Tariff regulation
the way forward
De Meern, November 2001
Logica Consulting B.V.
P.O.Box 172
3454 ZK De Meern The Netherlands
ECF0174/GMG/rul/N7573
Table of Contents
1 Summary 4
2 Objective of study 11
3 Theoretical framework 13
3.1 Introduction 13
3.2 Market Development 13
3.3 The role of Tariff Regulation 15
3.4 Effective Competition 16
3.5 Efficient Competition 23
3.6 Cost Allocation Systems 25
3.7 Accounting Separation 26
3.8 Price Caps 28
3.9 Requirements for tariff regulation 28
4 Tariff regulation in a European context 30
4.1 Tariff regulation by the European Commission 30
4.2 Other issues 36
4.3 Summary 37
5 Retail Tariff Regulation 38
5.1 Voice Telephony services 38
5.2 Leased line services 43
6 Wholesale Tariff Regulation 48
6.1 Fixed terminating and originating access 48 6.2 Fixed to Mobile terminating and originating access 60
6.3 Unbundled Local loop 68
6.4 Wholesale tariffs for leased lines 73
7 Other issues concerning tariff regulation 76
7.1 Flexible possibilities for intervention 76
7.2 Dynamic market definition 77
7.3 Price Squeeze issues 80
8 Conclusions 88
8.1 Retail tariff regulation in the Netherlands 88
8.2 Wholesale tariff regulation in the Netherlands 90
8.3 Other issues 95
Endnotes 98
Annex 1: Regulatory situation around Europe 100
A1.1 Tariff regulation in the United Kingdom 100
A1.2 Tariff regulation in Germany 104
A1.3 Tariff regulation in France 110
A1.4 Tariff regulation in Sweden 113
Annex 2: PSTN/Interconnection Benchmark definition 117 Annex 3: Additional analysis: Predictability of wholesale tariffs 118
Annex 4: Glossary 120
Endnotes to the annexes 123
1 Summary
European telecommunications regulation beginning with the Voice Telephony Directive in 1990 has sought to introduce competition into the
telecommunications market across the European Union (“EU”).
This has been done principally through the opening of the existing incumbents’
networks to new entrant operators who do not possess national networks, as do the incumbents. Much regulation has focused on requiring incumbents to provide access to their networks at cost based prices. In theory, this emulates the costs, which new entrants would bear if they were using a network, which they owned themselves.
However there are a number of difficulties and flaws with this reasoning stemming from the costing method used to determine incumbent network costs and from additional and unavoidable costs that new entrants must bear over and above those incurred by an incumbent offering the same services.
Before we go further, we should acknowledge that no regulatory system is perfect or will be suited to all situations. Most regulators acknowledge this and have adapted and developed regulatory policy both at the supranational (EU) level and at the national level. Indeed cost allocation methods and the type of costs that should be used to set cost based prices in different circumstances has been widely and fiercely debated across Europe. However, the aspect of
additional costs borne by new entrant operators because they incur both on-net
costs and off-net costs in disproportion to the on- and off-net costs incurred
by incumbents is an area that most regulators have failed to acknowledge or
address sufficiently. As retail costing based on the costs of theoretically
efficient operators becomes a widespread costing methodology, this is putting
increasing pressure on new entrants who, due to the additional off-net cost
burden they have to support, are finding themselves under increasing margin
pressure (price squeeze) and with ever decreasing hope of matching the
theoretical efficiency levels of the regulators cost models. This in turn is a
deterrent to investment by the new entrants. Unless the prices they can demand
for their services are high enough for them to cover their real costs and earn a
reasonable return on their investments they will ultimately cease to trade as
their investors seek other more lucrative investment opportunities elsewhere.
Regulators therefore have to be acutely aware that they must set prices at levels where the average operator can earn a reasonable return on its investment. This return on investment must give an incentive for less efficient operators to strive for greater efficiency, and must recognise that the most efficient operators will earn a higher rate of return on their investments than other operators – a just reward for their above average efficiency. Regulators should not be trying to squeeze every drop of blood from the market, but encouraging growth and ultimately self-sustaining competition.
To achieve self-sustaining competition, market players need to either be able to differentiate their offerings and charge a premium, or they need to be able to achieve a low cost base and make profits through greater margins than their competitors. We develop a theoretical framework, which looks at the requirements to achieve effective and efficient competition which will lead ultimately to self-sustaining competition.
The framework identifies a number of discriminatory cost advantages that incumbents enjoy over new entrant operators and suggests how these might be eliminated through regulatory measures. The key advantages enjoyed are:
• Historical assets (the cost of which has already been recovered thus leading to very low costs (sometimes zero) for certain assets used to provide services;
• On-net advantages; the new entrant has to route substantial traffic over an incumbent’s network as well as developing its own network. The
incumbent can route almost all of its traffic over its own network thus avoiding some of the interconnection charges new entrants must incur. As new entrants gain market share, this cost disadvantage will be spread more evenly between new entrants and incumbents, as the latter will find
themselves having to interconnect with new entrants to route traffic;
• Economies of scale, the incumbent will usually enjoy economies of scale and thus efficiency advantages over smaller new entrants.
To overcome these difficulties, it is suggested that wholesale prices of
incumbents be carefully monitored to ensure no “ghost costs” are included in wholesale tariffs and on the retail side, it is suggested that price floors be used to avoid unsustainable price squeeze for new entrants. Ghost costs are costs that are ‘nonexistent’ to the incumbent operator but are charged nonetheless in (wholesale) tariffs to new entrants. Ghost costs give the incumbent a
discriminatory cost advantage that may be used to undercut prices of new
entrants, thus severely damaging the position of new entrants, and potentially destroying the development of effective competition.
The report then examines the real regulation in the Netherlands of retail and wholesale services, and compares this to the application of EU legislation by a number of other regulators in Europe. We will see that a number of failures have arisen in the market due to regulation which is not always all
encompassing.
Retail tariff ceilings are used in EU regulation in order to protect the interests of consumers. Setting a ceiling to retail tariffs will also force an incumbent
operator to work more efficiently. When competition is effective however, the retail price control and efficiency stimuli will come from competitive market forces. Under effective competition, retail tariff ceilings are either irrelevant, because the market price is below the ceiling, or may even hamper effective competition if they cause a price squeeze. In markets where effective competition is occurring, or possibly even if the conditions for effective competition are present, retail tariff ceilings should be abolished.
In retail services, we see a number of instances where the margin between wholesale and retail rates is very narrow suggesting more rigorous allocation of costs in determining wholesale services. This may be achieved for instance by using a LRIC method or by using historic costs rather than current costs where this is appropriate, such as in the local access network. The use of price floors based on costs of an efficient new entrant operator are essential to ensure that price squeezing is minimised in the future.
On the wholesale side, a number of failures have arisen in the Netherlands due to:
• The high level interconnection tariffs – especially for call origination;
• The lack of predictability of tariffs which seem to be due to changing cost allocation policies and a non transparent costing system – Embedded Direct Costs;
• Lack of transparency, due to the use of non transparent cost allocation systems, incomplete implementation of accounting separation and OPTA’s decision not to publish accounting separation results;
• Lack of implementation of the non-discrimination policy. KPN Carrier
Service costs are not (yet) fully shared between all of KPN’s business
units leading to higher costs for certain wholesale services;
• Confusing stimuli for investing in infrastructure for efficient
interconnection with KPN, which may have caused operators to make incorrect and ineffective investment decisions.
Benchmarking with other countries reveals that attention needs to be paid not only to the cost methodology used, but also to careful identification of the relevant cost drivers. The evidence presented in this report suggests that these have a greater bearing on regulated prices than the cost allocation method chosen.
Concerning fixed to mobile tariffs, mobile operators have long enjoyed a lack of regulation. However this is now allowing them to cross subsidise services giving them an unfair advantage over fixed line new entrant operators for basic voice services. The high termination tariffs they charge for fixed-to-mobile calls allow them to subsidise call origination tariffs and undercut fixed telephony operators that do not enjoy these high termination costs. Regulators across Europe are beginning to look with concern at the termination rates charged.
Where mobile operators have a large market share on the national market for interconnection, they can be given SMP designation and their termination tariffs will be subject to regulation. In the Netherlands, we are concerned that the national interpretation of the EU legislation means that no mobile operator will be designated with SMP with cost orientation obligations. Although the use of the definition of the relevant market as ‘national market for interconnection’
is questionable in the context of mobile termination tariffs, we believe that at least KPN Mobile, but preferably also Libertel and other mobile operators, must be designated as SMP with resultant obligations for cost orientation. The general obligation for “reasonable tariffs” that is included in the Dutch
regulatory framework gives OPTA an additional possibility to force the other three mobile operators to lower their tariffs as well.
Concerning access to unbundled local loop services, OPTA has implemented a system that currently produces costs reflecting the true costs incurred by KPN for supplying local loop services (i.e. based on historic costs). However, the system will move towards a system based on current costs which will overstate the cost of provision and lead to unjustified profits for KPN, and price
squeezing for new entrants who are unlikely to have sufficient incentive to
develop competing local access infrastructure to provide existing services as
returns will be too low. New entrants may rather develop infrastructure along
with the market development for higher value services such as Video-on-
demand to the home and multimedia communications and entertainment to the home.
Leased lines is another area where current regulation in the Netherlands is currently failing to encourage efficient and effective competition. Presently there is no wholesale leased line offer available in the Netherlands. The key factor here is the lack of a sufficiently cost oriented interconnect leased lines offer, which would allow new entrants to compete effectively with KPN in all areas of the Netherlands, not just in those locations where they have their own infrastructure available. A LRIC based wholesale offer, combined with an adequate retail price floor will therefore encourage a healthy, competitive, retail market for leased lines. As soon as this will be achieved, new entrants will be able to compete effectively with KPN and the relevance of and emphasis on retail price ceiling regulation may decrease.
Finally, the telecoms market has developed and continues to develop very rapidly. The Netherlands uses very rigid definitions to determine what services can be regulated and which are absolved from regulation. This is now creating some embarrassing situations for OPTA that is unable to intervene in areas such as data and Internet related services where KPN clearly enjoys unfair advantages. It is suggested that a more flexible system be adopted like that of the UK where the level of competition in a market determines the extent to which the regulator can determine prices. This allows regulation to develop along with the telecoms market and should ultimately allow both the support of sustainable competition whilst it is developing and then the withdrawal of regulation as self-sustaining competition is achieved in more segments of the telecoms market.
Summarising, we recommend the following measures in regulating retail tariffs:
• Ensure an adequate retail tariff floor for all retail services on which the incumbent may have a discriminatory cost advantage. The tariff floor can be implemented by a Price Squeeze Test, provided that the test uses an adequately high mark up and that an adequate, cost oriented,
wholesale offer is available;
• Ensure that stimulating the long-term decrease of retail tariffs is
coordinated with the long-term development of wholesale tariffs,
thereby predictably avoiding price squeeze issues in the long run.
For wholesale tariff regulation, the following measures are recommended:
• For all wholesale services, tariff regulation must be based on a fully transparent Bottom Up Long Run Incremental Cost (BU-LRIC) cost allocation system. This will enable maximum transparency and predictability of tariffs as well as a sufficiently low level of wholesale tariffs. This will allow new entrants to compete as effectively as possible with the incumbent as well as ensure that the additional costs of
introduction of competing networks is compensated by overall increased efficiency;
• In order to ensure that wholesale tariffs are predictable and price squeezes are avoided in the long run, a Price Cap system for wholesale tariffs must be implemented that is (slightly) more aggressive than the price caps for retail services;
• The non-discrimination principle must be implemented in full in order to allow new entrants to compete effectively with the incumbent;
• Ensuring cost oriented fixed to mobile tariffs by at least designating KPN mobile with Significant Market Power on the combined market for fixed and mobile telephony. This is preferably followed by
designation of other large mobile operators or ensuring cost orientation of tariffs based on the ‘reasonable tariffs’ criterion;
• Cost oriented tariffs for the local loop must remain based on historic costs, including a small current costs allowance for maintenance and extension. Current regulation foresees full use of current costs within a few years. This however will create ‘ghost costs’ which effectively gives KPN a discriminatory cost advantage. European practice also shows that the use of current cost accounting for the local loop causes a severe price squeeze with retail telephony subscriptions, making it impossible for new entrants to compete in this area;
• Ensuring that KPN makes available a cost oriented (LRIC based) wholesale offer for leased lines;
• Implementation of tariff floors for reciprocal (terminating) wholesale tariffs in order to prevent that new entrants are forced to offer termination services below their cost level.
Finally we recommend the following general measures:
• In order to allow competition to become effective, OPTA’s Price
Squeeze Test must be improved in several areas. This includes a
substantial increase of the ‘retail mark up’ to reflect a/o a new entrants
network costs, and extension of the Price Squeeze Test to force tariff
floors for all services on which KPN may have a discriminatory cost advantage, including all services using the local loop and leased lines;
• OPTA must have the authority to flexibly intervene in markets and apply tariff regulation wherever KPN may have a discriminatory cost advantage;
• OPTA must have the authority to define markets dynamically, provided these definitions are (ex ante) harmonised with European market definitions;
• Accounting separation must be implemented minimally according to European recommendations. It must however also include
reconciliation data between cost allocation system and KPN’s statutory accounts in order to detect possible damaging charging of ‘ghost costs’.
The accounting separation information must be made publicly available.
2 Objective of study
The members of ACT -BT Ignite, Energis, Priority Telecom, Versatel and WorldCom- have concluded that, despite three to four years of full
liberalisation of the telecommunications market, there is still an insufficient basis for sustainable competition in the Dutch market.
The members of ACT asked Logica Consulting to provide them with an in- depth, factual and up to date view on tariff regulation in Europe and more specific in the Dutch market.
Purpose of the study is to provide the members of ACT with
recommendations for the way forward for tariff regulation in the Netherlands and where applicable in Europe to ensure a basis for sustainable competition.
Please note that the opinion of individual ACT members may differ from the views described in this report.
In this report a business economic approach is used for the way forward of tariff regulation that, instead of a general economic approach as taken by many authors in this field, deals specifically with the business economic conditions for new entrants to compete effectively with the incumbent operator. The business economic conditions are the key determinants for new entrants to be able to enter and operate in the telecommunications market on a sustainable basis.
This should be kept in mind when reading and discussing the report.
Chapter 3 of this report contains a theoretical framework, outlining the main objectives of liberalisation and how they should be achieved effectively. This framework is used later in this report to evaluate the effects of current regulation and its effects.
Chapter 4 contains an overview of the current European and Dutch regulatory
environment and the effect of the regulations on (sustainable) competition in
the Telecommunications market.
Chapters 5, 6, and 7 evaluate to what extent regulation and policies of the Dutch regulator OPTA have been successful in creating sustainable
competition with regard to compensating specific competitive advantages of incumbents as well as the effects of specific policies. Where applicable references will be made to practice in other European countries.
Chapter 7 contains conclusions and recommendations for the way forward.
3 Theoretical framework
3.1 Introduction
This chapter provides a generic framework for identifying the prerequisites of sustainable competition and its consequences for tariff regulation.
In the EU liberalisation model for the telecommunications industry the overall objective of liberalisation is “to benefit the EU citizen with freedom of choice, lower prices and innovation in telecommunications services” by establishing effective and efficient competition. Tariff regulation is one of the key instruments to realising efficient and effective competition in the telecommunications market.
Effective competition implies the creation of fair and attractive conditions under which new entrants can build up a sustainable competitive business. The development of effective competition is the prime method used by the EU to ensure that the EU citizen can always have a choice between competing operators for telecommunications services.
Efficient competition means that new entrants as well as incumbent
operators will work as efficiently as possible to minimise their operational costs and thus create a sustainable basis for low end user prices.
This chapter will first discuss the theoretical phases of the development of the liberalising telecommunications market. Then general requirements for
effective and efficient competition will be derived.
3.2 Market Development
The objective of liberalisation is to develop a monopoly market situation,
where new entrants need to be supported by industry specific regulation, to a
situation where competition will be ‘self-sustaining’. The development of the
market therefore can be divided into two separate phases: the ‘liberalisation
phase’ and the ‘self-sustainable competition’ phase. Obviously, in order to let
the competition develop, market conditions must allow competition to be
sustainable throughout both phases. In this report, we will focus on attaining sustainable competition during the ‘liberalisation phase’.
The ‘liberalisation phase’ is characterised by the existence of strong market players, mostly former incumbents that enjoy competitive advantages over potential new entrants. During the liberalisation phase, strong players must be restricted in using their competitive advantages to prevent or hinder new entrants from entering the market and/or developing their businesses. This can be achieved by creating a temporary, industry specific regulatory environment ensuring that new entrants get a fair chance to grow their business.
Figure 1, two phases of market development
The liberalisation phase can be said to end when we no longer require specific legislation to regulate the industry, but can rely on general competition law.
Identification of strong players is achieved by designating such parties as
‘parties with Significant Market Power’ (SMP) and applying additional obligations to these parties in order to:
• Level their specific advantages with new market entrants;
• Restrict their power to influence market conditions or charge excessive
tariffs for access to their networks.
Generally SMP operators have a share of 25% or more of a given market. This is the level at which it is thought possible to control prices independently of the rest of the market.
3.3 The role of Tariff Regulation
Tariff regulation can take place at two levels: wholesale and retail. To achieve long-term sustainable competition, regulators have to encourage competition at the infrastructure level. This is done by ensuring that a new entrant developing its own infrastructure can earn a reasonable return on the capital it invests in infrastructure. If regulated wholesale prices are set too low, this could
discourage infrastructure development. In order to allow new entrants to build out their infrastructure however, they first must be able to compete effectively with the incumbent operator. The development of infrastructure competition must therefore be considered a subordinate objective to the development of competition itself. This means that stimulating infrastructure roll out can only be justified as long as new entrants can compete effectively with the incumbent at the same time.
Most tariff regulation is applied to operators (usually incumbents) that operate as vertically integrated firms on the telecommunications market. These
operators:
• Provide retail services to consumers and businesses; and
• Provide wholesale services to new entrants.
Figure 2: position of new entrants relative to vertically integrated incumbent
During liberalisation, new entrants with little, if any, infrastructure, buy wholesale services from incumbents and sell retail services to end-users. Their margins are dictated by the level of the incumbent’s wholesale and retail prices, between which they have to operate.
Tariff regulation regulates both retail and wholesale tariffs and therefore impacts business of new entrants on the wholesale (cost) side as well as on the retail (revenues) side.
3.4 Effective Competition
In this report we will define effective competition as the situation where new entrants in a market can compete with incumbent operators on a sustainable basis, provided that they make the right decisions. This means that advantages of incumbents should be levelled or eliminated in such way that all players, including the incumbent, have a fair chance of success. Where advantages cannot be levelled, regulation must ensure that the incumbent cannot abuse these advantages. Once this situation is achieved we can talk of a ‘level playing field’. This situation has to be sustainable over the long term so that operators find the market attractive to enter and operate in. For effective competition to exist then, we require:
• The existence of a ‘level playing field’;
• Sustainability over a long period of time making the market attractive.
3.4.1 The Level Playing Field
During liberalisation, the historic competitive advantages of the incumbent should, be levelled so that they are competing on a level footing with the new entrants. This is analogous with the handicap system used in golf. This situation should last until the new entrants themselves have created sustainable
competitive advantages that match those of the incumbent in such way that competition is possible without the requirement to “handicap” any of the players.
In analysing what requirements underlie a ‘level playing field’ where all
operators have equal opportunities in creating sustainable competitive advantage,
we use the framework developed by Michael Porter
1. In achieving competitive
advantage, Porter distinguishes only two basic sustainable strategies:
• Cost leadership, where a competitive operator strives to minimise costs to achieve the lowest possible cost level, competing normally at level prices in the end customer market;
• Differentiation, where a competitive operator strives to achieve a differentiation advantage (e.g. by innovative services) that can be leveraged via premium end customer prices.
3.4.2 Equal cost minimisation opportunities
Following the above, in a ‘level playing field’, all operators in an industry must have equal opportunities to minimise their costs. Firms that have sources of cost advantages to which other firms do not have access to can harm competitors by lowering their prices below a (cost) level that cannot be achieved by their competitors.
Examining the costs of a telecommunications operator, we see that they can be divided into three component parts:
• Retail Costs: Costs for acquiring and maintaining the relationship with customers;
• Own Network Costs: Costs of the own network and organisation, including costs required for obtaining wholesale services (e.g. own network for interconnection);
• Wholesale costs: Costs charged by other operators for the use of their network and related facilities (i.e. interconnection and special access tariffs).
Figure 3: Cost structure of new entrants
If conditions exist to create equal cost minimisation opportunities then all operators in the industry can in principle offer a service against the same costs, provided that they use state of the art technology and organisation. This requires the existence of:
Wholesale Costs Own Network
Costs Retail Costs total costs
of new entrant
Margin on services Retail tariffs
Wholesale tariffs
• Non-discriminatory charging of wholesale costs;
• Adequate levelling of cost advantages of the incumbent;
• Adequate unbundling of services;
• An adequate degree of reciprocity of wholesale tariffs.
Non-discriminatory charging of wholesale costs
Operators that provide wholesale services to other operators incur ‘wholesale costs’ in doing so. The equal cost minimisation opportunities criterion dictates that these wholesale costs are charged equally to external operators obtaining wholesale services as well as to the internal retail service department of the providing operator. This means that the providing operator is not allowed to discriminate between charging costs to itself and to other operators when providing wholesale services. This is called the non-discrimination criterion.
An important side effect of the non-discrimination criterion is that in fact a vertically integrated providing operator must not be allowed to charge ‘ghost costs’, i.e. costs that it does not make in reality. This would violate the equal cost minimisation opportunities criterion because the costs for external
operators would be higher than the costs that it incurs for obtaining wholesale services by itself. Artificial costs may arise if an operator sets arbitrary tariffs or maintains different cost accounting principles internally than it does for
calculating tariffs to external parties. Tariffs of wholesale services that an incumbent provides to new entrants must therefore be cost oriented. This means that tariffs must never exceed the actual cost that an incumbent incurs to provide them.
2Tariffs may however be set to a lower level that reflects the cost level of an efficient operator. Such a case prevents the incumbent from charging other operators for its inefficiency.
Levelling of cost advantages of the incumbent
Due to its long standing existence in the telecommunications market and the fact that it has been able to build up its business in a monopoly environment, the incumbent has a number of specific cost advantages that allow it to provide services against (substantially) lower costs than new entrants can. These
advantages in general consist of:
• Economies of scale;
• On-net traffic advantages; and
• Possession of unique historic assets.
Economies of scale
Cost advantages due to economies of scale of an incumbent arise due to the fact that during liberalisation the incumbent still has a substantial market share and nation wide geographic coverage and can optimise its network and
operations accordingly. This will allow the incumbent to provide wholesale services as well as retail services against lower costs than new entrants can. This advantage will gradually disappear as market shares of new entrants grow and the market shares of the incumbent shrinks so that they become of similar size.
On-net traffic advantage
During liberalisation, new entrants roll out their own telecommunications networks, thereby in fact duplicating the network of the incumbent. This duplication causes a substantial additional ‘own network’ cost factor specifically for new entrants because, next to the cost of their own network, they also have to pay wholesale costs to the incumbent. The incumbent does not have this disadvantage because, at least during the early phases of liberalisation, it can route the vast majority of traffic via its own network.
Figure 4: Cost structure of new entrant versus incumbent
The on-net traffic advantage will gradually disappear as new entrants build out their own networks, grow in size and take market share from the incumbent.
As a consequence the incumbent will eventually pay wholesale costs to new entrants for a large share of its services. During the early phases of
liberalisation however, the on-net traffic advantage represents a substantial cost
advantage for the incumbent.
Unique historic assets
Incumbents usually have a large asset base that has been established over many years during their monopoly. Many of these assets will have been established at considerably lower cost than would be possible today, and many assets may have been financially recovered to a large extent during the monopoly time.
Such assets may now have a low historic value but may still be viable to use for many years or even decades more. Examples of such assets are the local loop, the copper access network used to connect end users to the incumbent’s local exchanges and an incumbent’s buildings. The low economic value of such assets can be arbitrarily used by the incumbent as a cost advantage over new entrants who are making large investments to replicate local loop networks.
Levelling of cost advantages
The existence of the above-mentioned cost advantages for the incumbent is contrary to the principle of equal cost minimisation opportunities and therefore dissuades new entrants. Due to these cost advantages an incumbent is able to provide services at a lower price than new entrants. Tariff regulation is an essential tool to level these cost advantages so that new entrants can compete effectively with the incumbent. In order to do this, tariff regulation must include:
• Low, cost oriented, wholesale tariffs in order to level cost advantages of the incumbent to the maximum extent possible. This may imply lowering wholesale tariffs to the cost level equivalent to a maximally efficient operator;
• Tariff floors for retail tariffs that prevent incumbents from lowering tariffs under the price and cost level that can be attained by new entrants;
Tariff floors can be implemented either as an absolute tariff floor, or, if a wholesale offer exists for the retail service considered, a minimum margin between wholesale and retail tariffs. The latter is also known as the ‘Price Squeeze Test’. If a tariff does not satisfy the Price Squeeze Test, either the retail tariff must be increased (i.e. implementing the price floor) or the wholesale tariff must be decreased.
Unbundling of wholesale tariffs
In obtaining wholesale services from the incumbent, new entrants must be able
to make their own “make or buy” decisions to minimise costs. This requires
that wholesale services and tariffs are unbundled into separate services so that competing operators are not required to purchase services they do not require.
Reciprocity of wholesale tariffs
In minimising their costs, operators may have equal opportunities in
minimising costs that are within their control, such as own network costs and retail costs. Wholesale costs are outside of their direct control and can only be influenced by negotiating tariffs with the party that they obtain wholesale services from. Such a party however is in a position to charge excessive wholesale tariffs because it fully controls a ‘bottleneck facility’: the access to customers. From the equal cost minimisation opportunities requirement follows that in order to let parties compete effectively with one another on the retail market, they must incur the same wholesale costs for access to (each) other’s networks. This is called the reciprocity principle. The reciprocity principle is primarily applicable to competing services that are provided in technologically and organisationally the same way. However, also in the case of technologically different services such as fixed and mobile telephony, the notion of reciprocity is relevant where, while competing on the retail side, terminating tariffs for mobile to fixed and fixed to mobile differ substantially.
Regulators may choose to implement reciprocity either (1) by trying to regulate wholesale tariffs of incumbent as well as new entrants directly or (2) by
allowing end user tariff differentiation, i.e. allowing incumbents and new entrants to discount excess or lower wholesale charges made by an interconnected operator into the retail tariffs. Both will result in equal wholesale costs for operators.
Reciprocity is an essential principle that may be used by regulators to prevent owners of bottleneck facilities from charging excessive tariffs for access to these bottleneck facilities. In a liberalising market however implementation of reciprocity must take into account the share of on-net traffic of the incumbent relative to the off-net traffic. Reciprocity only applies to off-net traffic. During the early phases of liberalisation, the incumbent has virtually no off-net traffic while new entrants have close to 100% off-net traffic. Enforcing reciprocity of wholesale tariffs would therefore amplify the incumbent’s on-net traffic
advantage instead of levelling it
3.
A further consequence of tariff reciprocity may be that tariffs are set below the
(reasonably achievable efficient) cost level of one of the parties involved. Such
a situation is particularly undesirable for new entrants and must be avoided by
setting a tariff floor equal to at least the wholesale cost level of a reasonably efficient new entrant.
Any form of reciprocity must therefore take into account cost advantages and disadvantages of each party and does not necessarily imply that tariffs between both parties are equal. Concluding we can state that, based on the reciprocity principle:
• Tariffs for wholesale services that are used to provide similar or competing services must not be set excessively high;
• Tariffs for wholesale services provided by new entrants, that are reciprocal must not be set under the cost level of a reasonably efficient operator.
3.4.3 Equal Differentiation Opportunities
Next to cost leadership, firms in an industry can create sustainable competitive advantage by creating a unique product offering or positioning. In a level playing field, operators should have equal opportunities in differentiating themselves from other operators offering innovative services at innovative prices.
Whereas, during liberalisation, the incumbent handles most traffic and
customers on its own network, new entrants may largely depend on wholesale services of the incumbent for providing new and innovative services. In order to give new entrants the maximum possible freedom in developing new services, the incumbent’s wholesale offer should allow them to make the right make or buy decisions. In order to do this, the incumbent’s wholesale offer must be sufficiently unbundled, i.e. split up in separate meaningful units that allow new entrants to purchase only the wholesale services they require and thus make the right make or buy decisions. The unbundling requirement also holds for retail services.
Service innovation leads by definition to new services that are not (yet) subject to tariff regulation. In order to protect new entrants from an incumbent (ab)using cost advantages (see previous section) to unfairly compete with new entrants, regulators must have the power to apply tariff regulation if, when and where necessary. This requires a regulator to be able to:
• Flexibly intervene in markets by applying tariff regulation;
• Dynamically define the markets to which tariff regulation must be made
applicable.
3.4.4 Market attractiveness
Next to the ‘equal opportunities (level playing field)’ criteria, market attractiveness plays a crucial role in providing a basis for sustainable
competition. Competition will not occur as long as potential new entrants find it unattractive to enter or operate in the market. Concerning tariff regulation, market attractiveness is predominantly determined by the predictability of tariffs and transparency of tariff decisions.
Attractiveness of a market is also heavily influenced by a regulator that can ex ante apply regulation when and where necessary. The effectiveness of
regulation in creating ex ante conditions under which effective competition can occur is determined by the requirements for effective competition as outlined in this section.
3.5 Efficient Competition
The objective of the EU of enabling ‘efficient competition’ is to realise lower retail prices in advance of, or in absence of price decreases due to competition.
The EU is concerned about protecting consumers from incumbents charging excessive prices. Efficient competition is achieved in the following ways:
• Limiting the incumbent’s retail and wholesale tariffs to such a level that they stimulate it to become more efficient and reduce costs.
• Allowing efficient new entrants to effectively compete with the incumbent on all services where the use of new, state of the art technology and organisation provides a higher level of efficiency than the incumbent has;
• Stimulating efficient interconnection of networks between new entrants and the incumbent.
3.5.1 Tariff ceilings
Limiting the incumbent’s retail and wholesale tariffs to a ceiling level will stimulate the incumbent to improve its efficiency, because this is the only method it can use to improve its margin on services. The maximum ceiling level must cover the incumbent’s actual costs plus a reasonable return on investment that allow it to sustainably continue providing a certain service.
Such an above cost ceiling level would provide incumbents with a mild
stimulus to improve efficiency. Theoretically, a more aggressive stimulus is
provided if tariffs are set based on the cost level of an efficient operator (plus a
reasonable return on investment), i.e. below the actual cost level of the
incumbent. As long as the incumbent does not meet this cost level, the incumbent will incur losses on the services that it can only eliminate by investing in network and operational efficiency. This method needs to be applied with care for although the incumbent may, over time, be able to match such demanding efficiency requirements, it may still be difficult for new
entrants to do the same due to their lesser economies of scale and because they initially have to replicate the network of the incumbent. It may therefore be wise to set efficiency measures based on a less than perfectly efficient operator.
3.5.2 Creating conditions for effective competition
An incumbent is also stimulated to work more efficiently where it faces
competition from efficient new entrants. This again stresses the importance of the existence of a complete, adequate and unbundled wholesale offer, and the existence of low barriers of entry, thereby exposing as many areas as possible for
competition by new entrants. As the core objective of liberalisation is to benefit the EU citizen by ensuring low prices through enabling effective competition, and only to a lesser extent by imposing (retail) price ceilings, the focus of regulatory activities should always lie on creating adequate conditions for effective competition. In that respect, the existence of retail tariff ceilings is indicative of the fact that competition is not yet effective and would suggest that the underlying incumbent’s wholesale offer is not yet fully adequate or that barriers of entry are too high.
3.5.3 Relevance of retail tariff ceilings
As described above, regulators may choose to implement retail tariff ceilings for the incumbent operator to protect the interests of consumers against potentially excessive prices of parties with market power. They must however only do this in absence of effective competition. If competition in a market is effective, market mechanisms and not artificially set tariff ceilings must ensure that prices are not excessive. Setting retail ceiling tariffs in an effectively competing market may even damage effective competition. If incumbent retail tariffs ceilings are set to a too low level, in particular new entrants, who initially may incur relatively higher costs than an incumbent operator, may find an insufficient margin between retail prices and their own costs which may force them to discontinue competition on this service. Setting retail tariff ceilings must therefore only be used by regulators as a secondary mechanism to protect consumer interests in absence of effective competition. The implementation of retail tariff ceilings must never jeopardise the development of effective
competition. Retail tariff ceilings can and should be removed as soon as
competition is deemed to be effective.
3.5.4 Efficient Interconnection
Efficient interconnection is of key importance to stimulate efficiency and to send the right economic signals to the market concerning make/buy decisions.
New entrants build their own networks to provide services to customers and have to interconnect with the network of the incumbent. This in fact
introduces an additional cost factor for new entrants to provide retail services, compared to the incumbent who does not have this disadvantage (the on-net traffic advantage). This has two important consequences:
• New entrants and incumbents must be stimulated to interconnect in the most efficient way. This means that new entrants and the incumbent must have stable and proper stimuli to invest in rolling out network infrastructure to each others networks at the lowest feasible hierarchical levels;
• Wholesale tariffs must be minimised and preferably reflect the cost level of an efficient operator in order to compensate for new entrant’s
additional network costs. Only then will retail tariffs be able to decrease as a result of competition.
3.6 Cost Allocation Systems
To ascertain the correct costs incurred in the provision of telecoms services, we need a cost allocation system. Cost orientation of tariffs means that tariffs are based on the actual or theoretical cost of providing a service plus a reasonable return on investment. Cost orientation implies the ability to determine the costs of providing a service to such an extent that costs can be allocated to units consumed by a particular service. This is achieved by a cost allocation system that relates investments and operational costs to units consumed expressed in ‘cost drivers’ such as traffic volume or transport distance.
Cost allocation systems can use different accounting principles to identify cost and allocate these to the cost drivers.
In practice the costs that are used in a cost allocation system are either based on Historic Cost Accounting (HCA), or on Current Cost Accounting (CCA).
Differences between HCA and CCA may occur where purchase price and
current price differ, e.g. due to the fact that technology tends to become less
costly over time and labour costs - for example digging - tend to rise over time.
Cost causation refers to the costs that are taken into consideration when determining tariffs. As one extreme, all costs that an operator incurs to provide a service regardless whether these can be justified or not may be taken into account. This approach is mostly referred to as the Fully Allocated Costs (or Fully Distributed Costs) approach. As the other extreme, only costs may be taken into account that would be incurred by a hypothetical efficient operator for providing a specific service. This approach is called the Long Run
Incremental Cost (LRIC) approach.
Cost allocation systems may be applied in a Top Down or Bottom Up fashion.
Top Down means that costs for specific cost elements are taken as a whole and distributed over realised ‘consumption units’ (utilisation) by dividing the total costs by the number of consumption units. A Bottom Up method in contrast takes costs of individual cost elements in terms of -for example- purchase price of assets, resources spent, etc., and determines the usage of those elements by an individual consumption unit. An advantage of a Bottom Up model is that it is very transparent. Disadvantages are that cost allocation of, in particular, organisational resources costs is difficult to allocate and that the Bottom Up approach may therefore not reflect reality
Tariffs that are based on the use of a cost allocation system are usually re- assessed on an annual basis. This is done in order to accommodate changes in input variables of the cost allocation system (e.g. an incumbent’s operational costs, costs of capital, etc.). Such annual re-assessment however may introduce substantial factors of unpredictability, which are undesirable. A method to circumvent this unpredictability is the use of a Price Cap system.
3.7 Accounting Separation
Accounting separation means the keeping of separate revenue and cost accounts for different activities. Accounting separation is required as a means to achieve a detailed and accurate account of the cost and profits made by an operator for a specific activity in conjunction with cost allocation systems that are used to ensure cost orientation. Accounting separation is of key importance for the following reasons:
• It allows objective verification of the cost orientation objectives of regulation and cost orientation policies;
• It allows objective verification of the effectiveness of the cost allocation
system selected to enforce cost orientation and to detect ‘ghost costs’ if
costs and profits of the cost allocation system are reconciled with those in the statutory annual report of the incumbent;
• If published it allows external parties to verify that regulation and policies have been properly implemented.
In order to be effective, accounting separation information therefore must:
• Show the costs and revenues of regulated services of the incumbent at least to the degree that costs, revenues, and return on investment can be identified;
• Show the effects of reconciliation of revenues and costs using the selected cost allocation system with the cost accounting system used by the incumbent for its annual statutory reporting, in order to reveal ghost costs;
• In order to ensure transparency: be publicly available to a sufficient level of detail so that market parties can verify adequate implementation of regulation and policies.
Fighting ‘Ghost Costs’
It is important to notice that accounting separation and cost orientation per se as required under ex ante EU telecommunications regulation, do not prevent a vertically integrated incumbent from cross subsidising retail services from
‘ghost costs’ charged on wholesale services. Nor do they prevent the vertically integrated incumbent from undercutting prices of new entrants who, due to the
‘ghost costs’ charged by the incumbent, incur higher costs for providing the same service than the incumbent does. Such practices of vertically integrated incumbents can only be dealt with ex post under competition law but even then, cross subsidisation practices will be extremely difficult to demonstrate. As procedures under competition law usually also take a very long time,
competition law is largely ineffective to ensure effective competition in the dynamic telecommunications market. Regulators must therefore ensure to fight ghost costs to the maximum possible extent using ex ante regulation.
Cost oriented tariffs may contain ghost costs due the information asymmetry where an incumbent operator may be able to ‘get away’ with allocating more costs to a service than would be justifiable from the view of the regulator. Another
important source of ghost costs is due to differences between principles of the cost
allocation systems used by the regulator and used by the incumbent internally,
which may overstate costs. An example is the valuation of historically low cost
assets on higher current cost (e.g. the local loop).
Ghost costs create a discriminatory cost advantage for the vertically integrated incumbent. Abuse of these cost advantages may be extremely detrimental to the competitive position of new entrants. In addition, if such abuse occurs, it is difficult if not impossible to fight under ex ante telecommunications regulation.
For these reasons, regulators must strive to fight ghost costs to the maximum possible extent, by:
• Preventing ghosts costs to occur, i.e. ensuring that the principles selected for cost allocation systems do not result in higher costs than a vertically integrated incumbent would calculate internally;
• Ensuring that accounting separation reports allow detection of ghost costs, by incorporating reconciliation results of the regulatory imposed cost allocation system with the incumbent’s statutory accounts.
3.8 Price Caps
As an alternative to the annual reassessment of tariffs based on one of the cost allocation methods described above, a Price Cap mechanism can be used. A Price Cap system would set tariffs to an appropriate starting level (based on FAC, EDC or LRIC) and in the following years the tariffs should be lowered so that they remain below the Price Cap. Tariffs can be changed by applying the following formula:
t
i+1= t
i(1 + RPI –X)
Where t
iis the tariff level in year i, RPI is the Retail Price Index (inflation) and X is the percentage with which operational costs go down due to anticipated efficiency improvements. The use of a price cap has several advantages over annual re-assessment of tariffs:
• Tariffs may become more predictable;
• The incumbent may have a better stimulus to become more efficient, since tariffs are not necessarily lowered at an equal rate to the cost improvements made by the incumbent.
3.9 Requirements for tariff regulation
Following the theoretical framework, tariff regulation should contain the
following elements.
3.9.1 Retail Tariffs
Retail Tariffs Effective Competition Efficient Competition
Cost orientation X
Unbundling X X
Price Floor X
3.9.2 Wholesale Tariffs
Wholesale Tariffs Effective Competition Efficient Competition
Minimal wholesale tariffs X X
Cost Orientation X X
Non-discrimination X
Predictability X
Full transparency X
Unbundling X X
Levelled Reciprocity X
Price Floor ico reciprocity X
3.9.3 Other requirements for tariff regulation
Retail Tariffs Effective Competition Efficient Competition
Dynamic regulation X
Price Squeeze measures X
4 Tariff regulation in a European context
This chapter provides an overview of the applicable provisions regulating the tariffs for interconnection, leased lines and voice telephony services. While discussing the implementation of regulation and its effects, the following structure is used:
• Retail services
• Voice telephony;
• Leased lines services.
• Wholesale services
• Fixed interconnection (termination and origination);
• Mobile interconnection (termination and origination);
• Local Loop Unbundling (LLU);
• Wholesale (interconnect) leased lines.
4.1 Tariff regulation by the European Commission
In 1987, the Commission issued its Green Paper on the development of the
common market for telecommunication services and equipment
4, with the
objective of introducing competition into the market to bring consumers lower
prices, freedom of choice and innovation. Liberalisation started in 1988 and
was due to be complete after ten years.
The liberalisation and harmonisation process is represented graphically below.
Figure 5: The EU step-by-step liberalisation process
