Telecommunications Interconnection: A Literature Survey

Prepared for


Asia Pacific Economic Cooperation (APEC) Organization




Julian K. Wright*

Eric K. Ralph*

D. Mark Kennet**






ABSTRACT:  This paper reviews a large sampling of literature from professional journals and regulatory agency publications that discusses interconnection between telecommunications networks.  Each paper is reviewed for its relevance according to a set of guidelines laid down by representatives of the APEC economies.  Ideas from these papers will be synthesized in a later report to form the basis of APEC guidelines for member economies on the appropriate management of interconnection.  We also develop a ¡°survey instrument¡± querying member economies on the current status of interconnection policy, and the results of this survey are summarized in an attached Excel worksheet.




The two most technically difficult, politically challenging, and important issues facing telecommunications regulators throughout the world are the appropriate roles for government in both regulating network interconnection and determining universal access policy.  Although this paper focuses on the first subject, as we shall see the two issues are quite closely related.


Recognizing the importance of the issue, the Interconnection Resources Project was approved by the Telecommunications Working Group (TEL) of the Asia Pacific Economic Cooperation (APEC[1]) organization at its 19th meeting in March 1999 (held in Miyazaki, Japan).  The Budget and Management Committee of APEC approved funding for the project from the Year 2000 Operational Account at its July 1999 meeting.  This report is part of a series of tasks to be carried out in response to the contract awarded by APEC to the authors to assist in the Project.


This Report is organized as follows.  First, we give an Overview of the issue and the challenges facing APEC (and for that matter, all) economies.  Second, we review a large sampling of professional literature addressing the issue.  Each article is classified according to a set of criteria identified by TEL as to its relevance for a practical policy implementation.  A short conclusion summarizes the literature review, and we conclude with a brief discussion of results of a ¡°survey¡± submitted to member economies.





Interconnection refers to the procedures by which independently-operated telecommunications networks are joined in such a way that they can function as a coherent whole. It is not difficult to show that subscribers of the two separate networks are made better off – and that therefore society is made better off – if interconnection enables users of each network to take advantage of what economists term ¡°network externalities.¡±


Historically, local telephone companies, whether regional or national, have been interconnected to one another just as they have been interconnected with long distance, mobile and international carriers. Until quite recently, however, interconnecting carriers were franchised monopolies and the kinds of interconnection involved was primarily restricted to monopoly‑to‑monopoly agreements. Furthermore, the terms of interconnection reflected government involvement to a much greater extent than market forces. With the advent of competition and a nascent ¡°network of networks¡± interconnection issues have come to the forefront of regulatory policy. Some of the issues associated with this new environment can be viewed as interim steps on a path to ¡°deregulation.¡± There is an increasing awareness, however, that regulatory oversight of interconnection arrangements may be required in the long run.


Interconnection between competing networks is a critical mechanism for introducing market forces into a monopolistic telecommunications environment. In many jurisdictions throughout the world, debate has focused on what are essentially variants of three major policy approaches.  In the remainder of this section, we describe the relative advantages of these interconnection compensation mechanisms.  We will consider (1) interconnection pricing based on long run incremental costs (LRIC), (2) the ¡°efficient component pricing rule¡±, and (3) a ¡°bill and keep¡± or ¡°sender keep all¡± approach, in which interconnecting networks would recover costs from their own end users, but not from one another.


A central issue lurking behind the entire discussion is whether parties are allowed to negotiate their own interconnection agreements.  In that case, the principles discussed in this document need only be used in the case the parties fail to agree.  In other words, these principles are not necessarily binding on the parties.  In general, although exceptions may exist, economists would argue that a freely negotiated contract between interconnecting parties should dominate any regulated or mediated solution.  The exceptions would occur in the event the agreement were set up as a form of collusion between operators; accordingly, many regulatory authorities reserve the right to review such negotiated agreements.


A variant of this question is that of whether the regulator sets terms, enforces a set of principles, or serves as a ¡°final offer mediator,¡± as in the case of Guatemala.


Cost-Based Interconnection Pricing


It is common in regulatory circles to argue that an efficient price for interconnection would be one based on the long run incremental cost of the services directly attributable to interconnection; however such a mechanism can only be dynamically efficient where lump sum taxes can be raised to cover shared network costs. Unfortunately, such lump sum taxes are infeasible. This means, that in practice, the interconnection price will not only be used to cover incremental costs, but also to provide a contribution towards other network costs, such as overheads.


In the remainder of this section we consider in more detail the nature of the costs that telecommunications firms incur.  A generic telecommunications network produces outputs consisting of access to the network and the carrying of traffic among customers.  Network providers incur costs in providing these outputs that must ultimately be recovered from their customers.  From the point of view of an individual network, an interconnecting network is a customer, although it may or may not be possible to distinguish between a network customer and a business or residential end-user customer. We distinguish between costs that are traffic sensitive (TS) or non-traffic sensitive (NTS) according to whether the cost varies with the number of minutes of traffic carried on the network.  We also define as attributable costs those NTS costs that can be directly attributed to individual customers (either end users or interconnecting networks).  Since traffic is inherently between or among customers, all TS costs are in this sense attributable to at least a pair of customers (see footnote 2). 


Costs are incurred in connecting end-users to the networks. Wireline networks connect users to end offices by means of a local loop, in which costs are largely NTS. Mobile networks connect customers to cell sites using wireless technologies, which are largely TS.  Both wireline and mobile networks contain switching capacity and interoffice trunking capacity.  The cost of switching capacity depends on the number of customers who can be connected through the switch, as well as on the number and duration of calls passing through the switch.  The cost of interoffice trunking facilities that connect local and tandem switches depends on the number and duration of calls transported. Since the facilities used for switching and interoffice trunking are available during all hours of the day and all days of the year, TS switching and trunking costs depend on the maximum, or peak period traffic that is carried in the network. Interoffice trunking facilities that are dedicated to the use of a single interconnecting carrier, such as a dedicated transport facility connecting an entrant¡¯s point of presence with an incumbent carrier¡¯s switching center, are characterized by NTS costs.


We next discuss the specific types of costs associated with interconnection: These costs include (1) costs incurred in provisioning each network for the seamless transfers of calls through appropriate software modifications of switches; (2) the cost of dedicated transport facilities connecting two networks; and (3) the cost of switching capacity (and possibly also some interoffice trunking) on each network to accommodate the peak traffic generated by the interconnection of the networks


Costs Not Directly Attributable to Interconnected Networks


Both wireline and mobile carriers incur costs in connecting end-users to their networks. NTS costs attributable to subscribers of a wireline local network should in principle be recovered by means of a flat rate charge to the customer causing the cost.[2]  That is, the price that an individual pays for access to the network should be set equal to at least the incremental cost of connecting that individual to the network. Customers of wireless mobile networks incur TS sensitive costs in connecting to the nearest local switch, since airtime is shared among multiple users. These costs might, in principle, be recovered in an efficient interconnection pricing regime. In the remainder of this discussion, however, we consider only the setting of interconnection prices between fixed carriers or for the termination of calls from a mobile carrier on a fixed network.


Networks also incur NTS costs that cannot be attributed to any particular subscriber. Carriers incur costs in upgrading their networks to enable them to interconnect with other networks—for example, provisioning each network for the seamless transfers of calls through appropriate software modifications of switches and related costs.  These costs are not traffic sensitive, nor are they directly attributable to any particular interconnecting network. Networks also incur overhead costs for corporate operations and other network functions (e.g., universal service and emergency services). Economic principles would indicate that each interconnecting carrier should recover its overhead costs from its own end-user customers.  This form of recovery would be efficient as long as it does not affect any customer's decision to subscribe to the network.  If subscription charges in excess of the incremental cost of access lead to customer defections, a social cost is incurred which must be weighed against the costs of other methods of overhead cost recovery.  These alternative methods could include TS charges on the network¡¯s own subscribers, a fixed interconnection charge on interconnecting carriers, or through a TS charge applied to interconnecting traffic.



Non-Traffic Sensitive Costs of Facilities Connecting the Networks


The dedicated facilities connecting two local networks are generally NTS, like end users¡¯ loops and the dedicated entrance facilities that long-distance carriers use to interconnect with local switched networks. The costs of these facilities could be shared and jointly provided on a ¡°meet point¡± basis, with neither carrier paying the other for the portion of the facility that the other provided. Alternatively an entering network provider could pay the incumbent network a non-traffic sensitive charge for the use of the incumbent¡¯s facilities.



Traffic-Sensitive Network Capacity Costs


TS costs are generally a function of the volume of traffic during peak load periods, not total traffic, since peak period demand determines how much capacity a network must construct and the cost of carrying an additional call off-peak is virtually zero.  Interconnection charges would more accurately reflect underlying network costs if TS costs were recovered through charges for network capacity, rather than through uniform charges based on aggregate minutes of use.   An example of capacity-based pricing is the traditional peak load pricing used in the electric and natural gas industries, in which a uniform price is charged to every minute of peak period demand so as to recover peak period capacity costs, while a lower price is charged for off-peak usage.  In the context of telecommunications, the off-peak costs of usage are negligible, so peak capacity costs should primarily be recovered in rates for peak period usage, with much lower rates for non-peak usage.


Under peak load pricing, users pay a uniform price per unit for every minute of peak period traffic.  This price is set to fully recover the cost of the capacity required to carry all peak period traffic.  Off-peak traffic does not create any cost because the capacity already exists and, ideally, the price for carrying off-peak traffic should be zero.  In practice off-peak prices may be greater than zero because peak period demand may not be well defined if there are extreme differences in the price of peak and off-peak traffic. 


Some existing interconnection arrangements are based on average cost pricing.  The average cost of capacity (i.e., the total cost of capacity divided by total peak and off-peak usage) is far lower than the average peak cost of capacity (i.e., the total cost of capacity divided by peak usage.)  While a charge based on average cost would generate the same amount of revenue as a charge based on average peak cost, it does not have any of the same efficiency properties. Such a pricing mechanism could create incentives for distorted customer usage patterns as well as anti-competitive consequences, if the incumbent carrier offers its own end-users peak/off-peak pricing while offering average cost pricing to interconnecting networks. During off-peak periods, average cost pricing puts interconnecting networks at a disadvantage in competing with the incumbent network for end-user traffic.  During peak periods, average cost pricing gives interconnecting networks an incentive to impose excessive traffic and capacity costs on the incumbent network.



Measurement of TS and NTS costs.


Perhaps as contentious as the choice of compensation mechanism for interconnection is the measurement of its economic cost.  While there is fairly wide consensus in the interconnection literature that the appropriate standard is long run economic cost,[3] the consensus on what that means is only very slowly emerging.

Economic theory can offer some guidance on this theme.  Standard theory tells us that firms make decisions based on prices and economic costs; in particular, firms base their decisions on the relationship between prices and forward-looking economic costs.  For this reason most often the literature takes the definition of economic costs to be forward-looking economic costs. These are the costs which would be incurred if a new service were to be provided, or avoided if an existing service¡¯s provision were to be ceased, assuming that all inputs of the firm can vary freely (thus the term ¡®forward-looking¡¯ or ¡®long-run¡¯). Considering the long-run economic cost ensures that the firm recovers all of its costs, not only operating and maintenance costs (which vary in the short run), but also

a reasonable return on its capital investment.[4] 

Accounting costs, on the other hand, are historical costs (embedded costs) as registered in the books of a firm and pricing based on embedded costs would normally fail to make that connection between economic costs and prices, thus leading to inefficiencies in the allocation of resources. Moreover, it is likely to be inconvenient to base prices or universal service obligation subsidies on information in the hands of the firms themselves (the well-known problem of asymmetric information). Only forward-looking economic costs can give operators in the market the right signals for entry, investment and innovation. The accounting auditing approach to cost assessment is by far the best known and the least information-demanding and time-consuming methodology. Accounting information is readily available to the regulator. The accounting approach relies on embedded costs recorded in the companies¡¯ books, so a regulation based on this methodology would be close to a cost-plus regulation, providing little incentives to the firms for cost minimization. Thus, productive efficiency could be thwarted.

Another problem with the accounting approach is that it is based exclusively on information provided by the firms, with no independent checking. This is typically all the information that the license or the regulator himself asks the companies for, and it should be clear after the former discussion that it is not enough, if the regulator wants to minimize the asymmetric information problem and the informational rents the firms can earn. So, to improve regulation, more information is needed. Because the firm will have no incentives on its own to reduce the asymmetry of information, the regulator should make an effort to collect information beyond accounting data. This will allow him to implement a cost proxy methodology that could enhance cost assessment for various regulatory purposes.

The alternative to accounting methodologies would be to use simulation exercises, based on technology and market parameters mainly, which rely much less on historical data, as is done for universal service computations in the United States. These alternative methodologies provide a non-discretionary framework within which regulators and firms can discuss with a significant degree of objectivity, and which could provide an independent check on the accuracy of firms¡¯ cost studies. However, lunch is not free: these alternative approaches require much more time and effort in both data collection and preparation as well as the time and effort spent on model design. Since accuracy only comes at a cost (the cost of information), there will normally exist a tradeoff.

In a paper otherwise devoted to universal service cost calculation, Benitez et al. (1999)[5] classify the proposed methodologies into two broad categories: financial models (as used in the UK and Australia) and engineering-economic models (as used in the USA). Both approaches agree on defining the relevant costs for regulation as incremental costs. The former[6] focuses on the overall financial performance of the firm with and without providing service to certain areas or customers (or groups of areas or groups of customers), thus concentrating on the costs the firm would avoid if it were to cease providing the service (avoidable costs) and the revenues it would stop receiving (foregone revenues). The analysis also takes into account a variety of factors that may indirectly affect the financial performance of the firm.  The cost of interconnection is defined as the cost a network would avoid if it ceased to provide the service to an interconnecting network (long-term avoidable costs) and the revenue that it would stop receiving (foregone revenue). Long-term avoidable costs include operating costs, depreciation and a reasonable return for the capital used. Long-term means the period of time in which all the assets are replaced, implying that all the capital equipment costs that the service provider would stop needing, should it disconnect the service to the interconnecting firm, shall be included in the analysis. Due to the fact that the long-term avoidable cost is an economic concept, the appropriate approach to evaluate the assets should be that of considering the asset replacement cost, such as it is considered, for example, in current costs accounting (CCA).

The measure of the long-term avoidable cost obtained from the cost information submitted by the incumbent service provider will show the costs incurred by that operator. However, an efficient degree of avoidable costs should be used when assessing interconnection cost:  the other operators should not have to pay for the incumbent service provider¡¯s inefficiency. Then, an efficiency adjustment must be applied to the long-term avoidable costs incurred so as to obtain an estimation of the efficient level of costs.[7]

It would appear that financial models share some of the undesirable properties of historical embedded cost (HEC) models (accounting/auditing approach). Both types of models rely on firm-reported cost data; the major difference is that the financial model uses a current cost accounting methodology rather than the HEC standard.[8]  Additionally, it may be difficult to obtain the detailed information on component costs required for interconnection purposes; typically, financial models are highly aggregated and do not consider de-averaged costs, which may lead to inefficient price signals.


Engineering-economic models have been developed in recent years as an alternative to the traditional econometric and accounting approaches to cost assessment. Engineering models offer a more detailed view of cost structures than is possible using econometric data. The engineering models (also known as cost proxy models) could enable the regulator to estimate the forward-looking economic cost of the service without having to rely on detailed cost studies that otherwise would be necessary. Proxy models can be useful for many regulatory purposes, such as determining levels of universal service support in high cost areas as well as pricing of unbundled network elements and interconnection. An economic cost proxy model begins with an engineering model of the physical local exchange network, and then makes a detailed set of assumptions about input prices and other factors.  The model calculates the cost of service that an efficient provider would incur if it were to enter the market at that point in time, utilizing current best-practice technology.  While the information demands of this approach are fairly rigorous, the results are free of the asymmetric information problem and the approach can generate externalities in the form of far greater understanding of network economics on the part of regulatory staff.  Of course, like any modelling approach, the results will only be as good as the assumptions made.



Opportunity Costs and ECPR


William Baumol and other economists have proposed a principle for compensation for interconnecting networks that they call the ¡°efficient component pricing rule¡± (ECPR).[9]  The ECPR theory, a variant of the long run incremental cost plus contribution approach, applies to situations where one carrier, typically an incumbent monopoly provider, can provide both a retail, end-to-end service and an input service such as a form of interconnection, that a new entrant needs to purchase in order to provide the retail service competitively.  Under the ECPR theory, the incumbent carrier's price for the essential input service should be equal to the "the input's direct per-unit incremental cost plus the opportunity cost to the input supplier of the sale of a unit of input."[10]   Proponents of the ECPR define the opportunity cost as the price of the final, retail service minus the incumbent's LRIC of supplying that service.  In other words, under the ECPR, the incumbent would be entitled to recover from new entrants all the net revenue that it would have obtained had it remained a monopoly and continued to provide all of the final service itself.


The proponents of the ECPR argue that it can lead to efficient entry signals to rival networks under the following conditions:  (1) the incumbent and the rival produce retail services that are perfect substitutes for one another; (2) there are no substitutes for the input service (i.e. the technology which transforms inputs into final outputs is one of fixed proportions); (3) the price of the final product is optimally determined, both before and after the entry of the competitive provider, either by market forces (in a contestable market framework) or by regulation; and (4) there are no sunk costs associated with entry.  If these conditions are met, the ECPR will yield efficient results in the sense that a rival producer of the final service will enter if and only if its total cost, including the purchase of the essential input from the incumbent, is less than the incumbent's total cost.  When the above conditions do not hold, however, static efficiency concerns require that the interconnection price should in general be lower, because the correct opportunity cost may be less than the difference between price and the incumbent's incremental cost of production.[11]  For example, when the entrant and the incumbent produce goods that are not perfect substitutes, the incumbent's outputs do not decline on a one-for-one basis with sales by rivals, and consequently revenue losses to the incumbent are less than implied by the ECPR.


Economists have also demonstrated that, when an entrant possesses market power or when there are sunk costs associated with entry, the ECPR does not yield optimal prices.[12] If, prior to entry of the new competitor, the prices of the final service are not optimal, then entry by rival firms can benefit consumers by reducing final product prices.[13]  If consumers will benefit from the introduction of a new service, then entry may be socially justified even if the entrant is a less efficient producer than the incumbent.[14] In each of these circumstances static efficiency dictates that the price of interconnection should be set lower than the ECPR level. Similarly, when firms can vary both quality and price, the ECPR can lead to sub-optimal outcomes.[15]


A more serious objection is that the ECPR requires the active participation of a regulator to set final product market prices efficiently, to generate incentives for innovation by incumbent firms, and implicitly to allocate production among incumbents and rivals by determining which firms own the right to serve a particular market (and hence which firms have the right to claim compensation for an opportunity cost of foregone revenue).  In a competitive market, no firm has an implicit property right to serve any particular segment of the market.  Hence opportunity costs do not arise as defined by proponents of the ECPR.


Bill and Keep (Sender Keeps All) Arrangements


Under ¡°bill and keep¡± compensation arrangements, there are no monetary transfers between the networks.  Instead, each network recovers from its own end-users the TS costs of both originating and terminating traffic over its network, as well as NTS costs.  Sender keeps all is equivalent to LRIC pricing if LRIC is zero for both networks.  When both networks have zero TS costs, as could be the case for off-peak traffic, a sender keeps all arrangement provides for static efficiency. For peak period traffic, however, the incremental cost of capacity is not close to zero, and sender keeps all arrangements are likely to promote inefficient interconnection arrangements.


To see this, suppose that the incremental cost of an additional unit of capacity is positive, but that both networks adopt bill and keep. Several distortions are likely to result. If firms choose retail prices, they will set retail prices below the incremental cost of calls.  This is because they face termination costs that are below true cost.  Also, the networks will exploit arbitrage opportunities, such as attracting firms that make a lot of outgoing calls (e.g. telemarketers). For such firms they receive the full retail price for these calls, but do not have to pay a cost to terminate them. Finally, networks will under-invest in network development, since they know that they can use the rival¡¯s network for free. Such behaviour is commonplace in the Internet, where bill and keep has been the normal practice between network providers (this is known as peering).


Thus in general bill and keep is not consistent with efficient usage pricing and does not promote entry by competing networks. 





In this section, we have collected a sampling of professional literature on the subject of interconnection from journals, working papers, government agencies, and intergovernmental agencies.  We provide a brief synopsis of each paper, and we classify each according to the principles set forth in the APEC-TEL Request for Proposal.


The classification key is as follows


(1)   The paper is useful outside the commercial/regulatory culture of a particular economy

(2)   The paper points toward international best practice

(3)   The paper points toward commonly accepted practice

(4)   The paper is appropriate for the APEC region

(5)   The paper has a practical element




ACCC, (1997), ¡°Access pricing principles - telecommunications,¡± mimeo,



Provides broad pricing principles for access pricing, the primary one being that access prices should be cost based according to TSLRIC (Total Service Long Run Incremental Cost).  Provides four pricing guides that allow a regulator to measure whether access prices are consistent with the broad pricing principles.  Also provides some guidance on how a TSLRIC approach can be implemented.




ACCC, (1998), ¡°An assessment of Telstra¡¯s access undertakings: a report for the ACCC¡±



This report, done by Ovum Pty, compares international interconnection charges with those proposed by Telstra in its 1997 undertaking. The study compares PSTN, GSM and AMPS originating and terminating access charges in Australia with several other, predominantly European, countries.   




APEC, 1997, ¡°Effective Interconnection in the APEC Region: A report for the APEC Telecommunications Working Group¡± ..//telwg/interTG/ovum.html



This study on interconnection issues in the APEC region provides a summary of the arrangements for establishing interconnection charges in APEC member economies.   It notes the preference exhibited in most interconnect frameworks among APEC member economies is for commercial negotiation between the operators, with some degree of regulatory oversight.  The best level of regulatory involvement in the setting of interconnection charges is discussed, recognising that this will differ across APEC member countries. 




APEC, 1999, ¡°APEC Principles of Interconnection (Finalized Text: May 14,1999)¡± ..//telwg/interTG/inter-new.html



Contains a set of eight fundamental principles for interconnection that are designed to be able to be adopted by the APEC member economies.  The following are member documents outlining how they have adopted the APEC principles:

APEC Principles of Interconnection: As Implemented in Singapore as of August 1999

APEC Principles of Interconnection as Implemented in the United States as of August 1999

APEC Principles of Interconnection – Canada September 14, 1999

APEC Principles of Interconnection: Japan February 24, 2000




Armstrong, M. (1998), ¡°Network Interconnection in Telecommunications¡±, The Economic Journal, 108: 545-564.



Provides a generic treatment of two-way interconnection.  Shows that interconnection prices can be used for collusion between two networks that are interconnected – that is, firms can increase per-minute interconnection prices to achieve per-minute retail prices that are at monopoly levels. Policy implication is that two-way interconnection charges should be regulated at cost (or below cost to the extent the firms¡¯ mark-up costs at the retail level).  These conclusions rest on the assumption that retail prices only include per-minute prices and that firms do not set monthly rentals.  




Armstrong, M., C. Doyle, and J. Vickers (1995), ¡°The Access Pricing Problem: A Synthesis¡±, Journal of Industrial Economics, XLIV (2): 132-150.



Shows how to modify the ECPR rule to achieve the Ramsey (optimal) approach to access pricing, when the standard assumptions used to justify ECPR are relaxed.  The paper argues that the opportunity cost of providing access needs to be adjusted (reduced) to take into account such things as product differentiation, bypass, and input substitution possibilities.  Assumes retail prices are regulated.  Implementation of such a Ramsey ECPR  requires information likely to be beyond the scope of a regulator.




Armstrong, M. and J. Vickers (1998), ¡°The Access Pricing Problem with Deregulation: A Note¡±, Journal of Industrial Economics, 46 (1): 115-121.



Extends the previous paper (Armstrong, M., C. Doyle, and J. Vickers (1995), ¡°The Access Pricing Problem: A Synthesis) to the case of retail price deregulation.  Shows that the optimal regulation of the margin between retail and wholesale prices entails the ECPR rule, but that regulation of the level of access prices rather than the margin leads to higher welfare.




Baumol, W. J., and J. G. Sidak, (1994b), ¡°The pricing of inputs sold to competitors¡±, Yale Journal on Regulation, 11 (1): 171-202.



The most cited reference to the ECPR access rule.  This rule states that the price of an input should equal its average incremental cost, including all pertinent incremental opportunity costs. These opportunity costs include any profit foregone caused by the competitor taking retail custom away from the access provider.  The main goal of ECPR is to avoid inefficient entry.  However, to achieve allocative efficiency this approach assumes downstream (retail) prices are regulated.  It recommends retail prices should be set between incremental and stand-alone costs.




Brennan, Timothy, (1997), ¡°Industry parallel interconnection agreements¡±, Information Economics and Policy, 9, 133-149.



Discusses the US telecommunications act of 1996 and the FCC¡¯s interconnection order.  Then argues that two-way interconnection will lead to some collusion through the setting of interconnection prices.




Brock, G. W. (1995), ¡°The economics of interconnection¡±, paper prepared for the Teleport Communications Group, April.

(1, 5)


An economist with AT&T provides three papers that analyse interconnection arrangements. Focus is on regimes that are administratively simple. In this regard, recommends ¡°Bill-and-Keep¡± in which originating networks do not need to compensate terminating networks for carrying their traffic.




Bureau of Transport and Communications Economics, 1995 ¡°Interconnection pricing principles: a review of the economics literature¡± Working Paper No. 17.



Provides a discussion of the possible alternative access pricing rules and their advantages and disadvantages. Concludes there is no single answer to what is the most efficient rule for pricing interconnection. Different regulatory, competitive and financial, and technological scenarios are likely to influence the efficiency, or optimality of any given rule in a particular scenario. Suggests the Ramsey approach to access pricing requires too much information to be practical. Also suggests there is a consensus that the access price should be between the limits of long-run incremental costs (allowing for a mark-up to recover fixed and common costs) and ECPR. The appendix contains a brief outline of interconnection pricing practices in Australia, France, Japan, New Zealand, United Kingdom, and United States.




Carter, M. and J. Wright, (1999), ¡°Interconnection in Network Industries¡±, Review of Industrial Organization, 14 (1): 1 - 25.



Provides a generic treatment of two-way interconnection with deregulated retail pricing.  Shows that interconnection prices can be used for collusion between two networks that are interconnected – that is, firms can increase per-minute interconnection prices to achieve per-minute retail prices that are at monopoly levels. The paper also allows a strategy-choice of refusing to interconnect and shows that forcing interconnection in such circumstances is not always welfare enhancing.  Like Armstrong¡¯s (1998) paper above, the results rest on assuming retail prices are solely per-minute charges.




Carter, M. and J. Wright, (1999), ¡°Interconnection in Local Networks: The Clear-Telecom Dispute¡±, The Economic Record. 1999



Discusses the Clear-Telecom interconnection dispute in New Zealand, explaining the economics behind it.  Explains why the entrant¡¯s preferred interconnection agreement may not be consistent with the public interest, but how the entrant will nevertheless attempt to alter their bargaining position by pushing for the threat of a regulated outcome that is in the public interest.




Carter, M. and J. Wright, (1999), ¡°Local and Long Distance Network Competition¡± Centre for Research in Network Economics and Communications Research Paper, University of Auckland.



Provides a model of two-way interconnection with deregulated (and non-linear) retail pricing, allowing for the termination and origination of long-distance calls.  Shows that provided networks have to agree to reciprocal interconnection charges, they will not be able to use two-way interconnection pricing to collude. Shows that the incentive firms have to set high access prices for termination and origination of long-distance calls is weakened since high access charges will intensify the competition to capture this access revenue by competing more vigorously for market share of the local telephone market.  Thus, argues that the main access pricing principles needed are reciprocity and non-discrimination (where equal services are being provided).




Cave, M., (1997), ¡°From Cost Plus Determinations to a Network Price Cap¡±, Information Economics & Policy, 9 (2): 151-60.



This paper describes the interconnection pricing policy in the UK, explaining how from 1997 many of BT's interconnect services will be freed from price control, as the relevant markets are regarded as effectively competitive. The paper also describes the methods of incremental cost modelling employed.




Crandall, Robert, (1997), ¡°Are we Deregulating Telephone Services? Think Again¡± Policy Brief #13 -- March 1997.



Discusses the 1996 Telecommunications Act in the U.S., and the many problems arising out of trying to implement the Act.  Argues that the unbundling requirement combined with the TELRIC approach to pricing access implies new entrants can simply lease all required facilities from established carriers at prices that are below the owners¡¯ actual costs.




Crocioni, P. and C. Veljanovski, (1999), ¡°Pricing calls to mobiles: analysis of the UK Monopolies & Mergers Commission Reports on mobile termination charges¡±, Telecommunications Policy, 539-555.



Provides a useful summary of the over 1000 page report by the UK Monopolies and Mergers Commission inquiry into fixed-to-mobile termination charges, contrasting the results of that report with the report by the UK telecommunications regulator (OFTEL).  Suggests that the termination of fixed-to-mobile calls by mobile networks is subject to competitive pressures. Although these are thought to be not yet strong enough to effectively discipline termination charges, this is considered a very much temporary problem.  Thus any regulation of termination charges should be temporary. The report also identifies the problems that arise if regulators try to force retail prices to be uniform or put very specific price caps on individual services.




Crook, John, (1995), ¡°Competition and Interconnection: Successes and Challenges¡±, mimeo. International Telecommunications Society Workshop at Wellington, New Zealand



Argues that ECPR, appropriately modified, provides a useful tool for identifying an equitable contribution from competitors towards an incumbent¡¯s fixed and common costs, with the aim of promoting competition and economic efficiency.




De Fraja, G. (1999). ¡°Regulation and Access Pricing with Asymmetric Information¡±, European Economic Review, 43 (1): 109-34.



This paper analyses whether the price charged to a competitor for the use of an essential input produced in conditions of natural monopoly should only reflect considerations of relative efficiency between the various potential suppliers. A model is used to show that this is not the case. Instead, the access price should be set ¡®pro-competitively¡¯: it may be socially optimal to award production to a firm less efficient than the owner of the network. This is due to the regulator's desire to reduce the cost of her inability to observe the allocation of costs within the regulated firm.




DeGraba, Patrick. (2000). ¡°A General Theory of Interconnection¡±, mimeo. FCC



Argues that an appropriately designed Bill and Keep interconnection regime can lead to the efficient deployment of facilities by carriers and efficient utilization be end users.  All that is needed is the appropriate contracts between carriers and end users.  This interconnection approach is denoted COBAK (Central Office Bill and Keep) and it is argued it is superior to the currently used standard of TELRIC in the United States.  For this to work, it would require customers to pay directly for the various components of their calls, so that customers receiving calls would also be required to pay for the component of the cost of the service that corresponds to receiving a call.  Abstracting from the transaction costs of this approach, it still begs the question what happens for components of the network that are not competed over – e.g. the bottleneck components.




Dogonaglu, U., (1999), ¡°Network Competition and Interconnection Pricing¡±, Ph.D. thesis, SUNY, Stony Brook.



Argues, using a model of two-way interconnection, the merits of an alternative principles of interconnection regulation – access prices have to be set as a fraction of retail prices and that both firms have to agree on the same fraction.  According to the model, this leads to cost based interconnection charges, if retail price competition is sufficiently strong.  This is a variant of ECPR in which the access price automatically adjusts as retail prices adjust, and in which the rule is applied reciprocally to both firms.




Economides, N., (1998a), ¡°The Tragic Inefficiency of M-ECPR¡±, Discussion Paper EC-98-01, Stern School of Business, N.Y.U.



Argues that application of the so-called "Market Determined Efficient Component Pricing Rule," the "Efficient Component Pricing Rule," and, in general, of pricing rules that are based on private opportunity costs would perpetuate pricing inefficiencies and result in lower social surplus than pricing which is based on social opportunity cost rather than private opportunity costs.  Two other papers by the author also make similar points (Economides, N. and L. J. White, (1995), ¡°Access and Interconnection Pricing: How Efficient is the Efficient Components Pricing Rule?¡± The Antitrust Bulletin, XL (3): 557-579 and Economides, N. and L. J. White, (1996), ¡°The Inefficiency of the ECPR Yet Again: A Reply to Larson¡± The Antitrust Bulletin, XLIII (2): 429-444.)




Economides, N., (1998b), ¡°Raising Rivals' Costs in Complementary Goods Markets: LECs Entering into Long Distance and Microsoft Bundling Internet Explorer¡±.



This paper shows that a local telephone network that has a monopoly, will increase the access price charged to a rival in a downstream market (such as the long-distance market) above cost.  The regulatory implication the authors draw is that the access price set by a vertically integrated monopoly that sells in a competitive downstream market, needs to be regulated.  If not, the vertically integrated monopolist may well foreclose the downstream market.




Economides, N., G. Lopomo and G. Woroch, (1996) ¡°Regulatory Pricing Policies to Neutralize Network Dominance¡±, Industrial and Corporate Change, 5 (4): 1013-1028.



Models a symmetric two-way interconnection case, and argues for the benefits of reciprocity to avoid the dominance of a single network, when networks can set differential retail prices for on and off-net calls.  The paper also argues for imputation tests, which stops a dominant network conducting a price-squeeze, and an unbundling rule, which reduces termination charges set by a dominant network that was engaging in pure bundling.




Ergas, H., (1998), ¡°Access and Interconnection in Network Industries¡± Network Economics Consulting Group (NECG). 1998



Considers various rules for handling competition issues in respect of access and interconnection.  Provides a useful survey of the practical problems associated with regulating access, and provides some suggestions for appropriate design of institutional mechanisms for resolving access disputes.  Outlines the strengths and weaknesses of various pricing rules for interconnection such as ECPR, TSLRIC, and price-caps.




Ergas, H., (1998) ¡°TSLRIC, TELRIC and other forms of forward-looking cost models in telecommunications: a curmudgeon¡¯s guide¡±. NECG. A paper prepared for the 1998 EU Competition Workshop at the Robert Schuman Centre of the European University Institute.



Discusses the practical issues involved with setting cost-based (forward-looking) access prices.  Details the many complications that need to be taken into account in constructing a TELRIC or TSLRIC estimate.




Ergas, H., and E. K. Ralph, (1994), ¡°The Interconnection Problem with a focus on Telecommunications¡±, Communications and Strategies, IDATE, 4 (16), 9-24.

(1, 2, 5)


Discussion of interconnection issues with a focus on telecommunications. Outlines why interconnection makes sense, considers a number of proposed regulatory devices, and a number of issues specific to telecommunications and the regulation thereof. Aimed at policy makers.




Ergas, H., and E. K. Ralph, (1996), ¡°The Baumol-Willig Rule: The answer to the pricing of interconnection?¡±, Trade Practices Commission working paper, Canberra, Australia.


Demonstrates the standard ECPR in a simple environment perpetuates monopoly outcomes, if downstream prices are not regulated. Further, shows that if firms can vary quality as well as price, application of the ECPR can even reduce welfare below monopoly levels.




Frieden, R., (1999), ¡°Last Days of the Free Ride? The Consequences of Settlement-Based Interconnection for the Internet¡±, Info. 1 (3): 225-38.



Looks at interconnection from the point of view of the Internet. It argues that settlement-free peering (the Internet equivalent of ¡°Bill and Keep¡±) will not last much longer.  The article tracks the evolution of internet interconnection and charging arrangements and draws comparisons with the arrangements of telecommunication carriers. Frieden argues that settlement-free peering increasingly is leading to network congestion and under-investment in new bandwidth. The article looks at alternative models for remedying these problems and the potential consequences. In particular it focuses on how pricing arrangements could change the nature of ISP relationships, from a peer-based to a hierarchy-based system.




Gabel, D., and D. F. Weiman, (1998), Opening Networks to Competition: The Regulation and Pricing of Access.



Ten chapters that cover selected theoretical material on interconnection pricing, together with a few case studies.  The most relevant chapters are a chapter titled ¡°Historical Perspectives on Interconnection between Competing Local Operating Companies: The United States, 1894-1914¡±, which studies how interconnection was dealt with in the early period of development in the U.S. telephone industry and ¡°On the Frontier of Deregulation: New Zealand Telecommunications and the Problem of Interconnecting Competing Networks,¡± which discusses New Zealand light-handed approach to interconnection.





Galbi, Douglas. (2000), ¡°The Economics of Transforming Network Interconnection and Transport¡± Common Carrier Bureau, FCC.



Argues the challenges of establishing interconnection relationships in voice telephony are slowing development of Internet services. In particular, regulation is encouraging competitive development that depends on  complex regulatory battles for interconnection, rather than towards the creation of new facilities and services. Galbi puts forward an alternative approach in which competing and independently owned, service interconnection points (SIPs) would stimulate development of local facilities.




Haring, J. and J. Rohlfs, (1997) ¡°Efficient competition in local telecommunications without excessive regulation¡± Information Economics and Policy, 9: 119-131.



Argues the excessive regulation that is required to bring in competition to local US telephony markets, can be replaced by a simple access pricing regime which only requires that call-termination charges be set on a symmetric basis (as described below), while allowing entrants (although not necessarily the incumbent) substantial retail pricing flexibility. Thus, according to this approach, the incumbent should be allowed to choose the price it charges an entrant for access to its network as well as the price it pays the entrant for access to the entrant¡¯s network, under the restriction that both prices be equal (symmetric). It is argued, given this restriction, the incumbent does not have an incentive to raise the access price above cost for the relevant network service, since this will lead to entrants that generate cash flow from the incumbent simply by specialising in the termination of calls. Likewise, the incumbent will not have an incentive to set the access price below cost for the relevant network service, since this will lead to entrants that generate cash flow from the incumbent simply by specialising in the origination of calls.  Essentially, any pricing away from cost by the incumbent will lead to profitable opportunities for arbitrage by competitors. Thus, according to this approach, the regulator does not need to know detailed cost information.  However, the author recognises some difficulties that arise from this approach, since with the same access price for different types of network services, arbitrage opportunities will arise.






Harris, Robert and C.Jeffrey Kraft, (1997), ¡°Meddling Through: Regulating Local Telephone Competition in the United States¡± Journal of Economic Perspectives 11 (4): 93-112.



Discusses the implementation of the U.S. Telecommunication Act of 1996. Describes the U.S. regulatory framework as one of extensively regulated competition.  Discusses the use of TELRIC (total element long run incremental cost) cost methodology in the context of the U.S. Telecommunication¡¯s Act.




Kahn, A., Tardiff, T., and D. Weisman, (1999), ¡°The Telecommunications Act at three years: an economic evaluation of its implementation by the Federal Communications Commission,¡± Information Economics And Policy 11 (4): 319-365



Provides an up-to-date discussion of the success, or otherwise, of the U.S. Telecommunications Act.  The Act calls for two forms of regulatory-assisted entry into local markets: lease of network elements and resale. The FCC has decreed that the charges for those elements and the resale discounts must emulate the costs of an ideally efficient firm. The paper argues that this standard is in fact not efficient, and the FCC's attempt to jump-start the entry of competitors in this way has short-circuited the competitive process itself and jeopardized achievement of the goals of the Act.




Kaserman, D. and J. Mayo, (1997), ¡°An Efficient Avoided Cost Pricing Rule for Resale of Local Exchange Telephone  Services¡± Journal of Regulatory Economics 11 (1): 91-107.



Provides an analysis of pricing access to wholesale reselling of retail services. Under the U.S. Telecommunications Act of 1996, incumbent local exchange telephone companies must make their retail telecommunications services available to resellers at a discount that reflects the costs that will be avoided by providing the services at wholesale rather than at retail. In this article, the authors develop a pricing methodology to apply to such wholesale services. The methodology, which is labelled the ¡°Avoided Cost Pricing Rule¡±, is designed to generate an economically efficient discount that may be applied to the incumbent local exchange carriers' services that are offered to resellers.



Kennet, D. M. and W. W. Sharkey (2000), ¡°Proposals for Interconnection Pricing in Peru.¡±  Mimeo, The World Bank.



Suggests two cost-based local network interconnection proposals to be adopted to mediate a pricing dispute between the Peruvian dominant carrier and two entrants.  In the simpler of the two proposals, an existing engineering model of the Peruvian network is used to calculate average TS costs throughout Peru.  This revenue flow to the incumbent is augmented by a fixed per-line charge determined by a benchmark calculation designed to be revenue-neutral with respect to the much higher usage charges discussed in the negotiations.  The fixed charge is in turn modified by the relative ubiquitousness of the networks:  A network is charged a fixed fee inversely proportional to the difference between the number of districts within each departamento that it serves and the number served by the incumbent, creating as a side benefit a small incentive for rural buildout.  The second proposal, meant to be implemented over a longer term, would use a more purely cost-based  approach, giving time to both the regulator and the companies to assess the validity of the model used.  While the paper is written specifically for the Peruvian case, the principles should be applicable in any developing country with low telephone penetration.



King, S., (1997), ¡°Access Pricing under Rate-of-Return Regulation¡±, Australian Economic Review, 30 (3): 243-55.



The paper models rate-of-return regulation of access prices, showing that the optimal access prices will depend on the degree of downstream competition. With imperfect price competition and fixed numbers of firms downstream, optimal access prices will ¡®mimic¡¯ downstream competition and reduce downstream profits. With free entry downstream, optimal access pricing should determine an optimal level of downstream participation.




King, S. and R. Maddock, (1999), ¡°Light-handed regulation of access in Australia: negotiation with arbitration¡±, Information Economics and Policy 11: 1-22.



The paper models a bargaining game between access providers and access seekers, when there is an outside regulatory option.  The authors show that the regulator can substantially influence the range of bargaining outcomes through both the expected outcome of arbitration and its interpretation of regulatory procedures. The paper focuses on the Australian regime, but provides a template to understand other so-called ¡°light-handed¡± regimes.




Knieps, G., (1997), ¡°Costing and Pricing of Interconnection Services in a Liberalized European Telecommunications Market¡±, Discussion Paper No. 39, Institut Fur Verkehrswissenschaft und Regionalpolitik Albert-Ludwigs-Universitat Freiburg.



Critically details the EU¡¯s approach to interconnection. Argues that the EU approach is flawed because it puts too many restrictions on the firm¡¯s pricing decisions and is inconsistent with economic efficiency.  The paper also provides a comparison of top-down versus bottom-up approaches to determining long-run incremental costs (LRIC).  Knieps concludes that engineering-economic models are inadequate for calculating decision relevant incremental costs for the established carriers. Outlines the various sources which lead to an underestimation of the real incremental costs.  Recommends top-down approaches based on cost accounting systems of established carriers. The paper evaluates the EU¡¯s ¡°best practice¡± approach.




Laffont, J.-J. and J. Tirole, (1994), ¡°Access Pricing and Competition¡±, European Economic Review, 38: 1673-1710.



Provides a survey of the role access prices play in achieving competition in different network industries (electricity, telecommunications, railways), where the network can be described as a natural monopoly.  The authors derive access pricing formulas in a model of optimal regulation under incomplete information. Their framework takes into account such things as fixed costs of the network, bypass, and incentives for firms to provide misleading information to regulators. 




Laffont, J.-J., and J. Tirole, (1996), ¡°Creating competition through interconnection: Theory and Practice¡±, Journal of Regulatory Economics, 10 (3): 227-56.



Suggests access/interconnection prices should be considered as part of a global price-cap.  The paper argues that setting access prices at the ECPR level, combined with a global price-cap, is the optimal form of regulation. The ECPR regulation prevents the incumbent carrying out a price-squeeze, since it ensures access prices are sufficiently below retail prices to allow efficient competitors to compete. A global price cap allows the incumbent to set access and retail prices according to an optimal Ramsey price structure, where the incumbent views its competitors¡¯ output as an output of its own.  However, the process of achieving Ramsey pricing is complex, likely to be extremely lengthy and requires reiterated regulatory decisions on prices, costs and incentives, making the mechanism vulnerable to manipulation.




Laffont, J.-J, and J. Tirole, (2000) ¡°Competition in Telecommunications¡±, MIT Press.



A textbook written in a style accessible to policy makers. Laffont and Tirole analyze regulatory reform and the emergence of competition in network industries using the latest theoretical tools of industrial organization, political economy, and the economics of incentives.  Chapters include: the introduction of incentive regulation; one-way access problems; the special nature of competition in an industry requiring two-way access; and universal service. The book concludes with a discussion of the Internet and regulatory institutions.




Laffont, J.-J, P. Rey, and J. Tirole, (1997), ¡°Competition between Telecommunications Operators¡±, European Economic Review, 41 (3-5): 701-11.



The paper analyses unregulated network competition, both in the mature and the transition phase of the industry. It considers competition in linear and nonlinear prices for customers, and also considers both nondiscriminatory pricing and termination-based discriminatory pricing. The paper investigates whether freely negotiated access charges are compatible with effective competition in the mature phase of the industry, and whether access charges can be used to erect barriers to entry in the transition towards competition. Last, it analyzes the meaning and impact of the Efficient Component Pricing Rule in the context of network competition.



Laffont, J.-J, P. Rey, and J. Tirole, (1998a), ¡°Network Competition: I. Overview and Nondiscriminatory Pricing¡±, Rand Journal of Economics, 29: 1-37.



Provides a generic treatment of two-way interconnection.  Shows that with linear retail pricing, interconnection prices can be used for collusion between two networks that are interconnected – that is, firms can increase per-minute interconnection prices to achieve per-minute retail prices that are at monopoly levels. Shows that this result no longer holds when retail pricing is non-linear. Explains what ECPR means in the context of two-way interconnection cases.




Laffont, J.-J, P. Rey, and J. Tirole, (1998b). ¡°Network Competition: II. Price discrimination¡±, Rand Journal of Economics, 29: 38-56.



Provides a generic treatment of two-way interconnection when firms are allowed to set differential prices, depending on whether a call terminates on one¡¯s own network or on a rival network.  In this case, network effects can still arise, even with network interconnection.  Despite this, the paper shows that such price discrimination can be welfare enhancing. 




Lapuerta, C. and W. Tye, (1999), ¡°Promoting Effective Competition through Interconnection Policy¡±, Telecommunications Policy, 23 (2): 129-45.



Argues, using a dynamic analysis, that the parity principle (that access prices be regulated equal to those that an incumbent would accept voluntarily), is largely inappropriate for setting interconnection prices in most current contexts.  The cases of the US and New Zealand are analyzed.  It is argued that interconnection charges are best set by legal or regulatory authority based on the costs of providing network access.




Lewis, Tracy and David Sappington, (1999) ¡°Access pricing with unregulated downstream competition,¡± Information Economics and Policy 11(1):73-100



Analyses optimal access pricing when downstream competition is unregulated but imperfect, and when the regulator is uncertain about the production costs of an unregulated competitor.  The analysis shows: (1) the regulator optimally sets access prices so as to tilt the playing field in the direction of the more efficient producer, rather than level the playing field as is often advocated in policy debates; (2) the optimal degree of regulatory intervention declines as downstream competition becomes more pronounced; and (3) the regulator optimally reveals to the incumbent supplier any information that arrives about the competitor's production costs.




Maher, M (1999) ¡°Access costs and entry in the local telecommunications network: a case for de-averaged rates¡± International Journal of Industrial Organization, 17: 593-609



Estimates cost functions of access costs at the local level for the U.S.  The study shows there are economies of scale in the provision of access to the local network and that these costs differ by geographic location.  The study suggests de-averaged cost-based rates (i.e. different rates in different areas) would not be prohibitively high and would not threaten universal service objectives.  However, the authors do argue that pro-competitive regulatory controls are still required to restrain dominant carriers. 




Michie, J., (1998). ¡°Competition Aspects of Pricing Access to Networks: What Are the Issues?¡±, Telecommunications Policy, 22 (6): 467-70.



The author provides a commentary discussing the questions of access pricing and the issues involved in actually deriving a price fixing mechanism for access to a network.




Neumann, K-H., J Arnbak, B. Mitchell, W. Neu. And I. Vogelsang. 1994 ¡°Network Interconnection in the Domain of ONP: Study for DG XIII of the European Commission¡± Final Report, wIK, Bad Honnef, Germany.



A report undertaken for the European Commission.  It reviews interconnection pricing theories and practical applications. It concludes in favour of cost-based access prices (such as average incremental cost, with a mark-up for legitimate revenue requirements along the lines of the Ramsey approach).




Office of Utilities Regulation (1999) ¡°Interconnection in Telecommunications, A Consultative Document¡±.



A consultative document developed for the government of Jamaica on a wide range of issues pertaining to interconnection in telephony.  The annex provides an international comparison of interconnection charges for 16 countries (including the APEC countries: Australia, Hong Kong, Japan, and the USA). The recommended best practice benchmark interconnection charge ranges derived by the European Commission are also included. The report contains a spreadsheet with the data.




Ovum, ¡°Ovum interconnect¡±



A quarterly report that can be purchased from OVUM. It provides an analysis of interconnect charges for fixed-line and mobile services, focusing on Austria, Belgium, Canada, Chile, Denmark, Finland, France, Germany, Hong Kong, Ireland, Italy, Japan, Mexico, Netherlands, Norway, Spain, Sweden, Switzerland, UK, and a sample of operators in the USA. 



Ovum, (June 1999) ¡°Implementing Cost-based Interconnect¡±



A one-off report that can be purchased from OVUM.  It provides a comprehensive study on implementing LRIC-based interconnection based on reform of an operator¡¯s accounting systems, on network models and on international benchmarks. 




Productivity Commission, 1999, ¡°International Benchmarking of Telecommunications Prices and Price Changes¡±, Research Report, AusInfo, Canberra, Australia.



Compares residential and business retail prices of various services across selected OECD countries. Provides a very brief summary of regulatory arrangements in different countries (as at June 1999).  In an annex it includes a more detailed survey of regulatory and institutional arrangements in Korea, Malaysia, and Singapore




Ralph, E. K., (1996) Regulating an Input Monopolist with Minimal Information: A local cap (, drawn from Ralph, E. K., (1996), ¡°Regulating an Input Monopolist with a Focus on Interconnection in Telecommunications¡±, Ph.D. thesis, Duke University.

(1, 5)


The rule Ralph develops regulates a vertically integrated access monopolist by allowing it to freely set prices on local access, local calls, and per minute interconnection fees, but the revenues earned on these, including revenues imputed to the monopolist for interconnecting to itself, cannot exceed a regulated cap. Implementing the mechanism requires comparatively little information as compared with other interconnection devices. The mechanism¡¯s efficacy is demonstrated for the case where downstream calls are priced at a constant per minute rate.




Salinger, M., (1998), ¡°Regulating Prices to Equal Forward-Looking Costs: Cost-Based Prices or Price-Based Costs?¡± Journal of Regulatory Economics, 14: 149-164.



This paper discusses how forward-looking costs can be defined and explains the role of depreciation and utilization in their calculation.  It shows the potential for competition, asset life uncertainty, and the installation of excess capacity in anticipation of demand growth to raise forward-looking costs. The potential for technological change that enhances the future value of an asset lowers forward-looking cost.




Schechter, P. (1996), ¡°Customer Ownership of the Local Loop: A Solution to the Problem of Interconnection¡±, Telecommunications Policy, 20 (8): 573-84.



Suggests the solution to the ¡°problem of interconnection¡± is for customers to own their own local loop and decide which networks and at what price these firms will have access to their part of the local loop.  There are contracting costs that need to be considered in evaluating such a proposal.  Moreover, the difficult issue of how consumers gain ownership of their part of the local loop, is something that needs to be further addressed.  



Sibley, David and Dennis Weisman, (1998), ¡°Raising rivals' costs: The entry of an upstream monopolist into downstream markets,¡± Information Economics and Policy 10(4):451-469



A regulated upstream monopolist provides an input to firms in a downstream market. If the monopolist enters the downstream market, a natural concern is that it will act so as to raise its downstream rivals¡¯ costs. An offsetting incentive is that a higher downstream price will reduce demand for the input, which reduces the monopolist's profit. Thus, the paper revisits Economides¡¯ (1998b) paper that showed that upstream monopolists face an incentive to raise rivals costs. Conditions under which one incentive dominates the other are derived.  Taking into account both incentives, the authors argue that the monopolist may desire to lower its downstream rivals¡¯ costs rather than raise them.  They conclude that that regulatory policy towards such downstream entry should not focus exclusively on the ability of the upstream firm to discriminate.




Spiller, P. and C. Cardilli, (1997), ¡°The Frontier of Telecommunications Deregulation: Small Countries Leading the Pack¡±, Journal of Economic Perspectives, 11 (4): 127-38.



Studies the performance and the problems of four small countries which have been on the forefront of telecommunications deregulation – Australia, Chile, Guatemala, and New Zealand.  Provides a comparison of the approaches to interconnection, equal access, unbundling and industry structure.




Valletti, Tommaso, (1998) ¡°Two-part access pricing and imperfect competition,¡± Information Economics And Policy 10 (3):305-323



Provides a model of a vertically separated industry with an upstream monopolist who supplies an essential input to two downstream Cournot firms. The model is relevant to the case in the telecommunications industry where trunk operators must have access to the local network of an incumbent firm to provide their long-distance service. The paper analyses two-part access pricing and input price discrimination under different regulatory settings, and it finds that discrimination may produce adverse welfare effects

when it is practised by the unregulated upstream firm.




Wright, J., (1999), ¡°International Telecommunications, Settlement Rates, and the FCC¡±, Journal of Regulatory Economics, 15: 267-291.



Models how international telecommunication operators set settlement rates.  Provides empirical estimates of the determinants of these rates for all countries versus the United States.  The analysis suggests that less developed countries will tend to have net incoming calls, and as a result will tend to negotiate high settlement rates that raise the price of international phone calls above the efficient level.  Suggests that introducing facilities based competition to international call services should alleviate such problems. 




Wright, J., (2000), ¡°Competition and Termination in Cellular Networks¡± Centre for Research in Network Economics and Communications, Research Paper, University of Auckland.



Models how mobile termination rates are set.  Higher mobile termination charges encourage lower mobile retail pricing, as mobile firms compete more aggressively to capture the lucrative termination revenue. Thus it argues that the optimal mobile termination charges are likely to be above incremental cost, due to the positive externality that arises from increased mobile penetration with lower mobile retail pricing.  This externality arises from the benefits received from other mobile customers and fixed-line customers when a new mobile customer joins the mobile network. 







In this Report, we provide an overview very briefly explaining some of the political and technical issues surrounding the setting of interconnection policy by APEC member economies.  We survey an extensive literature base on the subject, and classify each entry according to principles set forth by APEC – TEL.


Following the text of this report, we have included an Excel workbook containing the results of a survey questionnaire submitted to APEC member economy representatives concerning the status of interconnection policy in their respective home countries.  As can be readily seen, there is a wide variety of policy approaches to this vitally important issue, and apparently some desire on the part of members to understand others¡¯ approaches.


In the next set of tasks, we hope to be informed by representatives of the APEC economies regarding the special needs of each economy in addressing the issues in interconnection pricing.  Upon completing these discussions, we will begin the process of helping to create recommended policies for general agreement among the economies.


APPENDIX:  Survey Questionnaire




1.  What is the average per‑minute tariff for interconnection for companies wishing to provide service between your economy and another?


2.  What is the typical tariff structure?

           (a)  per‑minute charge only

           (b)  fixed monthly charge plus per‑minute charge (two‑part tariff)

           (c)  fixed monthly charge only (flat rate)

           (d)  Answer (a) plus a call set‑up charge

           (e)  Answer (b) plus a call set‑up charge

           (f)  Answer (c) plus a call set‑up charge

           (g)  Other ‑ please specify


3.  If your answer to Question 2 is not (a), please provide a typical interconnection tariff.


4.  How are interconnection rates determined?

(a)  Pure negotiation; firms negotiate without any government  interference

(b)  Negotiation, but with government intervention after  notification

         (i)   Government chooses between final offers

         (ii)  Government arbitrates/mediates between offers, no  appeal possible

         (iii) Government arbitrates/mediates between offers, but  parties may appeal

(c)  Government sets tariffs

(d)  State‑owned monopolist sets tariffs

(e)  Other ‑ please specify


5.  What is the number of service providers in your economy by type?

           (a)  Fixed‑wire local only

           (b)  Fixed‑wire, long distance domestic only

           (c)  Fixed‑wire, long distance international only

           (d)  Fixed‑wire, multiple services

           (e)  Mobile, excluding satellite

           (f)  Satellite

           (g)  Beeper

           (h)  Other


6.  Has the teledensity of your economy changed significantly since 1998?


* Network Economics Consulting Group Pty, Ltd.

** George Washington University and LECOM Associates, Inc.

[1]APEC membership:  Australia; Brunei Darussalam; Canada; Chile; People's Republic of China; Hong Kong, China; Indonesia; Japan; Republic of Korea; Malaysia; Mexico; New Zealand; Papua New Guinea; Peru; Philippines; Russia; Singapore; Chinese Taipei; Thailand; USA; Viet Nam.

[2] Strictly speaking, even these costs cannot be attributed to the customer at the terminal end of the access line. Other subscribers benefit both by being able to call this subscriber and by being able to be reached by this subscriber. In some instances, connecting the subscriber is not optimal except if these demands are taken into account. In no case where such demands are positive can it be unambiguously said that the subscriber at the end of the access line is the ¡°cost-causer¡±. Despite this, as a general rule, it probably is the case that the customer at the end of the subscriber line has a relatively high demand for the access line as compared with the sum of all other customers¡¯ demands for that line. This explains why such customers are usually identified as the ¡°cost-causer¡±.   The same can be said of the TS costs associated with a particular type of call—it is the call originator who typically gains the most from the call, and as a result the originator is typically identified as causing the cost. See for example, Henriet and Moulin, H (1996), ¡°Traffic-based cost allocation in a network,¡± Rand Journal Of Economics, 27 (2).

[3] See, for example, Commission Recommendation 98/322/EC (European Commission, 1998); or FCC Order 96-325 (CC Docket 96-98 and 95-185), "In the Matter of Implementation of Local Competition Provisions in the Telecommunications Act of 1996", United States Federal Communications Commission, 1996.


[4] The literature is not entirely clear on what a reasonable return for such investment is, but typically there is some recognition of the high degree of risk involved with large telecommunication¡¯s infrastructure investment. There is growing recognition, that facing sunk project costs, uncertain future demand, and technologies which might rapidly be made obsolete, there is a need for a larger return on capital to ensure dynamic efficiency.  These factors also suggest standard contestability theory should not be relied on to tell us what the efficient measure of costs is.  Salinger, M., (1998), ¡°Regulating Prices to Equal Forward-Looking Costs: Cost-Based Prices or Price-Based Costs?¡± Journal of Regulatory Economics, 14: 149-164 discusses several reasons why the normal return on investment needs to be adjusted upwards.

[5] Benitez, D.A., M. Celani, O.O. Chisari, M.A. Rodríguez Pardina and C.A. Ruzzier (1999), Minimizing the Costs of Universal Service Obligations in Argentina¡¯s Telecoms Sector Through a Cost Model, XII World Congress of the International Economic Association, August 1999, Buenos Aires, Argentina.

[6] See Oftel (1995, 1997).


[7] For the case of BT (British Telecom), the scope of the efficiency adjustment was taken from the analysis and the assumptions used by Oftel in the tariff revision for BT.


[8] Financial models and engineering-economic models should not be seen as rival approaches, but rather as complementary methodologies. For example, the avoidable costs mentioned above could be estimated through a cost proxy model.

    [9]William J. Baumol, ¡°Some Subtle Issues in Railroad Deregulation,¡± 10 Int'l J. Trans. Econ. 341 (1983); William J. Baumol and Gregory Sidak, Toward Competition in Local Telephony (1994); William J. Baumol and Gregory Sidak, ¡°The Pricing of Inputs Sold to Competitors,¡± 11 Yale J. on Reg. 171 (1994).

    [10]Baumol and Sidak, 11 Yale J. on Reg. at 178.

[11] Armstrong, Doyle and Vickers, 1995, Economides and White, 1995; Laffont and Tirole, 1994; Laffont and Tirole, 1995

[12]Armstrong and Doyle, 1994; Armstrong and Vickers, 1995 

[13] Economides and White, 1995.

[14] Laffont and Tirole, 1994.

[15] Ergas and Ralph, 1996.