the diverse division of responsibilities between the relevant authorities in the different countries than to shed light on the authorities' efficiency. All the regulatory authorities are financed entirely or partially by fees paid by operators. In Denmark, fees charged by the regulatory authority are supplemented by government appropriation. Competition authorities and consumer agencies are financed by government appropriation in all the countries. Budgets and number of employees regulating competition on telecommunication services in relevant government authorities are shown in table 2.4.

Table 2.4: Resources spent on regulation (2003)

Note 1: For regulatory authorities telecom cases comprise dispute mediation, imposition of sector specific competition regulation, handling of complaints over incumbent in regulated areas.
Note 2: Number of employees doing telecom cases at the regulatory authority in Denmark is estimated on the basis of the Danish government budget for 2003.

Sector specific private consumer boards have been established in both Denmark and Norway. In both countries, the board's expenses are covered by the industry. The industry is thereby given an extra incentive to address the underlying reasons for the complaints. For the industry, running the board itself implies an additional advantage, since the industry sooner than when public authorities are responsible for the case-handling gets a more accurate impression of which problems are most important and what kind of information consumers demand. It is not possible though to allocate all complaints to private boards since some problems of a general nature (e.g. marketing practices) often are covered by horizontal consumer regulation.

In most cases, cooperation between authorities is of an informal nature where information is exchanged and the specific delineation of responsibility for a given case is determined. There are no examples of cases with joint intervention from both NRA and NCA. If a case touches upon issues covered by different sets of legislation the relevant authorities deal only with the issues within their field of responsibility. So far, no difficulties worth mentioning have arisen in the allocation of cases between the relevant authorities.

Convergence between digital platforms is no longer just a technological opportunity but reality. Cable-TV networks offer Internet access in most Nordic countries and IP-telephony has already been introduced to business customers and in some countries to residential customers as well. Relatively high broadband penetration rates and increasing bandwidth are paving the way for video-on-demand services and even television distributed over PSTN. These developments make it increasingly difficult to define telecommunication and broadcasting as distinct sectors.

The blurring of boundaries between telecommunication and broadcasting raises questions about how the two sectors are regulated. Only in Finland is broadcasting content in part regulated by the same policy maker as telecommunication. In all the other countries, authorities or ministries separate from the telecommunication regulator are responsible for content regulation.

Horizontal integration of previously distinct industries warrants regulatory reforms aimed at more horizontal regulation. Such reforms do not necessarily imply merging of authorities or redistribution of regulatory responsibilities but at the very least an increased awareness from all interested parties.

Self-regulation
Self-regulation is encouraged by regulators in some Nordic countries. The industry thus gets a chance to find suitable ways to regulate issues of common interest without government intervention (often within a framework defined in regulation though). One of the advantages of self-regulation is that the industry may have a better understanding than a regulator of how a particular problem is best tackled, which may lead to more appropriate regulation than otherwise would have been the case. Self-regulation should not be mistaken for laissez-faire. Only cases where the industry unanimously adopts measures to avoid public regulatory intervention can be considered self-regulation. Use of self-regulation varies across the Nordic countries as shown in table 2.5.

Table 2.5: Use of self-regulation in the Nordic countries

The scope for self-regulation is limited though by differences of interests for incumbents and entrants as well as incumbents' relatively strong negotiation positions vis-à-vis entrants. The threat of regulatory intervention in case the industry fails to reach an agreement is therefore necessary to countervail incumbents' superior negotiating power and thereby provide an as level playing field as possible.

Principles for and extent of sector specific telecommunications regulation

Sector specific regulation is only called for, when competition law is inadequate in creating and maintaining effective competition. As sustainable competition gains foothold and as market failures are reduced or eliminated the need for sector specific regulation falls away and superfluous regulation can even be detrimental to effective competition. General competition regulation has been a powerful tool to combat different types of abusive behaviour by the incumbents in the Nordic countries, cf. the section "Nordic competition case law".

The European Commission explains when sector specific regulation is warranted in its explanatory memorandum for the relevant markets (page 11): "Ex ante regulation would be considered to constitute an appropriate complement to competition law in circumstances where the application of competition law would not adequately address the market failures concerned. Such circumstances would for example include situations where the compliance requirements of an intervention to redress a market failure are extensive (e.g. the need for detailed accounting for regulatory purposes, assessment of costs, monitoring of terms and conditions including technical parameters etc) or where frequent and/or timely intervention is indispensable, or where creating legal certainty is of paramount concern."

It can be of great relevance whether a market is regulated according to general competition regulation or sector specific regulation. Some telecommunication markets have proven very dynamic and technological progress has been the driver as well as the result of introduction of competition in the sector. Innovation and digitalisation of content and services are blurring the boundaries of previously sharply separated markets. If different competition legislation is applied on the telecommunication markets compared to e.g. media markets, operators from the two different market segments will not face a level playing field if competition arises between telecommunication and media companies.

Yet another risk is that sector specific regulation of the telecommunication markets might distort incentives for investment and innovation on the regulated markets. Market development may thereby be diverted in another direction than what would have been the case if general competition regulation had applied. Sector specific regulation should therefore always be kept at a minimum and leave the greatest possible scope for flexible and unhindered development of wholesale as well as retail products. In some cases, this implies a trade-off between maintaining regulation that can promote competition further and repealing regulation in order to lessen distortions. The same applies in cases where sector specific regulation addresses other issues (e.g. consumer protection) otherwise covered by horizontal regulation.

Some of these concerns are addressed in the new regulatory framework currently being implemented across the EEA countries. In the following, the new regulatory framework is briefly described and compared to the regulatory framework it replaces.

The previous and the new regulatory framework
The common European push for competition in the telecommunications sector was initiated in 1984 as the European council decided to promote liberalisation of markets for telecommunication services. The next significant development was in 1990 when it was decided to grant access to national incumbents' networks (open network provision) to entrants to facilitate retail market competition (Council Directive 90/387/EEC). At the same time (in Council Directive 90/388/EEC) a timetable for when different markets were to be liberalised was laid down.

Later amendments of Council Directive 90/388/EEC (particularly Directive 96/19/EC) lead to full competition by 1998 on all retail markets in the European Community and paved the way for convergence between PSTN and cable-TV networks (Directives 95/51/EC and 1999/64/EC). A selection of some of the most important directives is given in table 2.6:

Table 2.6: Liberalisation and harmonisation (98 Regulatory Package)

Even though EU initiatives for market liberalisation have been the model for market opening in all the Nordic countries, the liberalisation has followed different trajectories across the region. In Finland, the first steps to liberalise the markets were taken already in 1988 when markets for data transfer were partially opened for competition. The Finnish process of liberalising the terms for data transfer was finalised in 1990 when the market was completely deregulated. Finland thereby preceded the rest of the Nordic countries. Deregulation and liberalisation milestones are summarised in table 2.7.

Table 2.7: Liberalisation milestones

Source: Nordic competition authorities

The interconnection directive (97/33/EC of 30 June 1997)¹⁸ and the regulation on unbundled access to the local loop (2887/2000/EC of 18 December 2000) are of particular importance. In the interconnection directive the criteria for determining whether companies were to be regarded as having SMP were defined. The regulation prescribing entrants' access to the local loop gave entrants the opportunity to offer fixed line subscriptions and broadband access by using the incumbents' infrastructure but based on different technological platforms than the incumbents'. Entrants' independence of incumbents' technological platforms is essential for their capabilities to compete on product diversification. The best example is broadband access where entrants in many cases introduced broadband access before the incumbents did.

A review of the European regulatory framework for electronic communications began in 1999 with the adoption of the Communication Review. In March 2002 the Parliament and Council adopted a new package of sector specific regulation designed for more competitive markets and converging electronic communications technologies. The changes in regulatory framework are intended to facilitate roll-back of sector specific regulation as soon as competition emerges. A selection of the directives constituting central parts of the regulatory reform is given in table 2.8:

Table 2.8: Central directives in the 1999 Communications Review

Cross-border coordination is a demanding but necessary task. Only in this way can authorities reap the maximum benefits from each others experiences and create the prerequisites for a common market. Previously, coordination between NRAs was carried out within the Independent Regulators Group (IRG). But to induce a uniform use of the new regulatory framework across Europe a new cross-border network comprising NRAs and the European Commission (European Regulators Group or ERG) has been instituted (2002/627/EC). IRG and ERG now operate in parallel.

The new regulatory regime is only gradually being implemented. All the Nordic countries have implemented the community directives in national legislation within the timeframes envisaged in the regulation (2003 for EC members and 2004 for EEA members). NRAs are responsible for the market delineation and analysis of SMP in all the Nordic countries. This is an ongoing process and it is yet too early to tell whether it will lead to more or less regulation.

The new regulatory framework changes the criteria used to establish who and what to regulate as well as how the companies and markets in question can be regulated. One of the most significant differences between the new regulatory framework and the framework it replaces concerns the underlying principles for market analysis used to determine which companies qualify for regulation; companies with so-called significant market power (SMP).

Previously, companies with a market share in excess of 25 percent for a product defined in the sector specific regulation were to be subject to regulation. Other criteria were the company's turnover in relation to market size and the company's ability to influence market conditions. According to the new regulatory framework, only operators who posses and are expected to retain a dominant position on a relevant market, both as defined in competition law and jurisprudence, can be subject to regulation. It usually takes a market share in excess of 40 percent for a company to hold a dominant position according to competition jurisprudence. This higher threshold for market share will in itself restrict the group of companies qualified for regulation.

Passing from calculating market shares on the basis of single products to calculating them on the basis of product markets potentially encompassing competing products can likewise lead to lower market shares for candidates for regulation and thereby limit the number of cases in which sector specific regulation is applicable. In this way, regulators have to take new technologies and products into consideration when analysing the basis for regulation. The new principles for market analysis in the regulation of markets in the telecommunication sector should thus lead to a more technology neutral regulation as competitive pressure from products based on different/new technologies are taken into consideration when the possible need for regulation of the market for a given product is evaluated. Market dynamics are therefore better reflected in market analysis conducted under the new regulatory regime and a smoother and more rapid deregulation is facilitated as competition, possibly through new products, evolves.

Another major difference between the initial regulatory regime and the regulatory regime currently under implementation across Europe is the way competition problems detected in the market analysis are sought redressed. Previously, SMP-operators were automatically subjected to a full range of remedies (obligations vis-à-vis competitors). This approach has lead to a high degree of transparency for entrants and incumbents alike and has been easy to administer by regulatory authorities. Technological development, emergence of competing infrastructure and horizontal convergence have made it necessary to change this approach as the problems to be addressed to a much greater extent than before differ across both countries and markets within the industry. The one-size-fits-all approach has therefore been replaced by a requirement that regulatory authorities tailor regulatory intervention so that only remedies that are strictly necessary to redress problems detected are imposed on SMP-operators. This change of approach will add to regulatory uncertainty but is likely to lead to an overall reduction in the extent of remedies employed. On the other hand there is a risk that regulators may impose too many ex-ante obligations in order to prevent abuse of market power by SMP- operators. This can lead to an overloaded bureaucracy and high administrative costs.

The ambition with these new principles is to ensure that regulation better reflects market conditions and addresses the entrants' needs while at the same time leaving as much room as possible for the incumbents to compete. The main differences between the two regulatory regimes are summarized in a somewhat simplified manner in table 2.9.

Table 2.9: previous versus current regulatory regime

The regulatory reform will thus bring a greater coherence between horizontal competition regulation exercised by competition authorities and vertical competition regulation administered by regulatory authorities as sector specific regulation to a far greater extent than before will rely on general competition regulation.

Will the future bring more or less regulation?
As described above, the new regulatory regime is expected to provide a better basis for a normalisation of regulation of the telecom markets. But in the short run the result may turn out to be more regulation. Use of general competition law analysis, which the new regulatory framework is based upon, is more difficult when applied ex-ante than ex-post and there is no guarantee that a common analytical framework will create consistency. In order to prevent the abuse of market power by SMP-operators, there is thus a risk that national regulators impose too many ex-ante obligations which, in turn, could lead to an overloaded bureaucracy and high administrative costs.¹⁹

The relevant markets in which ex-ante regulation may be justified have been identified by the European Commission on the basis of three criteria, namely existence of high and non- transitory entry barriers, absence of effective competition and the insufficiency of competition law in achieving or restoring effective competition. Emerging markets should in principle be free from ex-ante regulation.

The current regulation of the 18 markets appointed by the European Commission for future regulation is summarized in table 2.10 to shed light on what can be expected as the new regulation is introduced:

Table 2.10: Number of operators subject to regulation prior to the new regulatory framework

Note: *subject to USO which theoretically does not depend of SMP status

Finland is in some aspects a unique case with its history of local monopolies that has brought a plurality in suppliers not seen in other countries. The higher number of SMP-operators in Finland compared to the other countries thus does not mean that competition is relatively poor in Finland but that the country is split up into smaller separate markets instead of having one national market. The challenges remain the same as in other countries because a local monopoly controlling infrastructure necessary to potential competitors has exactly the same possibilities to exclude competition on its "home market" as a national monopoly. When assessing the need for regulation it thus makes perfect sense to compare Finland with the rest of the Nordic countries.

A comparison between the old and the new regulation is never completely accurate as markets delineated according to the new regulation do not always correspond to the products or markets used under the previous regulation. But table 2.10 never the less gives an indication of the potential extent of the new regulation compared to the regime it is replacing.

Most markets designated for possible future SMP-regulation were also regulated under the previous regulatory framework. In most countries, markets for international roaming (market 17) and markets for broadcasting transmission services (market 18) were not regulated previously. Regulation of markets for mobile voice call termination (market 16) in Sweden and Finland has in part been inspired by the new regulatory framework though they preceded its introduction whereas regulation of mobile call termination in Norway dates back to 1998, i.e. before the new regulatory framework. The enlargement of the scope of regulation is likely to lead to more regulation in the short as well as the long run and Denmark is likely to follow the example set by its Nordic colleagues and regulate market 16.

As regulatory intervention may only be used when general competition legislation is insufficient to redress market failures, future need for regulation of market 17 (international roaming) depends on the development of competition jurisprudence. The European Commission has investigated possible excessive pricing on the wholesale market for international roaming in the UK²⁰. If the European Commission successfully intervenes against high wholesale prices using article 82 of the EC-treaty, regulatory intervention against pricing at these markets may therefore be un-called for.

An example of a market that is likely to become subject to less regulation is the market for access and call origination for mobile telephony. The number of SMP-operators appointed according to the old framework range from 2 to 3 across the Nordic countries. Under the future regulation, NRAs can only appoint more than one SMP-operator on a market if two or more operators have collective dominance, which is rather difficult to prove.

Even though the threshold for SMP-operators' market shares has been raised the new regulatory framework is likely to bring more regulation at least in the short term in most countries. Potential additional regulation is attributable to two effects from the new regulation. First, more markets have been appointed as candidates for regulation under the new regulatory framework than under the regulatory framework being phased out (including markets for mobile call termination, international roaming on mobile networks and broadcasting transmission services). Second, whereas the previous regulation primarily focused on the incumbent operators, the new regulation is neutral in its approach. As a consequence, entrants will probably face regulation of e.g. termination rates on fixed line as well as mobile telephony. So far, only the incumbents have been subject to such regulation in most countries.

The extension of regulation to more markets and in some cases more operators is due to a greater awareness of how markets work (see box 2.1) as well as different requirements for regulation as a consequence of more mature markets for e.g. mobile services. In the short term, NRA's will thus face greater responsibilities depending on the remedies NRAs choose to redress the competition problems identified in the market analysis. The more remedies are imposed the fewer cases are likely to arise but at the cost of stiffening market development as remedies implicitly may bring e.g. minimum standards for regulated products and upper limits on prices.

Box 2.1: Entrant squeezes incumbent

The liberalisation of telecommunications markets has thus generally required, and still requires, sector specific regulatory intervention across the Nordic countries as well as in the rest of Europe to secure market participants' access to essential infrastructure.

Remedies imposed on SMP-operators
Possible remedies that a regulator can impose on a SMP-operator include for instance transparency, non-discrimination, accounting separation, third party access, price control. All remedies aim at making incumbents and entrants as equal as possible when they compete at retail level.

In many cases, the remedies are difficult to use for detection of foul play. Both because annual accounts are finished late in relation to the period of time they cover but also because markets often develop and the appropriate delineation of markets used in the accounts therefore may change. Entries used for accounting separation therefore seldom correspond to relevant markets as intended in competition jurisprudence. As accounting separation in many cases is inadequate to discipline incumbents, correct wholesale price regulation becomes even more important because of wholesale prices' influence on incumbents' incentives to/benefits from attempts to foreclose entrants at retail markets.

Price regulation is not only one of the most important tasks; it is also one of the most difficult ones. In all the Nordic countries, access prices shall allow the incumbents to recover costs including a reasonable return on capital. But there are several different approaches as to how to determine costs using various cost bases and cost standards. A selection of different definitions of cost bases and cost standards is given in table 2.11.

Table 2.11: Principles for costs estimation

Different combinations of cost bases and standards are possible. The different approaches used in the Nordic countries are shown in table 2.12 for fixed line interconnection and local loop unbundling.

Table 2.12: Cost methodologies for calculating interconnection and unbundling charges

Source: the ninth implementation report.

Table 2.12 illustrates that setting prices equal to costs involves a range of choices by regulators with more than one possible correct answer as it is important to keep in mind that balanced investment incentives for both entrants and incumbents are just as important as low prices for entrants' access to incumbents' infrastructure.

When deciding how to allocate costs and who should cover them, regulators often turn to the principle of causation: expenditures should be covered by the entity that causes them to arise. This can lead to distortions in cost structures for entrants compared to incumbents and it is indisputable that entrants are handicapped by these practices as indicated in the following examples.

- In some countries, entrants are charged to cover the costs for being monitored by an incumbent employee when entering a collocation site. Allegedly, the surveillance is necessary to protect the incumbent and others using the collocation site from risk of damages to their equipment. Incumbents do not pay entrants for a similar surveillance service when incumbents access collocation sites.

- When entrants wish to exchange traffic with an incumbent, it is necessary to establish a transmission line connecting the two networks. In Denmark, entrants have to cover the total cost for the construction of this connection. This in part reflects that entrants need interconnection with the incumbent far more than the other way around. In Denmark, costs for construction of connections for interconnection between mobile networks are divided according to the volume of traffic sent by either company.

- In some countries, incumbent operators' interconnection rates have so far been regulated allowing incumbents to charge interconnection rates at three different price levels depending on where in the network hierarchy the call is received: local, single transit and double transit. Entrants' interconnection rates have so far been considered a commercial matter. In e.g. Denmark, the incumbent only pays entrants termination charges corresponding to incumbents own local termination rate irrespective of the termination service delivered by the entrant. The incumbent argues, supported by legislation, that calls terminating in another network often are routed differently than calls within the incumbent's network thereby creating additional costs for the incumbent. These costs are covered by only paying for local termination even when single or double transit termination is delivered by the entrants. Even though entrants face similar costs when their customers call customers connected to the incumbent's network entrants still pay full price for termination.

The examples given above illustrate a dilemma for regulators: efficiency at the cost of equal terms or vice versa. By allowing entrants to enter collocation sites a need for surveillance or access control is created. If costs are solely borne by entrants they face an uneven playing field. But to ask incumbents to pay as well is tantamount to asking the incumbent to defray extra expenses because an entrant uses the incumbent's site to compete with the incumbent. An advantage of the current regime for cost distribution between entrants and incumbents is that only entrants who are sufficiently efficient to cover all costs associated with introducing competition will enter the market. In other words: one avoids inefficient entry. A drawback of the current regime is that incumbents have no interest at all in carrying out investments that would reduce overall inefficiencies associated with operating more than one network, as the costs of these inefficiencies are borne by entrants alone. Current inefficiencies will therefore not be redressed.

Regulating access on markets undergoing rapid technological development

Telecommunications markets in Scandinavia and Europe are developing rapidly as technological development and increased demand for new services continuously change market conditions.

Because of the ever-changing market conditions national regulators face difficult tradeoffs particularly concerning access pricing. Access pricing is particularly complex because it is intended to promote several, and at times apparently conflicting, goals. In order to facilitate entry access prices should be as low as possible. But low access prices take away entrants' incentives to invest in new infrastructure. And without competing infrastructure markets will never be able to work unassisted by regulation. At the same time access prices are supposed to reflect incumbents' underlying costs in order to prevent inefficient entry and secure cost recovery for incumbents. Designing access price regulation is thus all about giving both entrants and incumbents the right investment incentives while keeping entry barriers low²¹. This is discussed further in the following with special attention as to how the unbundled local loop should be priced.

How best to promote competition over the local loop has been a controversial topic in telecoms regulation. Ideally, competitors would put an end to the incumbents' local loop monopolies by building their own networks. But building networks with coverage as the incumbents' networks is immensely expensive and time-consuming. Cable-TV networks generally provide coverage only in some areas (mainly the major cities) and mobile networks cannot yet offer broadband Internet access, at least not with full geographical coverage and the same data transmission speed as the fixed line network. Therefore, incumbents have been forced to share their fixed access networks with entrants.²²

Chang, Koski and Majumdar (2003) have examined access pricing regimes for local exchange carriers in the US and Europe. In addition, the relationship between various aspects of regulatory and institutional policy changes in Europe and access prices has been analyzed. Based on data from US, the authors find that lower access prices have promoted deployment of digital technology among US incumbent local exchange carriers. While the results for the EU are not statistically strong, being based on limited data, they suggest that access prices are lower where competition is more developed, where there is an independent regulatory authority and where retail price caps are in place. The study also finds that in countries where the access charges are higher than the EU average, telecommunications operators as a whole allocate more money towards investments. This is somewhat contrary to the findings established for the US. However, the authors underline that the European data are relatively sparse and likely to be insufficient. They also reflect institutional heterogeneity across nations. One therefore has to be careful about drawing too categorical conclusion on the relationship between different approaches to interconnection price regulation and investment behaviour in Europe.

As network costs to a great extent are fixed, it may appear reasonable to have fixed access charges. However, in order not to discriminate between small and large operators different combinations of fixed and variable charges are usually used. According to Peitz (2003) there are five general aims in the design of regulation. Regulation should:

1. Require as little information and data from market participants as possible.
2. Keep costs of regulation low and avoid an overloaded bureaucracy.
3. Ensure that regulatory measures are temporary rather than permanent and that superfluous regulation is repealed.
4. Ensure static economic efficiency with a particular focus on improving consumers' surplus.
5. Ensure dynamic efficiency so that incentives to invest give rise to socially optimal decisions.²³

As pointed out by Valletti (2003), access pricing regulation has been analysed in a static context but there is no developed analysis of the linkage between access pricing and incentives to invest. However, one can make some conjectures about the factors that should influence the choice of regulatory policy by drawing on literature in related areas, e.g. literature about research & development. One important factor is flexibility: entrants should be allowed to choose from different entry modes with possibly differentiated price structures and incumbents should be allowed to set different access charges subject to average constraints in the form of price caps. Another important factor is that willingness to invest depends on the perceived risk of investment.

Regulation of access prices and terms affects the return facility providers can expect on their investments. The more often regulation is updated to take into account e.g. recent market developments; the less predictable are access prices and terms faced by market participants. Consequently, expectations of the nature of future regulation affect incentives to invest and to upgrade networks. It is therefore, according to Valletti, advantageous to establish a set of (possibly changing) access pricing rules in advance or at least to commit the regulator to follow a set of objectives or criteria in advance.

A regulator can thus reduce investment risks by committing to an access pricing regime. Such a policy can also facilitate the gradual development of entrants' networks. When investing in new networks entrants need to be as certain as possible that future regulatory intervention against the incumbent does not erode the expected benefits from the investment.

One way to address the potentially conflicting objectives when access prices are determined (low prices to promote entry and high prices to promote construction of new infrastructure) is a model of dynamic prices, i.e. prices that change over time following a pre-determined trajectory. A model with dynamic access prices could promote competition in the short term and still provide incentives to invest in telecommunications infrastructure in the medium to long term, i.e. both static and dynamic efficiency would be achieved.

Cave and Vogelsang (2003) question that the best way to stimulate infrastructure investment is to have universally high access prices. According to the authors the matter must be considered in terms of the dynamics of competition not only between the incumbent and entrants but also among entrants. In addition, the dynamic process through which a single entrant progressively establishes itself in business is important. Experience shows that entrants do not immediately emerge as fully fledged facility-based competitors. A policy of high access prices in the early phase of the entrants activity may, according to the authors, "kill the entrants' business model stone dead". Constantly low access prices are, however, not a realistic long term solution as they deny cost recovery to network investors and thereby eliminate incentives to invest. The authors consider how to set access prices to ensure that both incumbents and by entrants are given an incentive to invest. They propose a model with dynamic access prices which rise over time as a solution.²⁴ Such an approach may even involve access prices which differ between operators according to the date of their entry.²⁵ This lowers entry barriers both because entrants' operating costs are lower initially when cash-flows usually are most strained and because losses from an early withdrawal from the market are reduced.

The final determination of the development of the rental charge depends on the regulator's preference for network duplication. Stable access prices at cost level are preferable if preferences for network duplication are weak. If, on the other hand, regulators favour local loop duplication and at the same time are concerned about entrants' short-term cash position they might choose to alleviate entrants' difficulties in the early phase of entry by charging a low initial rental charge which would then rise to a level above cost.

Entrants will generally be less concerned about regulated prices for replicable assets than prices for non-replicable assets. As it takes time for entrants to develop their asset base (and customer base) and that they will begin by investing in those assets which are most easy to duplicate, a regulatory policy of prices which rise over time – applied successively to assets in descending order of replicability – will facilitate the gradual development by entrants of their own networks.

This solution leaves open at least a couple of questions though. At what level should prices start, at what level should they end and how long shall it take for prices to reach their maximum level? Yet another question is the risk that incumbents do not recover their costs. Incumbents are burdened with a deficit during the period when access prices are set below costs. But incumbents will not recoup this deficit if entrants, as is the intention, build their own infrastructure when access prices are increased to levels above costs. Nevertheless, dynamic access pricing might offer the solution to the otherwise complex problem of how to promote construction of replicable infrastructure while keeping entry barriers low. Dynamic access pricing is discussed further in Chapter 4.

Nordic competition case law

Telecommunication markets in the Nordic countries are characterized by the strong presence of former monopolists. The incumbent providers thus still have monopoly positions on the local loop and hold dominant positions on most other markets. This is the reason why the majority of important Nordic competition practice on telecommunication focuses on abuse of dominant position.

Despite minor national differences, the Finnish, the Swedish, the Icelandic, the Danish and, as of 1 May 2004, the Norwegian competition regulations contain direct prohibition on abuse of dominant position and there is to a large extent a similar approach to the competition analysis in Nordic practice. Even though it is not possible to totally adopt the competition analysis performed under a different regulatory regime, it is possible to learn from the analytic approach in other cases.

Market definition
Before deciding if certain behaviour constitutes an abuse of dominant position, it is crucial to determine if the company in question is in fact dominant on the relevant market.

Defining the relevant market it could be tempting to use the markets defined by the Commission in Commission's Recommendation 2003/311/EC of 11 of February 2003 on Relevant Product and Service Markets. In this context it must be underlined that in competition cases, the exact relevant market must be defined case by case.

In regulated markets such as telecommunications it is often necessary not only to define the relevant market according to supply and demand substitution. One also has to consider legal demands for public service provision etc.

It is decided through Directive 2002/19/EC, the so-called access directive (cf. table 2.8) that every telecommunication provider is obliged to give other providers access to its net for exchange of traffic. Technically, every provider has an exclusive access to its own customers. Each provider thus has a monopoly position on other providers' access to its customers.

In the above mentioned recommendation from the European Commission it is established that relevant markets in cases on access to other providers' net or customers are to be defined as separate markets for each provider. This guideline has been followed in two Finnish cases – Fixed subscriber lines from May 2001²⁶ (upheld by the Supreme Administrative Court in April 2002) and Elisa Communication Oyj's Nettitaksa Internet Service from June 2001²⁷.

Abusive pricing
According to traditional competition practice, abusive pricing can take form as either excessive pricing or predatory pricing.

In 2003, the Danish Competition Authority examined a complaint on high prices for competitors' listing in the incumbent provider's national printed directories.²⁸ The Competition Authority did not find that the incumbent provider's earnings reached levels that justified allegations of excessive pricing.

There are no decisions in Nordic competition practice on abuse of dominant position by excessive pricing.

Both Icelandic and Danish competition authorities have analysed complaints on predatory pricing.²⁹ In none of the cases, the competition authorities found sufficient proof of predatory pricing and illegal cross subsidisation according to the AKZO test.³⁰ In the Icelandic case nr. 17/1999, however, the test was never applied as the incumbent refused to give information regarding the effect of high user discounts, which was decided to be to the incumbent's detriment. The Icelandic Competition Authority's decision was upheld through all instances of appeal.

Another way of approaching possible abusive pricing is by analysing the price margin left for competitors purchasing necessary connection services by the incumbent provider.

Danish, Finnish and Swedish competition authorities have investigated complaints on abuse of dominant position by margin squeeze. 28 April 2004, the Danish Competition Council decided that the incumbent provider of fixed line telephony to business customers had abused its dominant position by obtaining an insufficient profit to cover the significant business risks associated with providing the product.³¹

The Finnish Competition Authority is currently investigating if network operators on the ADSL market are applying price squeeze to their wholesale pricing. The Finnish investigations have not yet come to a final conclusion. In the Finnish case on Nettitaksa Internet service, June 2001, investigations on possible margin squeeze were abandoned as Elisa Communication changed its prices and thereby solved a possible problem.

In 1995 and 1996, The Swedish Competition Authority investigated complaints regarding the dominant provider, Telia's, prices for interconnection over the fixed network. The relatively high interconnection prices, especially on regional and long-distance national calls made it impossible for competitors to offer interconnected calls at reasonable prices. The Swedish Competition Authority issued two interim decisions on the subject. The case was closed in 1999.³² Due to the fact that the interconnection charges had decreased significantly since the interim decisions were taken, no further action was taken by the Competition Authority. Furthermore, the Swedish Competition Authority is investigating a possible margin squeeze on the market for broadband access (ADSL).

In a Danish case from May 2002,³³ the dominant provider (TDC) used its market power to charge high rates for other operators' termination in TDC's network while TDC demanded to pay significantly lower rates for termination in other operators' net. The Competition Council stated that TDC abused its dominant position both on the supply and demand market to obtain lucrative earning on termination.

Exclusivity and loyalty
Exclusivity clauses are a known way of tying customers. As other types of behaviour mentioned in this section, it can only be prohibited for dominant providers. For non-dominant companies it is a fully legal competitive means. For companies with a market share less than 30 pct., exclusivity clauses and other vertical loyalty measures are exempted from competition assessments in the block exemption on vertical agreements.³⁴

In October 2002, the Norwegian Competition Authority prohibited³⁵ Telenor Mobil (the dominant provider) from agreeing or demanding sole supplier conditions when concluding contracts with distributors of mobile subscriptions. The Competition Authority decided that the exclusive rights prevented the distributors' ability to sell other mobile companies' products and services in competition with Telenor.

Loyalty rebates is another means for a dominant provider to attract customers – and prevent existing customers from changing to other providers. The leading Community cases on the subject are the Hoffman-La Roche case³⁶ and the Michelin case.³⁷

The Icelandic Competition Authority has dealt with two cases on loyalty rebates on telecommunication services. In a case from 1999,³⁸ the incumbent provider lowered retail prices significantly on GSM services for frequent users just shortly before a new competitors entering to the market. The Icelandic Competition Council found that the discount scheme constituted an abuse of dominant position. In its case 34/2001, the Icelandic Competition Council ruled that Iceland Telecom had abused its dominant position on the market for telecom services by signing an exclusive contract and offering non-transparent end excessive discounts only to maintain its business with the Hafnafjörður municipality.

In august 2001, the Finnish Supreme Administrative Court ruled³⁹ that ownership discount granted by the dominant telephone co-operative was tying, discriminatory and non-costaccountable and comparable to an abusive loyalty discount.

After intervention⁴⁰ by the Norwegian Competition Authority in July 2002, the dominant provider in Norway, Telenor, had to change its so-called Mobilbonus programme. The Mobilbonus programme awarded customers points for using their mobile phones. The value of the points could be exchanged for other services by Telenor Mobil. Among other things, Telenor Mobil was ordered to pay the value of bonus points automatically to customers when they left the bonus programme or closed their mobile subscription. The Competition Authority also prohibited Telenor Mobil from making the financial value of the bonus points dependent on how the bonus points were realised. Furthermore, the Competition Authority prohibited Telenor Mobil from making the accumulation of bonus points depending on who the customer was calling. Finally, it was prohibited to allow bonus members to use bonus points to buy mobile phones.

In May 2003, the Norwegian Competition Authority prohibited⁴¹ the dominant provider (Telenor) from requiring a 70 percent adherence rate for offering discounts to members of the Cooperative Housing Association (NBBL). In addition the Norwegian Competition Authority prohibited the exclusivity clause, which forbade NBBL entering into agreement with competing operators. In the long term the Authority assessed the clauses to be liable to restrict competition, by shutting out current and possible future competitors.

A number of complaints to the Swedish Competition Authority questioned the validity of Telia's combination offers of analogue NMT and digital GSM services. Telia was the only operator able to offer its customers combinations of telecommunications services from NMT and GSM.

The Competition Authority made the overall assessment that NMT and GSM from a demand viewpoint possessed limited substitutability and for this reason should be regarded as comprising two different product markets. In 1996, the Swedish Competition Authority made a decision requiring Telia (the incumbent provider) to discontinue offering joint discounts for NMT and GSM subscriptions.⁴² The Competition Authority stated that such a discount system made it more difficult for competitors on the GSM market and that it contravened the Competition Act, i.e. the offers were illegal. Telia appealed the decision of the Authority to the Stockholm City Court which in 1998 upheld the Authority's decision. After this Telia appealed the decision of the Stockholm City Court to the Market Court. The decision was later revoked due to changed market conditions.

Discrimination and other abusive behaviour According to Community/EEA competition regulation and the similar regulation in Sweden, Finland, Norway, Iceland and Denmark, discrimination is abusive if a dominant company applies dissimilar conditions between competitors for equivalent transactions. Most Nordic competition authorities have dealt with cases on discrimination. There are examples of the incumbent provider charging significantly higher prices from new competitors than from older customers,⁴³ unjust discrimination with regard to payment for termination fees,⁴⁴ not providing access to fibre optic infrastructure on equal terms,⁴⁵ and discrimination between competitors by unjustified end-user price differences between on-net and off-net calls.⁴⁶ The previously mentioned Finnish ADSL investigation is also examining possible discrimination between service operators.

The incumbent provider might also abuse its dominant position by in other ways gaining advantages from its strong market position.

In 2001, the Icelandic Competition Authority received a complaint from the new entrant Halló! on the market for fixed line voice telephony. The incumbent had sent letters to all Halló!'s customers reminding them of Iceland Telecom's services. This was further followed by calls made to Halló!'s customers. The Competition Authority informed Iceland Telecom that it was likely to abuse its dominant position on the market for fixed line telephony by using information from the fixed line division's (wholesale market) database to market services only to those customers that had chosen carrier pre-selection (down stream market). Iceland Telecom agreed to negotiations with Halló!, which settled the case.⁴⁷

In 1999, the Danish Competition Council decided⁴⁸ that the dominant provider Tele Danmark (the later TDC) did not perform anticompetitive tying or in other ways abused its dominant position by offering the product "Duet". Duet customers could call at reduced rates from their fixed line to their mobile phone, and calls to the customers' fixed line phones were directed to their mobile phones if they did not answer the call at the fixed line phone. The competitors complained that they were unable to replicate the Duet product and that Tele Danmark used its advantageous market position on fixed line as leverage on the mobile market. The Competition Council found that competitors were able to offer similar products. Furthermore, Duet was overall profitable despite the reduced rates for calls from the individual customer's fixed line to his or her mobile phone.

In 1996, the Swedish Competition Authority decided that the dominant provider (Telia) would abuse its dominant position on fixed subscriber lines by introducing asymmetric prices to its own customers. Telia had planned to gradually introduce a new service called "transferred call" which would make it up to SEK 0.50 more expensive per minute for Telia's own customers to call customers or dial-up numbers in the network of other operators. The Competition Authority stated that the pricing which Telia wanted to introduce would have highly negative effects on competition, as customers would have very limited incentives to subscribe to fixed networks other than Telia's. The Competition Authority made an interim decision on May 1996 and a final decision on December 1996.⁴⁹ Telia appealed the decision to the Stockholm City Court. On March 2000, the Stockholm City Court upheld the Competition Authority's decision. Telia appealed the decision to the Market Court. The decision was later revoked due to the fact that Telia no longer planned to introduce such a pricing scheme.

Footnotes

15 The European Commission's directorate general for competition's (DG Comp) application of common European competition regulation is a notable exception. DG Comp has issued a number of recommendations and guidelines describing on a pre-emptive basis how the European Commission intends to interpret community competition law. But these recommendations are not binding and an infringement of recommendations and guidelines therefore has to be followed by a formal decision. Possible entrants thus cannot rely completely on being protected against abusive behaviour as described in the guidelines.

16 In Norway, the government chose to split up the government-owned companies Statnett SF (Norwegian powergrid company) and Statkraft (power generation company). Several other companies have chosen legal separation of generation and transmission companies.

17 The New regulatory framework is described further in the section "Principles for and extent of sector specific telecommunications regulation"

18 The interconnection directive was later amended through the directive 98/61/EC concerning numbering.
According to the amended directive, number portability should be possible. In addition, it should be possible for a subscriber to automatically have different types of calls connected by predetermined telecommunications operators (pre-selection). It should also be possible for a subscriber to bypass the pre-selected operator by dialling a prefix before the telephone number (call-by-call carrier-selection).

19 This has been expressed by RBB Economics (2003)

20 Press release at http://europa.eu.int/rapid/pressReleasesAction.do?reference=IP/04/994&format=HTML&aged= 0&language=EN&guiLanguage=en

21 This important dilemma is also addressed in "ERG common position on the approach to appropriate remedies in the new regulatory framework" section 5.2.2.3. ERG emphasizes that NRAs have to be careful to design the relative prices of the different options in relation to one another and in relation to the retail prices correctly. Too low a price on one level may e.g. inhibit investment on another level, where replication may be desirable.

22 See the Economist (2003)

23 Peitz also underlines that regulatory policy should clearly distinguish between two different market phases, namely infancy and maturity. A policy that is optimal in a mature market may not be optimal in an infant market and vice versa.

24 A policy of rising access prices over time has, according to the authors, the further advantage for entrants that the incumbent's incentive for price squeezes is reduced over time.

25 Cave and Vogelsang emphasize that such discrimination may be impracticable or unlawful in some countries.

26 Finnish Supreme Administrative Court's decision 150/690/2000, 18.5.2001: Elisa Corporation's fixed subscriber lines.

27 Finnish Competition Authority's decision 570/61/2000, 26.6.2001: Elisa's Nettitaksa Internet Service.

28 Danish Competition Council's meeting on 18 June 2003: Telia A/S complaining over TDC Forlag A/S.

29 Icelandic Competition Authority's decision nr. 17/1999: Iceland Telecom's GSM prices, Icelandic Competition Authority's decision nr. 41/2003: Iceland Telecom's closed user groups, Danish Competition Council's decision of 18 January 2003: Investigation on TDC's ADSL prices.

30 AKZO Chemie BV v. the Commission, case C-62/86.

31 Danish Competition Council's decision of 28 April 2004: Song Networks.

32 Swedish Competition Authority's decision on 26 May 1999 (case No 107/95).

33 Danish Competition Council's decision of 29 May 2002: Complain against TDC (filed by Consorte).

34 Commission Regulation (EC) No 1400/2002 of 31 July 2002 on the application of Article 81(3) of the Treaty to categories of vertical agreements and concerted practices in the motor vehicle sector, OJ 2002 L 203/30.

35 Norwegian Competition Authority's decision of 14 October 2002 no 85.

36 Hoffman-La Roche v. the Commission, case 85/76.

37 Michelin v. the Commission, case C-322/81.

38 Icelandic Competition Council's decision nr. 17/1999: Iceland Telecom's GSM prices.

39 Finnish Supreme Administrative Court's decision 240/2/2001, 15.8.2001: Päijät-Hämeen Puhelinosuuskunta's (telecom co-op.) ownership discounts.

40 Norwegian Competition Authority's decision of 10 July 2002 no 55

41 Norwegian Competition Authority's decision of 3 October 2003 no 58.

42 Swedish Competition Authority's decision of 26 June 1996 (case No 63/94).

43 Elisa Corporation's fixed subscriber lines, Finnish Competition Council's (FCC) decision 18.5.2001
(150/690/1999), no appeal to Supreme Administrative Court; Salon Seudun Puhelin Corporation's fixed subscriber lines, Finnish Competition Council's (FCC) decision 18.5.2001 (14/690/2000), Supreme Administrative Court's decision 22.4.2002 (1817/2/01); Turun Puhelin Corporation's fixed subscriber lines, Finnish Competition Council's (FCC) decision 18.5.2001 (15/690/2000), Supreme Administrative Court's decision 22.4.2002 (1842/2/01)

44 Finnish Competition Authority's decision of June 2001: Elisa Communication Oyj's Nettitaksa Internet Service.

45 Icelandic Competition Authority's decision nr. 21/1998: Iceland Telecom's fiberoptic infrastructure.

46 Following the Norwegian competition notice of intervention 18 June 2003, the incumbent reduced the price differentials considerably. In the same period the incumbent lost market shares. As a result, on 17 December 2003, the Authority considered an intervention to be redundant.

47 Icelandic Competition Authority's decision nr. 23/2002: Iceland Telecom's contact with Halló!'s customers.

48 Danish Competition Council's decision of 26 May 1999: Tele Danmark's Duet service.

49 Swedish Competition Authority's decision of 12 December 1996, case No. 587/96.

Version 1.0 October 2004 • © Danish Competition Authority.
Published by the Danish Competition Authority, www.ks.dk
Publication produced according to the standard for electronic publication set by the Government