Wholesale Competition in the International Telecommunications System

Networks and Spatial Economics, 5: (2005) 261–277©C 2005 Springer Science + Business Media, Inc. Manufactured in The Netherlands.

Wholesale Competition in the InternationalTelecommunications System

LIVIO CRICELLIDipartimento di Meccanica, Strutture, Ambiente e Territorio, Universita degli Studi di Cassino,Via G. Di Biasio 43 - 03043 Cassino (FR)email: [email protected]

FRANCESCA DI PILLODipartimento di Ingegneria dell’Impresa, Universita degli Studi di Roma “Tor Vergata”,Via del Politecnico 1 - 00133 Romaemail: [email protected]

MASSIMO GASTALDIDipartimento di Ingegneria Elettrica, Universita degli Studi di L’Aquila, Monteluco di Roio - 67100 L’Aquilaemail: [email protected]

NATHAN LEVIALDIDipartimento di Ingegneria dell’Impresa, Universita degli Studi di Roma “Tor Vergata”,Via del Politecnico 1 - 00133 Romaemail: [email protected]

Abstract

In the last ten years, the international telecommunications industry has undergone important changes from both thelegislative (regulatory reforms) and the technological points of view. The liberalization of the telecommunicationsmarket has, consequently, affected the dynamics of the bilateral agreements between international carriers. Inthis paper, the international wholesale market in which carriers compete over the tariffs in order to increase theirmarket share is analyzed. In particular, this paper examines the interactions between fixed telephony TLC carriersbelonging to various countries and analyzes the possible strategic choices of the market leader induced by changesin competitor behavior.

Keywords: two-way networks, international telecommunications, foreclosure strategy, interconnection tariff

1. Introduction

The international telephony industry is characterized by the need for interconnection be-tween carriers. Every international telephony carrier must conclude interconnection agree-ments in order to terminate calls to the destination country. Termination on the destinationnetwork represents a carrier cost, known as the interconnection or access tariff. The needfor interconnection between countries creates problems since they are characterized bydifferent market contexts, legislative restrictions, and catchment areas dimensions. These

262 CRICELLI ET AL.

differences have a reciprocal impact: changes in price level in one country, deriving fromeither the strategic choices of an individual carrier or from the implementation of economicpolicy, cause a large number of effects for the carriers in an interconnected country. TheFederal Communications Commission in the United States (FCC, 1996), in its directive oninternational interconnection tariffs, sets the interconnection tariff values payable by U.S.carriers to foreign carriers for outgoing traffic from the USA so as to bar the phenomenon of“whipsawing”; that is to say, the practice of applying a relatively high access tariff adoptedby foreign monopolies which, consequently, interferes with the competitive dynamics inthe American market. In fact, only large operators managed to sustain the interconnectioncost, causing the small carriers to be excluded from the market.

In the “two way” networks (Armstrong, 1998; Laffont et al., 1998; Cricelli et al., 1999),which characterize the international telecommunications market, reciprocal interconnectioncan have negative effects on social welfare. The operators can exploit the interconnectiontariff as an instrument of collusion in cases where the operators offer services with a degreeof differentiation so that they maintain a higher price level for the end user than wouldbe the case in a competitive scenario. In this case, in order to maximize social welfare,intervention by the Authority regulating the industry is necessary. On two way networksin asymmetrical conditions, Carter and Wright (1999) examined the threat of collusionsince the incumbent carriers possess the competitive advantage of brand loyalty towardsother competitors. The asymmetry, besides being relative to the size of carriers in the samemarket, can be applied to the size of interconnected carriers in different countries. Moreover,it is important to analyze the interactions between different carriers not in relation to thepossibility of collusion but with regard to a competitive scenario characterized by a greaterdegree of competition, similar to the actual context. Madden and Savage (2000) carried outsuch an analysis. They developed an econometric model to analyze the interactions betweenpricing choices and the market’s structure. Moreover, they demonstrated how a relationshipexists between reduction in market share of the larger carriers in the United States and thedecrease in international call price applied by foreign carriers.

Furthermore, the results obtained revealed the existence of a relationship between in-creased competition in foreign countries and reduced price levels not only from foreigncarriers, but also from U.S. carriers.

The aim of this paper is to investigate the economic interaction between carriers belongingto different countries and the possible impact that their interaction causes on the compet-itive scenarios in each country. With this aim in mind, the competitive situation has beenconsidered and the existing relationships between the strategic choices made by carriers intwo interconnected countries have been analyzed. Since the competition takes place bothin intermediate market and in retail market, the interconnection tariff and the price to theend user are taken into consideration as strategic decisional variables. With regard to bothmarkets, it is hypothesized that the operators may follow two distinct strategies: foreclosurestrategy and price competition. The results obtained from the simulation process are usedto demonstrate the economic advantage of implementing proposed strategies according tothe relative sizes of the carriers. This analysis assumes great importance because certainstrategic choices can place the competitiveness of the market in danger, which then requireseffective legislative intervention.

WHOLESALE COMPETITION IN THE INTL TELECOMMUNICATIONS SYSTEM 263

This paper is organized as follows. In the next section, an analysis of the competitivescenario is outlined. In Section 3, a model of the competitive context is proposed in which alarge number of carriers belonging to different countries operate. In Section 4, the range ofpossible strategic choices the carriers can make are examined by using a simulation process.Lastly, the conclusions drawn are presented in the final section.

2. The competitive scenario

Since the liberalization process began in the telecommunications industries of industrializedNations, the volume of international telephone traffic has experienced an average annualincrease of 15% (Telegeography, 2004), except between 2000–2002 (6%). In 2002, inter-national telephone calls throughout the world totaled 155 billion minutes. If Voice-over-IP(VoIP) were included (18.7 billions minutes), aggregate growth would came to approxi-mately 11.3%, slightly lower than growth in 2001. This growth has also been caused by theprogressive reduction in prices. For example, from the opening up of European carriers tocompetition, prices have decreased by approximately 40%. Price declines have outpacedtraffic growth, causing retail revenues from international traffic to decline from $72 billionin 2000 to $53 billion in 2002. Moreover, since 1998, many additional countries have joinedthe ranks of competitive markets, including Greece, Taiwan, China, and most recently, In-dia. Given their lower cost structures, new market entrants used low prices to pry consumersaway from the former incumbent monopoly carriers. Consequently, new carriers were ableto capture a substantial share of the market within two years of liberalization. Between1990 and 2002, new international carriers grew from 0.03 to 32% of the total internationalmarket. Data on PSTN International Traffic are summarized in Table 1. While PSTN trafficgrowth fell during the previous two years, VoIP has continued to surge, growing over 80%to 18.7 billion minutes of international voice traffic in 2002. The principal destinations forinternational VoIP traffic remain Latin America, Asia and Eastern Europe, with deregulat-ing markets, emerging competition, and relatively high settlements rates, underscoring thecontinued role of settlement rate arbitrage as the primary growth driver. While the desti-nations for VoIP traffic may not have registered much change over the previous years, thefield of players certainly has.

Table 1. Twelve years of changes in International Traffic (Telegeography, 2004).

Indicator 1990 1995 2000 2002

International Traffic (Billions of minutes) 33.5 61.6 132.7 155.2

Revenue from International Traffic (Billions of $US) 37 55 70 53

Countries allowing carrier competition 6 18 49 >50

International Traffic from competitive market (%) n.a. 49 87 90

Top 20 carriers’ share of world traffic (%) 86 72 50 48

Market share of new carriers (%) <1 5 31 32

264 CRICELLI ET AL.

One of the most surprising dynamics to emerge has been increased incumbent partici-pation in the VoIP market (Telegeography, 2004). Established traditional switched carriersare now beginning to compete directly with their VoIP counterparts. Nowadays, Teleglobeis the third largest carrier of international voice traffic transporting more than 11 billionsminutes per years using only VoIP technology (Telegeography, 2004). Moreover, in thelast ten years market forces and technological innovations have driven down prices andincreased traffic flows, stimulating competition. In this scenario, with the advent of Internettelephony, the idea of VoIP has no doubt played a relevant role in accelerating this process.In fact, internationally, in July 2002, 4,726 carriers had licenses to operate in the interna-tional telephony market. It is noteworthy that there were a few less than 600 carriers in theinternational telephony industry only as long ago as 1997. In the United States, the levelof competition in the industry has grown exponentially in recent years with over 1,800international telephony carriers compared to 175 in 1997. The European competitive sce-nario is characterized by a regime of competition in which the former monopolist, almostalways, retain the leadership. In particular, Italy has 120 international telephony operatorsin various market segments, with the ex monopolist Telecom Italia having 55.5% of thetotal international telephone call minutes in 2002.

The liberalization of the industry has induced a change from bilateral agreements be-tween national monopolies to agreements based on competitive criteria. The actual practiceof interconnection depends on a series of factors which produce notable variation in theinterconnection tariff. The cost of interconnection depends on various elements:

– the type of place and the characteristics of the destination (e.g. large/small city, ruralarea, etc.);

– the typology of the destination (e.g. fixed/mobile network);– the quality of the transmission.

For example, the carriers support a higher cost to terminate a call in a developing countrywhere they are obliged to negotiate with a local monopoly which has the greater bargainingpower. Furthermore, with the aim of reducing transmission costs, many carriers adopt prac-tices such as call back or refile which bypass the traditional system of rate making. Thesealternative transmission methods, even though costing less, suffer the disadvantage of infe-rior quality. The legislative framework in Europe is composed of European Union legislationprotecting the possibility of operators negotiating “transfrontier agreements” between themfor access and interconnection on a “commercial basis”, or in effect, negotiations are notsubject to legislative restrictions (EC, 2002).

The determination of the interconnection tariff and the price for the end user therefore rep-resent a crucial factor in a complex competitive context such as international telephony. Infact, the competition in this industry is not only based on attracting the national subscribersfor international telephone calls, but also on the determination of the interconnection tarifffor the termination of calls made in foreign countries. The determination of the intercon-nection tariff assumes greater complexity because they are fixed in a reciprocal manner bythe operators of the two countries who wish to connect. Therefore, the competition is atwo stage process: firstly, competing on a national level in order to have the greater number


of subscribers for international telephone services, and secondly, competing on an inter-national level in order to make more favorable agreements with foreign operators for thetermination of calls.

3. The model

Two fixed telephony national markets are considered, namely market A and market B inwhich n and m competitors respectively operate, as shown in figure 1. An asymmetricalscheme is assumed for both markets where carrier 1 is the incumbent while the other n − 1(or m − 1) carriers are the followers. The retail price of the international call is paid bythe subscriber who makes the call, while the subscriber who receives the call supportsno cost at all. Each carrier is able to operate in both an intermediary market in whichinternational connections take place, as well as in the retail market, offering full servicecoverage. Competition among carriers therefore takes place on two levels: firstly, in theretail market where the decisional variable is the price to the end user, and secondly, in theintermediate market where the decisional variable is the interconnection tariff. In order tomodel this competitive scenario, the model of Carter and Wright (1999) is considered forthe case with two carriers.

Figure 1. The competitive scenario.

266 CRICELLI ET AL.

Figure 2. Top ten international telephony operators (Source: Telegeography, 2004).

Market asymmetry is necessary because it represents the competitive reality. As illustratedin figure 2, in many of the industrialized countries the leadership of the competitive marketsis taken by the ex-monopolists. These retain a strong competitive advantage which theyobtained during the years in which they were the only carrier in the market. Therefore, the exmonopolist subscribers have greater brand loyalty and switching costs. Nowadays, in highlycompetitive environments, concentrating on improved consumer loyalty to brand permitscarriers to maintain a dominant and lasting position in the international telecommunicationsscenario. The utility function for a consumer subscribing to a generic operator j is given by

U = u(q j ) + θ j + v0

where, q j is the volume of calls; u(q j ) is the utility arising from the consumption of quantityq j ; θ j is the additional benefit derived from the connection with operator j ; v0 is the fixedsurplus caused by the use of the interconnection service.

Consumers are uniformly distributed on the interval [0, 1] with a value x , while thecarriers are located at 0 and 1 + β, where β measures the degree of asymmetry betweenthe competitors and represents the extra benefits that an entrant must offer to persuadeconsumers to switch from the incumbent. Therefore, β can be considered as a measure ofbrand loyalty. When β > 0, carrier 1 has a greater market share than carrier 2 when pricesare equal. Thus, when β > 0, carrier 1 is considered as the incumbent and carrier 2 is theentrant. There is not brand loyalty when β = 0 and, if carriers price are equal, carrier 1 willhave half of the market. Moreover, such brand loyalty can represent the superior servicesprovided by the incumbent. By subscribing to carrier 1 and 2, a consumer with value x


receives extra benefits given by the following relationships:

θ1 = β

2σ+ 1 − x

2σand θ2 = x

2σ

The variable σ represents the degree of substitutability between the carriers. For lowvalues of σ , a carrier is able to price higher than its competitor, holding all its market share.As σ tends to infinity, θ becomes an irrelevant variable with respect to the utility function.Consumers will choose to connect to a carrier only as regards to price and the carrier withthe lowest prices captures the whole market. Considering a two part tariff (a per unit pricep j and the constant fee in order to be connected g j ), then gross consumer surplus is:

Z (p j ) = maxq

{u(q j ) − p j q j }

and the net consumer surplus obtained from connection to carrier j is:

w j = Z (p j ) − g j

A subscriber localized at point x will be indifferent between the two networks if receivedextra benefits are equal (Carter and Wright, 1999):

w1 + β

2σ+ 1 − x

2σ= w2 + x

2σ

and since S1 + S2 = 1, the following is obtained:

w1 + β

2σ+ 1 − S1

2σ= w2 + 1 − S2

2σ

Carter and Wright model is an extension of the simple Hotelling model of product dif-ferentiation incorporating brand loyalty. Our aim is to extend this model to n operators.The larger part of the literature concerning Hotelling model concentrates on duopoly mar-kets only. The question of how the number of companies affects the equilibrium outcome islargely ignored. There are exceptions in de Palma et al. (1987) who treat the three firm case ina probabilistic framework, and Lancaster (1979), Salop (1979), Novshek (1980), and Econo-mides (1989, 1993) who do not restrict the number of firms. In the extension proposed in thispaper, the market share for each carrier is calculated in order to represent the real competi-tive scenario of international telecommunications. Therefore, analysis of the mathematicaland economic characteristics of market share distribution among carriers arising from thisextension is now necessary. Assuming the existence of n and m carriers in market A and Brespectively, the equilibrium market shares of the carriers of market A can be determinedfrom the following system of equations representing the indifferent position of subscribers:

w1 + β

2σ+ 1 − S A

1

2σ= w2 + 1 − S A

2

2σ

w2 + 1 − S A2

2σ= w3 + 1 − S A

3

2σ′ (1)′′

268 CRICELLI ET AL.

wn−1 + 1 − S An−1

2σ= wn + 1 − S A

n

2σ

subject to:n∑

j=1

S Aj = 1

By solving system (1), the market shares in A are given by:

S A1 = 1 + (n − 1)β

n+ 2σ

n

{nw1 −

n∑j=1

w j

}

and

S Aj = 1 − β

n+ 2σ

n

{nw j −

n∑j=1

w j

}for every j ≥ 2 (2)

The analysis of Eq. (2) shows that the level of market share for the incumbent clearlydecreases with the number of followers, as shown in figure 3.

It is hypothesized that price competition occurs on the intermediate market so that eachcarrier chooses to interconnect with the destination country carrier that offers the lowesttariff. Therefore, the profit made by each carrier, which depends on both the outgoing trafficand the fixed fee, can be further increased if the carrier fixes a lower interconnection tariffcompared to other competitors. Consequently, the profit function of generic carrier j candiffer according to the interconnection tariffs chosen by all the carriers in the market; inparticular:

π Aj = (

pAj − cA

j − t ABj

)q AB S A

j + (r A

j − f Aj

)S A

j if t B Aj � t B A

i ∀i �= j

with i = 1, . . . , n

Figure 3. Relationship between market share and number of carriers.


π Aj = (

pAj − cA

j − t ABj

)q AB S A

j + (r A

j − f Aj

)S A

j + (t B A

j − cAj

)q B A

nif t B A

j = t B Ai

∀i �= j with i = 1, . . . , n

π Aj = (

pAj − cA

j − t ABj

)q AB S A

j + (r A

j − f Aj

)S A

j + (t B A

j − cAj

)q B A if t B A

j ≺ t B Ai

∀i �= j with i = 1, . . . , n (3)

where: pAj is the price to the end user applied by carrier j in market A; cA

j is the cost ofthe operator j of originating or terminating a call, that are assumed to be equal; t AB

j is theinterconnection tariff paid by operator j in market A for traffic towards the market B; t B A

j isthe interconnection tariff applied by operator j in market A for traffic coming from marketB; q AB is the total quantity of calls from market A to market B; q B A is the total quantity ofcalls from market B to market A; S A

j is the market share of operator j in market A; r Aj is

the total fixed fee; f Aj is the fixed cost supported by carrier j in market A.

The analysis of expressions (3) shows that the first term in each equation representsthe profit arising from the outgoing traffic from market A to market B, the second termrepresents the profit related to the fixed fee in order to be connected to the network, and thethird one, if it exists, represents the profit deriving from the termination of incoming trafficand only depends on the interconnection tariff charged. The incoming traffic is therefore notdivided up according to the market share owned, but is gained by the carrier which offersthe lowest interconnection tariff. This solution is a result of the price competition that takesplace in the intermediate market.

4. Strategic carriers behavior

The proposed model can be used to analyze the possible strategy carried out by the incum-bent on the intermediate market and aimed at impeding and preventing the entry of newcompetitors. The leader can choose a market foreclosure strategy (Tyrole, 1997) using thecompetitive advantages it possesses: brand loyalty and degree of substitutability. The aim ofthis analysis is to compare the profits obtained from a strategy of price competition (throughprofit maximization) with the profits resulting from the market foreclosure strategy in orderto establish the best choice for the incumbent, and also taking the possible reactions of thefollowers into consideration. It is assumed that the carrier who offers the lowest intercon-nection tariff is able to guarantee complete service coverage. The carriers compete on pricein the retail market and on interconnection tariff in the intermediate market.

Competition in the retail market

If a linear demand function (4) is considered

q AB = a − bpAj (4)

270 CRICELLI ET AL.

from the first order conditions in (3) the price which maximizes the profit of the incumbentis given by:

pA1 = 1

2

(a

b+ cA

1 + t ABj

)(5)

It can observe from (5) that the retail price of the level of interconnection tariff paid isincreasing in order to complete the call in the other countries.

Competition in the intermediate market

In the analysis of the competition in the intermediate market, it is assumed that in orderto gain the entire volume of incoming calls, the incumbent effects a foreclosure strategy(Ordover et al., 1990). As stated in literature, there are many tools used to achieve marketforeclosure (Tyrole, 1997; Krattenmaker and Salop, 1986; Aghion and Bolton, 1987). Thestrategy considered in the paper identifies the level of interconnection tariff that cancelsthe rival profits. This strategy is possible since the profit function is characterized by twoprincipal components: the first is obtained in the national market using the price to the enduser and the fixed fee, while the second is derived from the interconnection. Consequently,the high level of traffic volume originated by the incumbent allows such a strategy to becarried out. Moreover, the incumbent can obtain a lower level of tariff for the outgoingtraffic due to the greater volume of originated calls. So, a linear relationship can be assumedbetween the interconnection charge and the quantity of international calls interconnectedby each carrier. This relationship is given by:

t B Aj = K − kq B A (6)

Notice that the tariff choice only determines the carriers that terminate the calls andnot the aggregate volume of calls. This hypothesis is based on a practice carried out bynumerous carriers of international telephony who set quantity discount on the terminationcharge. Therefore, in order to obtain its strategic tariff, the incumbent sets the followersprofits equal to zero and by using Eq. (3) for market A obtaining:

t B A∗j = cA

j − (pA

j − cAj − t AB

j

)S A

j

q AB

q B A− (

r Aj − f A

j

) S Aj

q B A(7)

The same is true for market B:

t AB∗j = cB

j − (pB

j − cBj − t B A

j

)SB

j

q B A

q AB− (

r Bj − f B

j

) SBj

q AB(8)

By substituting Eq. (7) in (8), the interconnection tariff of follower j (9) operating inmarket B specifically depends on the number of carriers, on the cost sustained, and on theprice level:

t AB∗j = (

1 − S Aj SB

j

)−1

cBj − (

pBj − cB

j − cAj

) SBj q B A

q AB− (

pAj − cA

j

)S A

j SBj

−(r A

j − f Aj

) S Aj SB

j

q AB− (

r Bj − f B

j

) SBj

q AB

(9)


Figure 4. Trend in the price to the end user charged by the incumbent carrier in A compared to competitivepressure in A and to the level of prices to the end user in B.

An analogous expression could be derived for market A. The incumbent in market Achooses its own interconnection tariff t B A

1 as follows:

t B A1 = t B A∗

j − ε (10a)

and the same is true for market B:

t AB1 = t AB∗

j − ε (10b)

Figure 4 shows the relationship between the price to the end user charged by the incumbentin market A, the number n of competitors operating in market A, and the price level to the enduser in market B. When market A is characterized by relatively low competitive pressure,the growth of the price level to end users in market B causes the incumbent in market A toincrease its price. Since the demand function for international calls is also linear for marketB, when the price level to the end user increases, the quantity of calls directed to marketA decreases, and consequently the profit component of the incumbent in A derived fromthe interconnection also declines. As a result of having to balance the loss of profit on theintermediate market, the incumbent increases the price to the end user.

When competitive pressure increases in market A, the reaction of the incumbent is toreduce prices. The impact of price choice operated in market B compared to market A isalso shown by the effect on the interconnection tariff of the incumbent in market A causedby the retail price in market B, as shown in figure 5. This effect increases with the growthof the ratio Q = q B A

q AB , representing the traffic imbalance. Figure 6 shows the relationship

272 CRICELLI ET AL.

Figure 5. Trend of the interconnection tariff of the incumbent carrier in A compared to the level of prices to theend user in B and to the size of the catchment areas.

Figure 6. Trend in the profit made by the incumbent carrier in A compared to competitive pressure in A and tothe level of prices to the end user in B.


between the profit of the incumbent on market A, the level of retail prices in market B, andthe number of carriers in market A. When competitive pressure is relatively low in marketA and the price level to end users in market B increases, the profit of the incumbent inmarket A grows. This increase is obtained from the growth of the incumbent’s price, whichhas already been analyzed in figure 4. Vice versa, under the hypothesis in which market Ais highly competitive, the profit for the incumbent diminishes.

The other strategy which the incumbent may adopt is that of allowing other carriers toenter the intermediate market. In this case, the incumbent fixes the tariff that allows profitmaximization. By substituting (6) in (3), the following profit function is obtained:

π A1 = (

pA1 − cA

1 − t ABj

)q AB S A

1 + (r A

1 − f A1

)S A

1 + (t B A1 − cA

1

)(−t B A1

k+ K

k

)

and, from the first order conditions, the following interconnection tariff is determined:

t B A1 = K + cA

1

2(11)

With the aim of analyzing the advantage of a horizontal foreclosure strategy for the incum-bent, the profits obtained by both strategic choices were compared. Figure 7 shows how theprofit obtained by the incumbent in the case of price competition using profit maximizationtariff (11) is always greater than that made through the horizontal foreclosure strategy by

Figure 7. Trend of the profit of the incumbent carrier in A in the case of price competition and in the case ofstrategic foreclosure in conditions of low competitive pressure (n = 2).

274 CRICELLI ET AL.

Figure 8. Trend of the incumbent carrier profit in A in the case of price competition and in the case of strategicforeclosure in conditions of high competitive pressure (n = 30).

tariff (10), whatever the price level fixed on market B. The greater the competitive pressurein market A, the greater the incumbent suffers a loss from horizontal market foreclosurestrategy (figure 8). In fact, with the increase in the level of competition, the new carriersentering the market fix a lower interconnection tariff, obliging the incumbent to reduce theforeclosure tariff further. Otherwise, if the incumbent chooses the tariff maximizing profit,the new carrier entering the market can fix a tariff equal to:

t B Aj = t B A∗

j + ε (12)

This tariff allows the follower to obtain a positive profit on the intermediate market andin turn to lower the price in the retail market in order to obtain market share.

The positive profit in the intermediate market is also possible since the use of InternetProtocol for wholesale voice telephony allows the new entrant to operate with lower costs.In fact, tier 1 players, such as AT&T and Sprint, are sending more traffic to wholesale VoIPproviders.

Such continued movement from PSTN to VoIP involves increased cost savings for tra-ditional operators. Figure 9 shows how the price of the follower is a function of both thenumber of carriers present in market A and the price level to the end user in market B. Anincreasing price level in market B causes a reduction in the quantity of calls interconnected,and, consequently, a fall in profit in the intermediate market for the follower in market A.In this case, as its total profit will decrease, the follower cannot lower the price to the enduser, which on the contrary, will tend to increase. The comparison with figure 4 shows howthe price of the follower is lower than that of the incumbent. Moreover, figure 10 showshow the market share of the follower grows when it chooses the tariff (12) and how in-crease in this effect reduces the price level in market B. From the simulations carried out,


Figure 9. Trend in the price to the end user charged by the follower carrier in A compared to competitive pressurein A and to the level of prices to the end user in B.

Figure 10. Trend in the market share of the follower carrier in A compared to the price levels to the end user inboth market A and in market B.

276 CRICELLI ET AL.

it can be seen that in the first phase of the competitive dynamics, it is advantageous forincumbent to compete on price, obtaining ever greater profits than those obtained by usinga foreclosure strategy, whatever the strategies chosen by the carriers in the other market.When the incumbent allows followers to enter the intermediate market, the followers mayin turn choose a strategy which allows them to gain a market share in the retail market. Inthis case, the follower market share increases as a result of the fall in price levels to the enduser in the other market, whereas the market share of the incumbent suffers a reduction.With the aim of maintaining its market share, choosing a strategy of horizontal foreclosurewith the consequent threat to competitiveness in the market is advantageous for incumbent.

5. Conclusions

The increased competitive pressure that has developed in the major industrialized countriesin recent years has impacted on international telephone traffic agreements. Previously,when a monopolistic regime operated internationally, the common interconnection tariffwas determined according to bilateral agreements between carriers in different countries.

From the beginning of liberalization, the interconnection tariff immediately experienceda radical decrease and at the same time became important for strategic leverage in the choiceof foreign carrier. The evolution of the competitive scenario has brought about convergencetowards a context characterized by the presence of a multiplicity of carriers and a marketleader (normally the ex-monopolist). Such an asymmetry of the relative sizes of the carriersallows the incumbent to negotiate a larger traffic volume and to be able to operate a strategyof horizontal foreclosure towards the other carriers. By not considering the range of possiblereactions from the followers, the foreclosure is a losing strategy since the simulations carriedout in this paper reveal that the incumbent obtains a better profit by competing on price.

If, on the other hand, the incumbent allows followers to enter the intermediate market,the followers may in turn choose a strategy of gaining share in the retail market. In thiscase, the market share of the incumbent suffers a decline. Considering this potential loss, theincumbent may find advantageous the choice of the strategy of horizontal foreclosure. Thisstrategy of the incumbent in the intermediate market represents a threat to competitivenessin the market, revealing the need to operate policy choices at the international level. Since thedetermination of interconnection tariff by the incumbent carrier is so important, regulationis both appropriate and necessary, which must be agreed at a supranational level in order toprotect the competitive dynamics in the international market.

Acknowledgment

Authors would like to thank anonymous referees for their useful suggestions and comments.

References

Aghion, P. and P. Bolton. (1987). “Contracts as a Barrier to Entry.” American Economic Review 77, 388–401.Armstrong, M. (1998). “Network Interconnection in Telecommunications.” Economic Journal 108, 545–564.


Carter, M. and J. Wright. (1999). “Interconnection in Network Industries.” Review of Industrial Organization 14,1–25.

Cricelli, L., M. Gastaldi, and N. Levialdi. (1999), “Vertical Integration in International Telecommunication Sys-tem.” Review of Industrial Organization. Netherlands: Kluwer Academic Publishers, vol. 14, No. 3, pp. 337–353.

de Palma, A., V. Ginsburgh, and J.-F. Thisse. (1987). “On Existence of location Equilibria in the 3 Firm HotellingProblem.” Journal of Industrial Economics, 36, 245–252.

EC. (2002). EC Directive 2002/19 of the European Parliament and the Council of 7 March 2002 on access to, andinterconnection of, electronic communications networks and facilities, published in OJ L 108, 24/04/02.

Economides, N. (1989). “Symmetric Equilibrium Existence and Optimality in Differentiated Product Markets.”Journal of Economic Theory 47, 178–194.

Economides, N. (1993). “Hotelling‘s “Main Street” with More than Two Competitors.” Journal of Regional Science33(3), 303–319.

FCC. (1996). “IB Docket 96/261 In the Matter of International Settlement Rates.” Washington, USA.Krattenmaker, T. and S. Salop. (1986). “Anticompetitive Exclusion: Raising Rival Cost to Achieve Power Over

Price.” Yale LawJournal 96, 209–295.Laffont, J.-J., P. Rey, and J. Tirole. (1998). “Network Competition: I Overview and Nondiscriminatory Pricing.”

Rand Journal of Economics 29(1), 1–37.Laffont, J.-J., P. Rey, and J. Tirole. (1998). “Network Competition: II Discriminatory Pricing.” Rand Journal of

Economics 29(1), 38–56.Lancaster, K. (1979). Variety, Equity, and Efficiency. New York, USA: Columbia University Press.Madden, G. and S.J. Savage. (2000). “Market structure, Competition and Pricing in United States International

Telephone Markets.” Review of Economics and Statistics 82(2), 291–296.Novshek, W. (1980). “Equilibrium in Simple Spatial (or Differentiated Product) Models.” Journal of Economic

Theory 22, 313–326.Ordover, J.A., G. Saloner, and S. Salop. (1990). “Equilibrium Vertical Foreclosure.” American Economic Review

80(1), 125–142.Salop, S. (1979). “Monopolistic Competition with Outside Goods.” Bell Journal of Economics 10, 141–156.Telegeography. (2004). Country Traffic Statistic. San Diego, USA: Primetrica Inc.Tyrole, J. (1997). The Theory of Industrial Organisation. Cambridge, Massachusetts, USA: MIT Press.

Wholesale Competition in the International Telecommunications System

Documents