removing the non-discrimination obligation in the Dutch fixed call termination market?

(1)

What is the likely impact of

removing the non-discrimination obligation in the Dutch fixed call termination market?

Prepared for KPN

September 2008

(2)

Oxera Consulting Ltd is registered in England No. 2589629 and in Belgium No. 0883.432.547.

Registered offices at Park Central, 40/41 Park End Street, Oxford, OX1 1JD, UK, and Stephanie Square Centre, Avenue Louise 65, Box 11, 1050 Brussels, Belgium. Although every effort has been made to ensure the accuracy of the material and the integrity of the analysis presented herein, the Company accepts no liability for any actions taken on the basis of its contents.

Oxera Consulting Ltd is not licensed in the conduct of investment business as defined in the Financial Services and Markets Act 2000. Anyone considering a specific investment should consult their own broker or other investment adviser. The Company accepts no liability for any specific investment decision, which must be at the investor’s own risk.

(3)

1 Overview 1

1.1 Background 1

1.2 Summary of OPTA’s findings 1

1.3 Structure of the report 2

2 When would bilateral negotiations lead to low FTA

rates and/or B&K? 3

2.1 When is a B&K arrangement likely to be non-strategic? 3 2.2 When is a B&K arrangement likely to strategic? 4 2.3 Which factors may lead to the emergence of B&K

arrangements in the Dutch FTA market? 5

3 What are the likely effects of low FTA B&K

agreements on the retail fixed telephony market? 7

3.1 Benefits of bilateral reduction of termination charges 7 3.2 Implications of removing the ND obligation for competition 8

4 Conclusions 11

List of tables

Table 3.1 Types of competitor in fixed line telephony and their respective interconnection arrangements 9

List of figures

Figure 3.1 Wholesale payments between operators 10

(4)

1 Overview

1.1 Background

Oxera has been asked by KPN to provide an independent economic commentary on specific aspects of OPTA’s draft decision on the fixed call termination market.

¹

The main features of OPTA’s decision are that all operators providing fixed terminating access (FTA) are considered to have significant market power (SMP)—there are symmetrical obligations on all parties except in relation to transparency (only KPN needs to publish a reference offer) and to accounting separation (only for KPN)—and, of relevance here, that the non-discrimination (ND) obligation has been withdrawn and bilateral termination agreements below the price cap are therefore now possible. From May 2008, all operators were subject to such an ND obligation; before then, only KPN was subject to this obligation.

²

With the removal of the ND obligation, operators would have the opportunity to discriminate between different purchasers of FTA, by charging some operators a lower termination rate than others. It would not be possible to exceed the price ceiling set by the regulator, implying that operators would have the option to charge any tariff between zero and the price cap (assuming negative rates are unlikely).

This report examines the reasoning behind OPTA’s proposed withdrawal of the ND obligation in the FTA market and its potential effects on the retail fixed telephony market.

1.2 Summary of OPTA’s findings

As set out in its draft decision of August 15th 2008, OPTA does not consider price

discrimination to be a potential problem for competition unless FTA rates are such that they lead to a situation of margin squeeze.

³

The imposition of a price cap is considered to be sufficient by OPTA to remedy problems of margin squeeze and excessive prices.

⁴

OPTA states that price discrimination below the price cap would not raise competitive

concerns. According to OPTA, price discrimination can be rational, non-strategic and efficient in markets with bilateral buying relationships.

⁵

Moreover, the regulator finds that price

discrimination could positively influence competition, and subsequently increase overall welfare. OPTA identifies two examples in this regard—a reduction in FTA rates could:

– reduce billing costs;

– provide a better basis for the use of retail offerings with fixed prices for unlimited calls.

⁶

1 OPTA (2008), ‘Marktanalyse vaste gespreksafgifte’, ontwerpbesluit, August 15th.

2 It is Oxera’s understanding that OPTA’s decision of April 29th 2008 to impose ND obligations on all operators, as opposed to just KPN, followed from a ruling in 2007 of the Trade and Industry Appeals Tribunal (College van Beroep voor het bedrijfsleven, CBb), in which the CBb found that OPTA’s justification for the non-imposition of a ND on operators other than KPN was not sufficient. See OPTA (2008), ‘Wijzigingsbesluit voor de wholesalemarkten voor gespreksafgifte op afzonderlijke vaste openbare telefoonnetwerken’, April 29th.

3 OPTA (2008), ‘Marktanalyse vaste gespreksafgifte’, ontwerpbesluit, August 15th, p. 78, para 328.

(5)

OPTA finds that differences in FTA may arise due to normal commercial considerations.

Firms may take different views as regards to what comprises profit-maximising behaviour, despite facing the same circumstances. According to OPTA, firms should have the

opportunity to implement those views. It is OPTA’s belief that this would enable the market to reward the most efficient behaviour by selecting the most efficient operator.

⁷

OPTA also states that with an ND obligation in place, operators need to reach agreements with all purchasers of FTA before they reduce the level of FTA. In OPTA’s view, this may reduce competition between FTA providers.

⁸

1.3 Structure of the report

The report is structured as follows.

– Section 2 discusses the strategic and non-strategic factors that may lead operators to agree on low bilateral FTA rates, possibly even bill-and-keep (B&K) arrangements, once OPTA has removed the ND obligation. This section then assesses which of these factors are likely to be present in the Dutch market.

– Section 3 analyses the potential effects of B&K agreements in the market. In particular, it explores the likelihood that the benefits identified by OPTA in its draft decision will emerge. This section also assesses the potential distortive effects that the removal of the ND obligation could have on the retail fixed telephony market.

– Section 4 concludes.

(6)

2 When would bilateral negotiations lead to low FTA rates and/or B&K?

The proposed removal of the ND obligation raises questions about the possible rationale and conditions under which operators are likely to lower FTA rates from their current levels.

Answering these questions will facilitate a clearer understanding of the potential effect that OPTA’s decision may have on competition and consumer welfare. Central to the discussion is whether a reduction in FTA rates would be of a strategic or non-strategic nature.

The term ‘strategic’, as used here, describes firms’ actions that are intended to raise profit at the expense of rivals or to reduce competition. It encompasses both cooperative (collusive) and non-cooperative (unilateral) actions. Not all strategic actions are automatically anti- competitive in the context of competition law or telecoms regulation. This also depends on the likely effects such actions have on competition—only a strategic action that is likely to have a material effect on, or distort, competition could be deemed anti-competitive. On the other hand, when the underlying rationale for firms’ actions is not intended to influence the environment in which they compete, those actions can be described as non-strategic.

Examples include price reductions that stem from efficiency improvements. Non-strategic actions are generally more benign from the perspective of a competition analysis.

In its draft decision, OPTA states that operators’ incentives for agreeing on low bilateral FTA rates can be normal and non-strategic. OPTA provides two examples of potential non- strategic incentives for lowering FTA, namely lower billing costs and providing a better basis for the use of retail offerings with fixed prices for unlimited calls (see section 1.2 above).

This section examines whether FTA agreements below the price cap, following the removal of the ND obligation, would be motivated by strategic or non-strategic factors. If firms have strategic incentives, there could be a risk that the potential benefits of removing the ND obligation may not materialise and, furthermore, that the agreements could have the effect of limiting competition in the market.

It is important to note that it is difficult to predict the level below the price cap at which FTA rates would be set in bilateral negotiations. This report therefore refers interchangeably to low FTA rates or B&K agreements.

2.1 When is a B&K arrangement likely to be non-strategic?

The rationale to engage in a non-strategic bilateral reduction in termination rates can be explained by the following factors.

– Net FTA payments between the negotiating parties are balanced. Asymmetry in existing FTA payments can reduce the probability of reaching an agreement over low reciprocal termination rates. As illustrated in the simple formula below, net FTA payments are a function of traffic balance and the current FTA rates charged by the negotiating parties. Operators are more likely to agree on low reciprocal FTA rates if net FTA payments are close to zero; otherwise, the net recipient of FTA payments would be worse off by moving to B&K.

Net termination payment

1

=

(FTA

1

x incoming traffic from operator 2) – (FTA

2

x outgoing traffic to operator 2)

Equation 2.1

(7)

– Traffic between the negotiating partners is balanced. A critical assumption in the economic literature on mobile-to-mobile termination rates that leads to the emergence of reciprocal termination rates, and bilateral agreements on B&K in particular, is that consumers make on-/off-net calls in proportion to each network’s market shares, and, hence, that traffic between firms is balanced.

⁹

In practice, the traffic balance depends on the calling profiles of the negotiating partners’ customers. If traffic between the

negotiating partners is not balanced, the probability of balanced FTA payments is low.

Hence, the non-strategic incentives of the operator with traffic inflow surplus to agree on B&K would be dampened.

– Negotiating parties have similar cost structures. If negotiating parties have

significantly different cost structures, the operator with the more expensive network may want to agree on a relatively higher FTA rate than the operator with the cheapest

network.

– Operators can achieve significant cost savings. Bilateral B&K agreements could be non-strategic if the avoided transaction costs were sufficiently high to offset potential losses from the negotiating partners, such as forgone net FTA revenue. Examples of these cost savings, as cited by OPTA, could be lower billing and contractual costs.

In practice, all the above factors interact with one another to shape the incentives of operators when negotiating FTA rates. The likelihood that these conditions are met in the Dutch market is discussed in section 2.3.

2.2 When is a B&K arrangement likely to strategic?

B&K agreements may also be strategically motivated—ie, aimed at achieving a competitive advantage in the market. The main mechanism through which this can be achieved would work as follows.

The two (or more) operators involved would agree a B&K arrangement to achieve a competitive advantage over third parties in the retail market by means of retail price discrimination. A B&K arrangement would remove the per-minute (variable) cost of calls terminating from networks that are part of the agreement. This allows the operators to lower outgoing retail prices to these networks, while keeping outgoing prices to networks outside of the agreements the same (or possibly even increase them). This is a form of on-net/off-net price discrimination, where on-net is defined more broadly to include all networks that are part of the low (B&K) FTA agreements.

The economic literature suggests that such price differentials attract customers as a consequence of tariff-mediated network externalities. Put simply, the larger the customer base on which the calls terminate ‘on-net’, the higher the benefit for an additional subscriber to join the network. This effect is only possible if operators can discriminate between

purchasers of FTA (otherwise, reduced FTA would have a proportionate effect on originating calls across the board).

¹⁰

Another, related, strategic driver behind low or B&K FTA arrangements is that bilateral price reductions may have a positive impact on the traffic balance of the negotiating parties in relation to operators that do not form part of the agreements. Depending on the respective price elasticities, retail price differentials between on- and off-net calls facilitated by B&K

9 See, for example, Valletti, T.M. and Houpis, G. (2005), ‘Mobile termination: What is the ‘right’ charge?’, Journal of Regulatory Economics, 28, 235–258. The simplifying assumption of balanced calling patterns with reciprocal termination rates has also been made in Laffont, J. and Tirole J. (2000), Competition in Telecommunications, p. 189, Cambridge University Press, Massachusetts.

10 For a discussion, see Laffont J.J., Rey, P. and Tirole, J. (1998), ‘Network Competition II: Price Discrimination’, RAND Journal of Economics, 29, 38–56.

(8)

arrangements may provide incentives to end-users to make more on-net calls (ie, calls within, to and from the networks with B&K arrangements) and fewer off-net calls. The parties with B&K arrangements would benefit financially from this situation if the forgone on-net revenues were more than compensated by potentially higher off-net prices, as well as by a reduction in outgoing termination payments to networks not part of FTA (B&K) agreements.

2.3 Which factors may lead to the emergence of B&K arrangements in the Dutch FTA market?

This section explores whether a bilateral reduction in FTA rates and/or a B&K agreement between some or all players in the Dutch fixed telephony market could be motivated by the non-strategic or strategic factors described above.

2.3.1 Balance of termination payments

Under the proposed regulatory arrangements, FTA rates of KPN and other operators are based on the termination costs of KPN.

¹¹

Assuming symmetric termination costs, Equation 2.1 described above can therefore be restated as follows:

Net termination payment

1

=

FTA

KPN

x (incoming traffic from operator 2 – outgoing traffic to operator 2)

Equation 2.2 As can be seen, the critical factor in the balance of FTA payments becomes the net traffic flows between the negotiating parties. A net receiver of traffic would have a net revenue surplus and would be worse off by agreeing to lower the FTA rate below FTA

_KPN

(and even more so by entering into a B&K arrangement), whereas a net sender of traffic would have a net revenue deficit and would welcome a reduction in FTA rates below FTA

_KPN

.

Oxera understands that, currently, KPN is a net receiver of traffic relative to other fixed telephony operators. This is confirmed in a study by Lexonomics on behalf of OPTA.

¹²

Similar data is not publicly available for traffic patterns between other fixed and cable operators, but it is not unreasonable to assume that, overall, the market is characterised by at least some degree of traffic imbalance. Traffic balance depends on an array of factors, such as pricing schemes and customers’ calling patterns. Unless all of those factors that influence the proportion of incoming and outgoing calls are very similar between the negotiating parties, balanced traffic would be unlikely.

In any event, unbalanced traffic would make B&K arrangements less likely to occur for non- strategic reasons. By implication, any B&K arrangements that do occur in such

circumstances are more likely to be for strategic purposes. In particular, for the party that is a net receiver of calls, the B&K arrangement would be unattractive except for the strategic rationale—its FTA profit sacrifice would have to be recovered through some other means, such as higher profits in the fixed retail market (once market share has been gained), or cross-subsidy from other activities.

11 In its draft decision, OPTA proposed the withdrawal of asymmetry in regulated FTA, implying that price caps, which are based on KPN’s cost structure, apply to all operators in the market. Tariffs for operators other than KPN are currently regulated according to delayed reciprocity, reflecting KPN’s more efficient scale. Source: OPTA (2008), ‘Marktanalyse Vaste

gespreksafgifte’, ontwerpbesluit , August 15th

12 Lexonomics (2007), ‘De rol van tegenwerkende kopersmacht bij de beoordeling van aanmerkelijke marktmacht van Tele2/Versatel en UPC/Priority in vaste gespreksafgifte’, Rapport van Lexonomics voor OPTA, October 12th.

(9)

2.3.2 Cost savings

In its draft decision, OPTA states that a reduction in FTA rates may lead to lower billing costs.

¹³

In practice, however, these cost savings would be very small or even negligible unless an industry-wide B&K agreement could be reached. This is because billing systems and functions involve relatively large fixed costs, which can at most only be partly avoided if a particular sub-set of traffic flows enters into a B&K arrangement. This is explained in greater detail in section 3. Hence, cost savings are unlikely to be a non-strategic rationale for any B&K arrangements that might emerge in the Dutch FTA market.

2.3.3 Conclusion

The analysis presented above indicates that there do not appear to be strong incentives for operators in the Dutch market to enter into bilateral reductions in FTA rates, other than for strategic reasons.

In the Dutch retail market for fixed telephony, such strategic incentives exist in particular for the cable operators, which do not compete against each other at the retail level because their networks do not overlap. Changes in the retail offerings of one cable operator would have no immediate effect on the operations of another cable operator. Strategic alignments between cable operators may lead to lower FTA rates, which, in return, would reduce the variable per- minute costs of terminating calls at the other cable operator’s network. As a result, cable operators may be able to charge retail tariffs on a flat-rate basis for calls between their networks (‘on-cable’ calls as a variant of ‘on-net’ calls). Cable operators would thus be able to obtain a competitive advantage over other market players.

As noted above, not all strategic actions are automatically anti-competitive in the context of competition law or telecoms regulation. This also depends on the effects that such actions are likely to have on competition—only a strategic action that is likely to have a material effect on competition could be deemed anti-competitive. This is further addressed in section 3 below.

(10)

3 What are the likely effects of low FTA B&K agreements on the retail fixed telephony market?

This section assesses the likely effects that the emergence of bilateral FTA arrangements below the price cap (in particular, B&K agreements) could have on the Dutch retail fixed telephony market.

The analysis focuses on:

– the likelihood that the possible benefits of B&K identified by OPTA actually materialise (section 3.1); and

– the competitive distortions that these agreements may introduce into the market (section 3.2).

3.1 Benefits of bilateral reduction of termination charges

As explained above, OPTA makes the point that agreements below the price cap, and B&K in particular, can be of a non-strategic nature. It provides two examples in this context: direct cost savings in billing expenditure associated with B&K, and the possibility that B&K

arrangements translate into retail tariffs that are beneficial to end-customers.

3.1.1 Avoided billing costs

To assess the potential for cost savings with respect to billing, it is important to understand the underlying cost structure of the billing functions of a fixed operator.

Wholesale billing functions of operators are typically integrated such that a single system is used to monitor all incoming and outgoing traffic, as well as to estimate payment balances for all possible bilateral flows. Similarly, there is a separate integrated billing system to monitor all types of retail traffic.

In other words, the integrated wholesale billing system of any given operator not only

monitors traffic to and from operators with which a B&K arrangement may have been agreed, but also the traffic of all other operators for which termination rates continue to be positive.

Similarly, in order to bill retail customers, other types of traffic would still need to be monitored.

For example, if cable operators were to agree on B&K among themselves (and, as argued above, they may have incentives to do so, given that they do not compete directly in the retail market), it is estimated that the traffic that would be part of the agreement would be around 10–20% of total traffic.

¹⁴

Operators agreeing on B&K among themselves would therefore still need to operate billing systems to monitor all other wholesale and retail traffic flows.

Setting up a billing system involves a relatively large proportion of fixed costs, which have to be incurred regardless of the level of traffic. Consequently, the incremental billing costs of any particular pair of traffic flows, such as traffic between two cable companies, would be very low, if not zero. Therefore, a bilateral B&K agreement between two cable operators, for example, would be unlikely to achieve cost efficiencies in this respect.

14 This is based on OPTA’s SMP analysis, where it is stated that cable-based residential telephony accounts for 10–20% of the market. Source: OPTA (2008), ‘Marktanalyse Vaste gespreksafgifte’, ontwerpbesluit, August 15th, p. 99.

(11)

Only if B&K arrangements were expected to be an industry-wide norm might it be possible to achieve significant savings in billing costs. However, this would require the entire billing system as it stands to be phased out, and, hence, there would need to be zero payments between operators, also with respect to international traffic and interconnection with mobile operators. As regards retail billing costs, the cost efficiencies would be similarly limited by the likelihood that, even if cable-to-cable traffic were based on flat rate tariffs, end-users would still have to pay usage-based prices for other types of voice traffic.

3.1.2 Flat-rate tariffs to consumers

Another potential benefit outlined by OPTA is that the transition to B&K arrangements would provide a more effective basis for fixed prices for unlimited calls, which could be welfare- enhancing. Flat-rate pricing would result because, with a B&K agreement, the operators would no longer pay wholesale charges on a per-minute basis.

While it is reasonable to assume that such tariff structures may become more common as a result of B&K agreements, it is unclear whether these tariffs would be welfare-enhancing.

Improvements in consumer welfare could be expected if the following factors are observed:

– consumers have a good understanding of the various features, including the pricing structure, of the different offers available in the market;

– consumers have a good knowledge of their actual usage patterns, such that they are able to choose the product that offers them the best value for money; and

– as an extension to the point above, it should be relatively easy to compare the products and services of different providers.

If a sub-group of operators in the Dutch market were to introduce flat-rate pricing following a B&K agreement over FTA rates, these conditions would not necessarily be met, and it is unlikely that the potential benefits, as outlined by OPTA, would materialise in full.

3.2 Implications of removing the ND obligation for competition

3.2.1 Distortion between cable and other operators

An important policy question is how the establishment of a bilateral B&K arrangement would affect different competitors in the market. This sub-section extends OPTA’s analysis of the effect on competition resulting from the removal of the ND obligation.

As mentioned above, while it does not seem likely that providers of fixed telephony services (PSTN or VoB) will enter into B&K agreements for non-strategic reasons, cable operators, on the other hand, may have strategic incentives to enter into these type of FTA agreements. As discussed above, bilateral FTA reductions enable differentiated retail prices in the form of

‘on-cable’ rates, and cable operators could exploit the network externalities arising from such rates in order to gain further market share.

The important consideration in this regard is whether this competitive advantage has a distortive impact on overall competition in the market. Therefore, the focus of the analysis is on the impact that strategic B&K agreements between cable operators may have on the ability to compete and on the growth prospects for other competitors.

Different types of competitor, and their respective interconnection arrangements, are summarised in Table 3.1. Cable operators had a combined market share of 22% in 2007, which is expected to increase to a level of 46% in 2011.

¹⁵

The growing market share of cable operators could be explained in part by their strong position in the adjacent TV market, which

15 OPTA (2008), ‘Onderzoek marktontwikkelingen elektronische communicatiediensten 2008–2011 (VKA)’, June 12th.

(12)

the cable operators can exploit when competing with offers that combine TV, broadband and fixed telephony. Indeed, bundled ‘multi-play’ offers account for 70% of the Dutch broadband customer base, and around 50% of these bundles include TV broadcasting services.

¹⁶

The implication is that cable companies are well positioned in the fixed line market, and in that sense, B&K arrangements in FTA between cable operators have the potential to have a discernible effect on the competitive environment in the Dutch market.

Table 3.1 Types of competitor in fixed line telephony and their respective interconnection arrangements

Description Operators

Market share¹

KPN

Based on its own infrastructure KPN 65%

Wholesale line rental (WLR)/carrier pre-selection (CPS)

WLR is a wholesale product that enables entrants to replicate end-user subscriptions over KPN’s network. Entrants can route end-users to their network through a CPS system

Various operators 6%

Local-loop unbundling (LLU)

Potentially complemented by own or leased backhaul infrastructure, new entrants can provide voice services through proprietary networks by means of LLU

Voice-over broadband (VoB)

Having conveyed the call over the IP network, VoB providers interconnect to fixed (or mobile) networks through gateways, and subsequently pay the standard FTA price. VoB is provided by alternative operators—eg, in conjunction with broadband offered by means of LLU and/or bitstream access

Online, Tele 2, Telfort and other DSL operators

7%

Cable

Based on their own infrastructure (VoIP) Ziggo, UPC, Delta 22%

Glass fibre

Operators owning glass fibre network infrastructure

are interconnected with KPN at a regional level Various operators 1%

Notes:

¹

Market shares may not add up to 100% due to rounding.

²

KPN charges entrants the terminating access for calls terminating on its network. For calls outside KPN’s network, termination (possibly packaged with transit service) is purchased from KPN, which would operate as an intermediator, or directly from the receiving operator.

Source: Oxera, based on OPTA (2008), ‘Marktanalyse Vaste gespreksafgifte’, ontwerpbesluit , August. Market data from Verdonck, Klooster & Associates B.V. (2008), ‘Marktontwikkelingen 2008–2011, Addendum bij Marktontwikkelingen 2007-2010 Ontwikkelingen van en scenario's voor de consumentenmarkt voor telefonie, omroep en breedbandinternettoegang’, Onderzoek_marktontwikkelingen_2008-2011_bijlage_telefonie120608, June 12th in OPTA (2008), ‘Onderzoek marktontwikkelingen elektronische communicatiediensten 2008–2011 (VKA)’, June 12th (http://www.opta.nl/asp/publicaties/document.asp?id=2669).

An important characteristic of the network configuration of providers other than KPN and the cable networks (ie, WLR, CPS, LLU and DSL-based VoB) is that it is based on regulated access products provided over KPN’s network.

¹⁷

In this respect, the interconnection with, and call conveyance to, other networks is provided by a third-party operator. Unless the third- party operator has access to the low (or zero) FTA agreements with the cable companies, the wholesale price that these operators would pay to terminate a call on those networks would be similar to that of KPN.

Thus, because FTA rates account for a significant proportion of the costs of an outgoing call, KPN, and all other operators that rely on its inputs, would not be in a position to replicate the offers of cable companies that enter into B&K arrangements and will therefore face a

competitive disadvantage compared with cable. Because cable operators had a 22% market

16OPTA (2008), ‘OPTA Annual Report and Market Monitor 2007’, May, p. 78.

17 As stated in OPTA’s consultation, the regulated products discussed in Table 3.1 are to remain part of the regulatory framework in the Netherlands. OPTA (2008), ‘Fixed Telephony: Broadband and Leased Line Preliminary Draft Decisions:

Context and Perspective’, OPTA/AM/2008/201603, July.

(13)

share in 2007, and this share is generally expected to increase over the coming years (up to 46% in 2011), the impact of this competitive distortion is material.

¹⁸

3.2.2 Illustration of the distortion

Figure 3.1 provides a simple illustration of the arguments outlined above. Operators

engaging in competition for the same end-users (ie, in the territory of cable network 1) face different costs for terminating calls on the other B&K operator’s network (cable network 2).

As illustrated in the diagram, cable operator 2 would only provide a zero termination charge (C

CTC

) to the other B&K operators with which it does not compete in the same retail

markets.

¹⁹

Consequently, it could be difficult for other market players to compete with the cable operators’ retail prices for calls terminated on cable networks (R

c

), as those operators must pay the standard FTA charge (C

2

, which is higher than C

CTC

).

Indeed, an entrant could not equally provide call services to cable operators’ networks if the retail charge for cable-to-cable traffic charged by a cable operator were lower than the

‘standard’ FTA charge, plus the entrant’s other unavoidable costs, including wholesale charges for call origination and transit (ie, R

c

< W

OT

+C

2

+retail costs).

Figure 3.1 Wholesale payments between operators

Cable network 1 retail price R_C

KPN fixed network

FTA charge C_CTC=0

FTA charge C₂> C_CTC

Entrant (WLR / CPS) retail price R_E

Wholesale charge for origination and transit W_OT

Cable network 2

Consumers in area 1 retail price R_E

retail price R_C

Cable network 1 retail price R_C

KPN fixed network

FTA charge C_CTC=0

FTA charge C₂> C_CTC

Entrant (WLR / CPS) retail price R_E

Wholesale charge for origination and transit W_OT

Cable network 2

Consumers in area 1 retail price R_E

retail price R_C

Source: Oxera.

18 Based on market data from Verdonck, Klooster & Associates B.V. (2008), ‘Marktontwikkelingen 2008–2011, Addendum bij Marktontwikkelingen 2007-2010 Ontwikkelingen van en scenario’s voor de consumentenmarkt voor telefonie, omroep en breedbandinternettoegang’, Onderzoek_marktontwikkelingen_2008-2011_bijlage_telefonie120608, June 12th in OPTA (2008),

‘Onderzoek marktontwikkelingen elektronische communicatiediensten 2008–2011 (VKA)’, June 12th (http://www.opta.nl/asp/publicaties/document.asp?id=2669).

19 In the simplified framework set out in Figure 3.1, same logic applies in the territory of cable network 2.

(14)

4 Conclusions

OPTA’s draft decision on FTA proposes to remove the ND obligation on all operators. In this report, Oxera has discussed some of the implications of this removal for competition in the retail fixed telephony market.

In particular, the report has analysed the non-strategic and strategic incentives that operators may have to engage in B&K arrangements. The analysis suggests that non-strategic reasons for bilateral FTA agreements—including cost savings and benefits rising from simplified tariffs—are not likely to explain the emergence of bilateral reductions in FTA rates. On the other hand, the analysis suggests that if these agreements were to arise, it is likely that this would be for strategic reasons.

In the Dutch retail market for fixed telephony, such strategic incentives exist in particular for the cable operators, which do not compete against each other at the retail level because their networks do not overlap. As a result, cable operators may be able to charge retail tariffs on a flat-rate basis for calls between their networks (‘on-cable’ calls as a variant of ‘on-net’ calls).

KPN, and all other operators that rely on its inputs, would not be in a position to replicate the offers of cable companies that enter into B&K arrangements and will therefore face a

competitive disadvantage compared with cable. Because cable operators currently have a significant market share (22% in 2007), and this share is generally expected to increase over the coming years, the impact of this competitive distortion is significant.

As far as Oxera is aware, OPTA’s removal of the ND discrimination would be one of the first such decisions in European telecoms regulation. Oxera has reviewed national regulators’

decisions on FTA in eight other EU Member States.

²⁰

– In five Member States—Belgium, Finland, Germany, Ireland and Sweden—the ND obligation is imposed on all fixed telephony operators, as it is currently in the Dutch market.

– In one Member State—Italy—the ND obligation is imposed on 18 of the main operators.

– In two Member States—Spain and UK—the ND obligation is imposed on the incumbent only (as in the Dutch market before OPTA changed this following the court decision referred to in section 1).

20 Sources: SMP decisions for the respective countries: Sweden: Post- och Telestryrelsen (2004), ‘Beslutsutkast: EkomL for foretag med ett betydande marknadsinflytande pa den svenska marknaden for smatalsterminering I individuella allmanna telefonnat via en fast anslutningspunkt’, p. 17, May; May; Ireland: Commission for Communications Regulation (2007), ‘Market Analysis – Interconnection Market Review - Fixed Wholesale Call Termination Services’, October; Commission del Mercado de las Telecommunicaciones (2004), ‘Proyecto de medida relativo a la definición de los mercados de terminación de llamadas en las redes públicas individuales de cada operador de telefonía fija, el análisis de los mismos, la designación de operadores con poder significativo de mercado y la propuesta de obligaciones específicas’, July UK: Oftel (2003), ‘Review of fixed geographic call termination markets’, Augustl; Germany: Regulierungsbehörde für Telekommunikation und Post (2005), Notifizierung bezueglich Zuführungs-, Terminierungs- und Transitleistungen im öffentlichen Festtelefonnetz, February, European Commission (2006), decision concerning Case DE/2006/0402; European Commission ( 2007), decision concerning Case DE/2007/0679;

Italy: European Commission (2006), decision concerning Case IT/2006/0407 and IT/2006/0408; European Commission (2008), decision concerning Case IT/2008/0777; Belgium: Belgisch Instituut voor postdiensten en telecommunicatie (2006), ‘Besluit Betreffende De Definitie Van De Markten, De Analyse Van De Concurrentievoorwaarden, De Identificatie Van De Operatoren Met Een Sterke Machtspositie En De Bepaling Van De Gepaste Verplichtingen Voor De Markten Van De Cluster “Vaste Telefonie”, Geselecteerd In De Aanbeveling Van De Europese Commissie Van 11 Februari 2003’, August 11th; Belgisch Instituut voor postdiensten en telecommunicatie (2006), (2007), ‘Betreffende De Definitie Van De Markten, De Analyse Van De Concurrentievoorwaarden, De Identificatie Van De Operatoren Met Een Sterke Machtspositie En De Bepaling Van De Gepaste Verplichtingen Voor De Markten Van De Cluster “Vaste Telefonie”, Geselecteerd In De Aanbeveling Van De Europese Commissie Van 11 Februari 2003’, March 7th. Brussels, European commission (2006), Decision concerning Case BE/2006/0435, Case BE/2006/0436, Case BE/2006/0437, Case BE/2006/0438, August 14th. Finnish Communications Regulatory Authority (2007), ‘Markkina-analyysi kiinteaan puhelin verkkoon laskevan puheliikenteen markkinoilla’, December.

(15)

It appears from all these decisions that the ND obligation in FTA markets is an area that has generally received relatively little attention to date. Given that OPTA is one of the first regulators to remove the ND obligation altogether, it would be in line with good regulatory practice to carefully consider and balance all the arguments for and against such a decision.

The arguments presented in this report are intended to assist OPTA in carrying out such an

analysis.

(16)

removing the non-discrimination obligation in the Dutch fixed call termination market?

What is the likely impact of