Competition law and regulation in the Turkish telecommunications industry: Friends or foes?

ARTICLE IN PRESS Telecommunications Policy 34 (2010) 233–243

Contents lists available at ScienceDirect

Telecommunications Policy URL: www.elsevierbusinessandmanagement.com/locate/telpol

Competition law and regulation in the Turkish telecommunications industry: Friends or foes? S- ahin Ardıyok a, Fuat Og˘uz b, a b

ACTECON Competition & Regulation Consultancy, Maya Akar Center No: 100/10, Esentepe 34394, Istanbul, Turkey Bas- kent University, Department of Economics, Ankara, Turkey

a r t i c l e in fo

JEL classifications: K23 L43 L51 Keywords: Competition policy Regulation Telecommunications Turkey

abstract The interaction between national competition authorities (NCAs) and national regulatory authorities (NRAs) plays a vital role in institutionalizing competition policy during regulatory reforms. Questions about jurisdictional authority over competition policy are far from settled What role should NCAs play in regulated industries? Should we see NRAs and NCAs as complements or substitutes? This paper attempts to discuss these issues within the context of the Turkish telecommunications industry. Recent events in this industry point to relative strengths and weaknesses of a legally powerful NRA against a NCA. We address the complementarity issue as an empirical question and dispute its practical viability in a hostile environment where two agencies differ on the role of competition. The Turkish telecommunications industry shows that legal ambiguity surrounding competition policy creates inefficiencies and increased power struggles. & 2009 Elsevier Ltd. All rights reserved.

1. Introduction The interaction between national competition authorities (NCAs) and national regulatory authorities (NRAs) plays a vital role in institutionalizing competition policy1 during the implementation of regulatory reforms in certain industries. The Trinko case2 in the United States, the enlargement of the European Union and the increasing number of NRAs around the world create both theoretical and empirical issues concerning the role of competition rules3 and NCAs in regulated industries. From a political, economic, and legal standpoint, the answers to questions such as ‘who has the authority over what?’, and ‘to what extent?’ authority can be used are quite blurry. What role should NCAs play in regulated industries? Should NRAs and NCAs be seen as complements or substitutes? This paper tries to discuss these issues within the context of the Turkish telecommunications industry. Recent events in this industry point to the relative strengths and weaknesses of an NRA with strong legal authorities versus an NCA. Herein,

 Corresponding author.

E-mail address: [email protected] (F. Og˘uz). Competition policy refers to the set of rules designed to promote and protect competition and restrict monopoly practices. These rules include oversight of mergers, prohibition of price fixing and agreements (tacit or explicit) for sharing the market and other behavior that might restrain competition. 2 Verizon Communications Inc. v. Law Offices of Curtis V. Trinko, LLP, 540 U.S. 398 (2004). 3 Competition rules comprise rules and remedies to prevent anti-competitive contracts or practices, monopolization or abuse of dominant position and merger controls. 1

0308-5961/$ - see front matter & 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.telpol.2009.10.002

ARTICLE IN PRESS 234

 S. Ardıyok, F.Oguz / Telecommunications Policy 34 (2010) 233–243

the issue of complementarity as an empirical question is addressed and its practical viability in a hostile environment where two agencies differ on the role of competition rules is disputed. Many countries differ in their emphasis on the role of NCAs in the regulated industries. At one extreme, NCAs have broad powers over regulated industries, as in Australia and New Zealand. At the other end, in countries such as the United States, both state (PUCs) and federal (FCC) regulators control competition issues by means of heavy regulation, which leaves a limited room for NCAs. Turkey, as a latecomer, tends to come closer to the US example. However, the jury is still out on how Turkey should and will proceed and political interests are not yet settled. Competition is a relatively new issue in terms of legal structure in Turkey. Although the Constitution of 1982 has a provision (article no. 167) ordering the state to prevent cartelization and monopolization, the Act which brought forward the necessary legal tools was enacted in 1994 and the NCA did not even take office until 1997. Following on from the 1990s, the economic crisis of 2001 and privatizations within the banking, energy and telecommunications sectors galvanized regulatory reforms in these industries with an independent regulator being established for each industry. As a rule, the Turkish the NCA’s authority covers all industries. The NRAs create room for themselves through exceptions and immunities. In the absence of any exception or regulation,4 the NCA is supposed to have the final say on competition issues. Exceptions do not have a systematic pattern and follow industry-specific expectations. Furthermore, political and business interest groups do tend to have some power over immunities. From a legal and political point of view, the design and implementation of competition rules and regulatory relations are far from clear. Depending on the issue at hand, competition rules may become secondary to political preferences. Should there be general competition rules applied to all industries with a limited number of exceptions? Or, should a long list of exceptions and immunities be attached to the competition and regulatory rules? Politicians, regulators, and economists all tend to answer these questions quite differently. This study maintains that competition rules, implemented fully and indiscriminately, should have authority in these industries. They should also complement regulatory legislation for permanent market failures. More importantly, competition rules should be seen as a restraint on the anti-competitive behaviors of regulators. This is not the case in the current legal framework in Turkey. Political tensions and legal problems occur from time to time. For example, some NRAs do not recognize the authority of the NCA in their areas. In some cases, legal authority could not find any political support. In some other situations, legal immunities have been introduced to overcome competition issues. In this paper, the issues are considered from the perspective of complementarity. The next section addresses the complementarity issue. The advantages and disadvantages of general competition rules and sector-specific regulation are debated. Then, the discussion turns to the recent experience in the Turkish telecommunications industry. In the last section, relative efficiency and effectiveness of alternative mechanisms are compared and contrasted. The paper concludes by summarizing the findings.

2. Antitrust vs. regulation: complementary or substitute? The study of the relationship between competition rules and regulation is relatively new. Theoretical literature on the interaction between agencies (NRAs and NCAs) having the power to interfere in markets goes back to McCubbins et al. (1987, 1989). In this setting, legislative activity is structured as a multilateral principle-agent problem. NCAs and NRAs have conflicting goals and react to different incentives. Recently, Carlton and Picker (2006) offered a general framework based on McCubbins, Noll and Weingast’s work.5Newbery (2006) offers another discussion of the subject, with a differing view on the fundamental issue. While Carlton and Picker generalize their preference of competition rules over regulation, Newbery makes a distinction between the telecommunications and electricity industries, the latter possibly requiring a more extensive regulation than telecommunications does. Beato and Laffont (2002) also point to the conflict between the social goals of regulatory policy and efficiency considerations of a general competition policy. Moreover, when the regulatory reform is closely related to privatizations (as in the case of Turkey) the emerging industrial structure will be influential over the potential for competition in the industry. Some of the issues discussed below have their origins in the current privatization and liberalization reform in the telecommunications industry. Institutional environment may thus play a more important role than technical considerations. For example, in the European context, Knieps argues that technical regulations are justified for a more competitive environment. On the other hand, competition policy should take a primary role on interconnection and network access problems (Knieps, 1997).6 This may not hold for many other countries because of the continuing process of regulatory reform.7 4 The Turkish NCA does not have a legal standing of filing a suit for any NRA regulation. Even this regulation evidently lacks a sound legal foundation to overrule completion rules. 5 See McCubbins et al. (2007) for a recent recapitulation of this approach. 6 See, Buigies (2003), Economides (2005), Geradin and Kerf (2003), MacAvoy (1998), and Shelanski (2002) for further discussions of the relationship between competition policy and regulation in telecommunications. 7 On the relationship between privatization, regulation and competition, see particularly Zhang et al. (2005) and Parker et al. (2005).

ARTICLE IN PRESS  S. Ardıyok, F.Oguz / Telecommunications Policy 34 (2010) 233–243

235

The relationship between competition rules and regulation is sometimes discussed in the literature in different settings. Some authors emphasize the distinction between implicit and explicit regulation. Some others focus on ex-post and ex-ante regulation. Both classifications refer to complementarities between agencies. In cases of implicit regulation (e.g., New Zealand), regulated firms are allowed to maximize their profits, as long as they comply with competition rules. In explicit regulation, competition rules do not apply and any action that is not banned in the regulatory legislation is assumed legal. This provides incentives to seek and extract rents from consumers. If regulatory capture is widespread, implicit regulation is widely accepted as a better rule. In industries where interconnection and third party access issues create inconsistencies between regulatory preferences and competition rules, not only do political pressures bring about conflicts of interest between agencies but technical requirements do as well. And, telecommunications are all about interconnection and access to bottlenecks.8 Why do we regulate network industries instead of putting them under the enforcement of competition rules? The economic justification of regulation is based on the co-existence of natural monopoly and network externalities. For example, if more than just one firm can provide long-distance telephone service efficiently, there cannot be a strong economic ground to support a vertically integrated monopolist owning the sole local telephony network. This is the case even though there may be social and political reasons.9 Economides (2006, p. 97) puts forward three reasons for choosing regulation rather than relying only on competition rules. First, market failures may make it impossible to reach competition in the industry. Second, social goals do not fit well with economic efficiency. Finally, social and private benefits may differ. On the other hand, Spiller (1996) gives more emphasis to political preferences on the existence of regulatory institutions. For the new institutionalist school, the power of politics over these industries provides the ground for regulation.10 From an institutionalist perspective, utilities are almost always part of a political game. Scope and scale economies, asset specificity, and massive consumption gives way to political interests and politicized individuals who want to take part in the decision-making at all stages (Spiller & Tomassi, 2005). For example, widespread use of electricity by households and other industries may create hold-up problems in a free contracting system. The history of energy and telecommunications is full of many ingenious ways to infiltrate politics into economics. In this environment, there is no easy way to separate ‘political’ from ‘economic’ or purely ‘technical’ considerations. 2.1. Complements or substitutes? Why do regulators exist? Historical roots of regulatory activity and its modern economic explanations present conflicting views on the issue. Theoretical reasons for regulators are mainly about the difficulty of setting rates, ambiguities about the efficient levels of production and similar economic problems of natural monopolies in network industries. Historical accounts of regulatory activities provide a different story with an emphasis on ‘political’ nature of decisionmaking in these industries (Benson, 2005; Spiller & Tomassi, 2005, pp. 522–523). Political economy emphasizes the elimination of regulatory capture and governmental opportunism. While NRAs with a nature of independence are introduced to reduce rent-seeking activities, it is now widely believed that they are both a means and an end to regulatory capture.11 Are regulations political or economic in nature? The answer to this question sheds some light on the issue of complementarity. Following the literature, it is argued that regulation is more about restraining governmental opportunism (and creating new opportunities for rent-seeking in some cases) and competition is more about restraining market opportunism (and creating new opportunities for anti-competitive behavior in some cases). Which institution should have power over anti-competitive actions in regulated industries? Carlton and Picker (2006) focus on political considerations. According to them, the institution which can say ‘no’ most easily to politicians and special interest groups should have the final say. There are four potential candidates for the task: Judiciary, the NCA, the NRA, and political branch like Ministry of Transportation in Turkish Telecommunications Agency. Even though the political branch may hold a strong gateway12 to effective competition law enforcement, its role eventually would be limited to determine macro policies in that industry. However, due to its inherited nature, the political branch is prone to influence by political considerations and interest groups.13 Therefore, the political branch should be placed on the far end of Carlton and Picker’s aforementioned scale. That 8

This description excludes rent-seeking and immoral behaviors, which definitely supply an unwelcome addition to conflicts of interests. Before the modern era of universal services obligation (USO), the main justification asserted by the vertically integrated monopolists was crosssubsidies benefiting rural areas and local users of telecommunications supported by politicians and NRAs. 10 Historically, politics apparently has the upper hand. The early regulators in the United States (Armentano, 1982; Carlton and Picker, 2006; DiLorenzo, 1996) and other countries (Spiller and Levy, 1994) support this thesis. The introduction of NRAs in Turkey, similarly, has much to do with the political environment, rather than economic considerations (C - etin and Oguz, 2007). 11 Theory of economic regulation (of Chicago) and the public choice theory (of Virginia) justify this view. 12 In Turkey, Ministry of Transportation can practically block liberalization of a market by delaying the approval of minimum licensing fees for new entrants. 13 This feature actually caused the emergence of the policies to establish NCAs, and NRAs having some amount of independence from the political branch. 9

ARTICLE IN PRESS  S. Ardıyok, F.Oguz / Telecommunications Policy 34 (2010) 233–243

236

said, while NRAs are more sensitive to social needs and political pressure, the judiciary is, theoretically, the most independent. The NCA remains in the middle of these dimensions. In Turkey, courts are not an option in choosing alternative institutions in order to establish and implement competition rules. As the US Supreme Court plainly expressed in its Trinko Decision,14 courts are ill-suited for a role of regulator in terms of identifying the proper price, quantity, and other terms of doing business.. In other word, the courts of first instance lack expertise and resources of the work that the NCAs and the NRAs are doing. However, a court’s role in supervising these institutions at an appeal level is no doubt vital. The existence of NCAs and NRAs somehow legally restricts the role of the political branch in overseeing the day-to-day operations of the industry. Thus, here the focus is on two feasible alternatives and discussion of their strong and weak points.

2.2. Advantages of involving competition authority The defense of complementarity is based mostly on technical considerations. First, when the industry has natural monopoly characteristics, there is a need for a regulator. The regulator should aim to improve competition in the industry. The inability of an NRA to deal with anti-competitive behaviors adequately requires an NCA to deal with antitrust issues in the market. On the other hand, NCAs do not have adequate capabilities to solve institutional and price-related problems.15 Industry-specific NRAs are better equipped to know the structure of the industry. Thus, a well-defined legal structure having two institutions with different responsibilities can improve social welfare and competition. Put differently, there is much to be gained from specialization. Further, competition rules can be used to control regulations in the industry. Anti-competitive regulations may be restricted through this process. Particularly, an independent competition agency may put a cap on rent-seeking behavior during privatizations and deregulations. An NCA may provide a check over an NRA’s decisions and actions. NRAs are more prone to interest group politics. The institutional environment and daily regulatory burdens increases the interaction between the industry and the regulator. This situation makes regulators more sensitive to the problems of the producers. In regulated industries, producers are usually a collection of large firms, and consumers are a large group of society. Thus, the benefits of collective action become higher than the costs for producers. The opposite is true for consumers. Competition rules can play a constraining role over the decisions of NRAs. A typical example is a merger situation. Mergers in telecommunications are first evaluated by the NRA. Then, the NCA considers these mergers from a competition law perspective. Non-economic considerations may force the NRA to approve anti-competitive mergers. The privatization of Turk Telekom in Turkey supports this argument. During the privatization of Turk Telekom, the NCA insisted on divestiture of the company’s cable TV assets. Shedding these assets was vital for enabling facility-based competition in the long run. Contrary to the NCA’s holdings, the NRA set forth that Turk Telekom had to be privatized as a whole entity. Second, a purely economic scrutiny may eliminate some of the anti-competitive pressures, assuming that an NCA is not under political pressure and makes its decisions solely on economic principles. Strong political pressures may align the NCA with the NRAs, as discussed below. Last but not least, competition rules and an NCA are less influenced by political pressures. As a result, competition rules do not change as often as regulatory rules are amended. Within the last decade, regulatory rules have seen many changes yet the Competition Act in Turkey has remained unaltered.

2.3. Advantages of pro-NRA law First, NRAs have a comparative advantage in the relative speed of decision-making in network industries. Many complex issues such as tariff-setting and licensing require quick action.16 NCAs, being general institutions, cannot be as responsive as NRAs. Network industries require swift action in cases of crises. Competition authorities are not sufficiently quick to response to economic crises; at least they are less responsive than NRAs. Even NRAs can dawdle as the last Californian energy crisis showed (Ardiyok, 2006a). The same is true for Turkey. In the financial crisis of 2001, the banking regulator could not react to the apparent signs for calls for regulation and in the end lacked the ability to intervene in a timely manner.17 Social and political considerations provide ground for an independent regulatory agency. NCAs are less sensitive to social needs. For example, there can be cases where it is more efficient not to provide any telecommunications service, yet 14

Verizon Communications Inc. v. Law Offices of Curtis v. Trinko, 124S. Ct. 872, 879 (2004). Australi NCA having a special department dealing with regulatory issues cannot be deemed an exception to this argument. 16 This is true under the assumption that there is no regulatory capture or political pressure. Here, the Chicago approach and the institutionalist view differ. For Spiller and institutionalists, governmental opportunism is not an ad hoc and exogenous variable. It is part of the model. See Spiller and Tomassi (2005) for a general discussion. 17 The reasons are controversial, yet it is widely believed that error played a more prominent role than regulatory capture. The Turkish telecommunications regulation has another story: the regulator was deliberately slow in resolving roaming issues and some other cases. Political capture, apparently, was the reason. 15

ARTICLE IN PRESS  S. Ardıyok, F.Oguz / Telecommunications Policy 34 (2010) 233–243

237

social considerations may require it, as in many rural areas. Specifically, the argument for essential facilities and issues surrounding interconnection make an NRA the most likely candidate for regulatory control. The social requirement for regulatory activity depends on institutional factors. In some industries, essential facility doctrine does not hold a strong ground and innovations play a more important role than natural monopoly characteristic (e.g. mobile phone services). In these industries, competition rules become relatively more efficient than regulation and should thus be given priority. In cases where cross-subsidies and other social goals prevail, regulators try to both limit entry into the market and to restrain competition in order to create rents to be distributed. Competition rules may limit the effectiveness of this kind of regulatory action. In some other cases, an NRA may want to control entry into a market in order to reduce possible risks in cases of crises. However, in most cases, rent transfers find more empirical support than these arguments. Another justification for regulation comes from network effects. Strong network externalities limit the ability of NCAs to correct anti-competitive behavior. The size of externalities depends on many factors and may be smaller in reality than theory would suggest (Economides, 2006). The above discussion assumes that there is a ‘categorical’ difference between NCAs and NRAs. Pressure groups can capture NRAs more easily (i.e., less costly) than NCAs. In regulated industries, lobbies get more benefit for the same cost, due to the lower transaction costs of collective action. An alternative approach may be to relax this assumption and argue that there is no categorical difference between them. As public choice theory argues, if NCAs behave like NRAs in regulated industries, they may easily become part of a political wealth transfer game.18 Different goals and tools create tensions between NCAs and NRAs. The Turkish telecommunications industry is a case in point. The NRA holds the view that it has full authority over the industry, including the development and enforcement of competition rules. This view has created an institutional environment where anti-competitive behavior easily flourishes. In the end, while theoretical considerations would support complementarities between the NCA and the NRA, political reality makes them rivals. The Turkish telecommunications industry is again an example of this view. In the rest of the paper, the Turkish telecommunications industry is considered from this perspective. Before going into the details of the Turkish experience, it is beneficial to explain the situation in the EU given that Turkey must adopt the EU’s telecommunications acquis if Turkey’s EU membership is to be considered. Examining the situation in Turkey through the lens of the EU’s requirements helps to illuminate the issues in the Turkish telecommunications industry. The EU issued a new regulatory framework19 in order to deregulate all portions of telecommunications industry in order to support the concept of an information society.20 A major amendment in the framework which requires NRAs and NCAs to complement each other is stated in the framework directive.21 In the Article 3 of the directive it is stated that ‘‘y Member States shall ensure, where appropriate, consultation and cooperation between those authorities, and between those authorities an national authorities entrusted with the implementation of competition law and national y National regulatory authorities and national competition authorities shall provide each other with the information necessary for the application of the provisions of this Directive and the Specific Directives.’’ On the other hand, the EU obliges NRA’s to talk in competition law language while designing their respective regulations (Streel, 2003). Additionally, the Commission has limited the scope of the markets to be regulated by the NRA’s by announcing a list and requiring a heavy burden of evidence to add additional markets to this list. That means the NRA’s are not able to enlarge their scope of jurisdiction without a market analysis procedure, which would be overseen by the Commission (Keagan, 2004). The EU legislation seems to draw a clear line of complementarity. However, real-life outcomes may differ from country to country. For instance, in the Netherlands a protocol between the NRA and the NCA is in force and works considerably well. That said, almost the same protocol which was signed between the Turkish NRA and NCA has not generated friendly cooperation. This suggests that supporting institutions, such as the legal system, traditions of bureaucracy, and the approach of relevant politicians, results in different outcomes under the same set of rules. 3. The Turkish experience: telecommunications Competition does not have a long history in the Turkish telecommunications industry.22 Until 2004, fixed telephony services were under state monopoly and provided by Turk Telekom. The mobile segment had only three firms and the industry was heavily regulated. The establishment of an independent regulatory agency in telecommunications (NRA), combined with pressure to deregulate the industry, was a major step towards a more competitive market structure. International evidence supports the importance of competition for telecommunications (Newbery, 2003). However, political and economic dimensions of competition differ in local and long-distance fixed-line businesses. Mobile telephony 18

McChesney and Shughart (1995) provide a collection of classic articles of this view. Turkey has a duty to adopt E telecommunications acquis on its way to membership. 20 For a detailed description of objectives of New Regulatory Framework, see Hocepied (2002). 21 Directive 2002/21/EC of 7 March 2002 on a common regulatory framework for electronic communications Networks and service. 22 For general discussions of the regulatory environment in the Turkish telecommunications markets, see Atiyas and Dogan (2007), Atiyas (2006), and Burnham (2007). In this paper, a general introduction to the industry is not provided and readers are referred to above mentioned works. 19

ARTICLE IN PRESS 238

 S. Ardıyok, F.Oguz / Telecommunications Policy 34 (2010) 233–243

has a different market structure. It is widely believed that the local fixed telephony market when compared to longdistance has the characteristics of a natural monopoly. The magnitude of specific assets is much smaller in the mobile industry (Spiller & Tomassi, 2005, p. 521). This is one reason why private companies are more interested in the mobile market in the recently liberalized markets of East Europe and Turkey. ‘Competition’ became a buzzword in the 2000s with the establishment of a watchdog agency in Turkey. The expectations from consumers were lower prices and better services. Yet, the current situation in mobile telephony is a settled oligopoly of three firms that charge almost the highest prices in Europe (Atiyas & Dogan, 2007, p. 502). The political nature of the regulatory structure and the legal ambiguity about the role of the NCA helped to transform a first-mover advantage in a newly established sector in mobile services into an institutional barrier to entry.23 The competition act (Act no. 4054) governs competition rules in Turkey.24 The Competition Agency (NCA) has the authority to oversee competition issues in all industries with very few industry-specific exceptions.25 The Act bans restraints of competition through agreement, concerted practices and decisions of association of undertakings; abuse of dominant power and mergers to achieve a dominant position in any industry. The legal framework in telecommunications does not provide any direct immunity or exception from the NCA investigations. The governing statue for the telecommunications industry is The Act on Electronic Communications (No. 5809). The provisions about which agency has the right to investigate competition infringements in the industry are still contentious. Particularly, Act No. 5809 favors the NRA in sharing the responsibility. On the other hand, it has some provisions which explicitly accept the NCA’s authority to oversee competition infringements. This provision requires the NCA to take the NRA’s opinion before rendering its final decision on cases regarding the telecommunications industry. Nonetheless, the opposite is not necessarily true, as the NRA is not obliged to consult with the NCA on its decision related to competition rules in the industry.26 Due to an absence of the straightforward allocation of legal authority with respect to enforcement of competition rules in telecommunications, the NCA and the NRA have conflicting views on the role of competition rules in this industry. The NRA claims that the NCA lacks any authority to enforce competition rules in telecommunications markets (Atiyas, 2006). In addition, the NCA is not able to question the NRA’s regulations with respect to competitive issues on legal grounds. In other words, the NCA does not have any standing to sue on any NRA decision or any regulation even if they are against the law (Competition Act). Parties, which may have legal standing, like firms or associations have never challenged NRA decisions. Based on the findings of regulatory capture theory, it is expected that firms in a specific industry prefer an industryspecific regulatory agency to a strong NCA. In this vein, telecommunications firms in Turkey have strategically supported the NRA for its authority over the industry. In practice, the NCA is tougher as regulatory capture theory predicts and more comprehensive in investigations. For example, the NRA has never carried out raids like the ones made by the NCA in almost all industries. The same is true for penalties and remedies for anti-competitive behavior. Consequently, the NRA’s enforcement capacity remains more limited than that of the NCA. From a historical perspective, the NCA’s attitude toward competition issues in telecommunications has changed over time. During the initial stage, it was more eager to deal with anti-competitive behavior in telecommunications. Yet, later on, it has became more reluctant. The footprints of this drift can be traced from the NCA’s decisions in the industry. As an implied policy, the NCA prefers not to initiate an investigation into potentially anti-competitive behavior on which the NRA has an overseeing regulation. The Judiciary may play a vital role where regulation theory points out problems like forum shopping27 between regulators whose lines of authority are not clear as is the case in Turkey. Nevertheless, the competency of courts or judges has been far behind that of regulators and market players. Eventually the Council of State, the appeal body for both NRA and NCA decisions in Turkey, has remained reluctant to solve the jurisdiction of authority problem between these regulatory agencies. This problem is faced mainly by firms that have applied to the NRA and/or the NCA in order to prevent incumbents’ anti-competitive actions. As expected, the incumbents have brought forward the issue of the authority of the NRA and the NCA in telecommunications and tried to keep the focus away from the substance of the subject matter. Prospectively, the NCA has taken an active role by suggesting the drafting of a protocol, taking The Netherlands’ protocol between NMa and OPTA as an example. After lengthy negotiations, the protocol was signed in 2002 between the NRA and the NCA to ease potential conflicts. It outlines rules for interaction and consultation. However, the implementation has so far not been adequate. The lack of any legal enforcement makes it part of a strategic game between agencies.

23 When controversial issues came before it, the Danıs-tay (The Council of State) has had a conservative approach to these cases. One of the reasons is the lack of technical capacity in the Danıs-tay in those cases. The inability to understand relevant issues thoroughly leads the Danıs- tay to fall into type 2 error in its decisions. 24 Turkey follows the European Union rules on regulation. As it aspires to be a member to the Union, it restructures its legal body according to the EU model. 25 For example, banking mergers and acquisitions with an amount of 25% or less are immune from the application of Act no. 4054. 26 While it is uncertain whether this omission is intentional or just a political mistake, its endurance gives the signal that politicians are not disturbed by the situation in the industry. This point also strengthens the authority of the NRA. 27 Forum shopping describes the practice adopted by some firms to get their case handled in the institution which is most likely to render a favorable judgment.

ARTICLE IN PRESS  S. Ardıyok, F.Oguz / Telecommunications Policy 34 (2010) 233–243

239

The rest of the paper looks more closely at a number of cases and tries to elaborate the tension between the two agencies and address the consequences of inconsistent views on competition.

3.1. Cases Historically the decisions of the NCA in telecommunications and the NRA’s position in those cases can be categorized into two periods. In the first period, which lasted until 2006, the NCA was very passionate about pursuing any anticompetitive actions in telecommunications. The NRA, in this period, was trying to get together the regulatory pieces in the industry. It can be asserted that the NRA’s institutional capacity was not strong enough to resist political pressures, which aimed to protect the state-owned incumbent operator Turk Telekom. This situation created a base for the threat of regulatory capture and was reinforced by faulty choices during the institutional establishment of the NRA. To begin with, the NRA was not formed as a brand-new institution, like the NCA, but built on the Wireless Department of the Ministry of Transportation. In the Wireless Department, the majority of the staff had been engineers and normally they did not have a sense of regulatory literacy. Additionally, it recruited most of its staff from Turk Telekom.28 In the first period the NCA took a market oriented approach in spite of the NRA’s politically oriented position. In those first years, the NRA was not sufficiently ready to remedy market failures in telecommunications. Most of the major regulations were not in place and its enforcement capacity was not coherent enough to deter incumbent firms. Nevertheless, a regulatory process was in place and the NRA was issuing new licenses to new operators. In this environment, incumbents could adopt strategies to discourage actual entries. Under unsophisticated knowledge of complementarity in competition rules, normally it was expected that the firms would only apply to the NRA that claimed that it had authority to investigate competition infringements. However, rival operators mostly preferred to apply to both the NCA and the NRA. One reason might be the NCA’s good reputation in other industries and its detailed and consistent procedures to handle the cases.29 Particularly, those who can benefit from more competition will likely prefer the NCA; but incumbents tend to deal with the NRA. On the other hand, due to low litigation costs in Turkey, complainants try to increase their chances by applying to both institutions, without noticing this strategy might have created an opportunity of forum shopping for defenders or incumbents. This strategy might be suicidal for newcomers. Incumbents have discovered that both institutions are eager to investigate competition issues and by favoring the NRA in the challenge, they could create room for a dispute of substitution between the NCA and the NRA. Unfortunately, in the first period, this dispute limited both institutions and distanced them from the merits of the cases. Hence, very long legal considerations are observed in the NCA decisions which try to explain the allocation of authority or responsibility over competition rules in telecommunications. After the general framework of the first period, it is better to review the cases that have the potential of illuminating regulatory institutions’ behavior in that period. Due to a lack of major regulations, which could remedy permanent market failures, most of the cases in the first period were about monopolization or abuse of a dominant position. The first one is a landmark case concerning Turk Telekom initiated by a complaint of the association of internet service providers (TISSAD).30 The NCA investigated a wide range of markets where competitive entries occurred after deregulation in the industry. At that time, the NRA was just established and the Ministry of Transportation was still a major authority over the industry. Thus, initiating a legal inquiry about Turk Telekom meant questioning the political will of the Minister who appointed the management of the state-owned undertaking. The NRA joined to the case later on, and in its opinion asserted that it had jurisdictional authority and initiated its own investigation of the dispute. In spite of NRA resistance, the NCA concluded the case and rendered the first administrative fine to a state-owned undertaking, Turk Telekom.31 In its decision on the dispute of Turk Telekom v. _IhlasNet,32 it was stated that the NRA’s opinion was taken by the NCA. In its opinion, the NRA claimed that _IhlasNet also applied to the NRA and the power of authority to investigate competition infringement vested in the NRA. The NCA went on with its investigation, disregarding the NRA’s opinion, and decided that Turk Telekom wasn’t abusing its dominant position. The only exception where the NRA referred a case to the NCA and did not say a word about the NCA’s investigation authority, was a complaint by the GSM operator _IS- TIM about concerted practices on national roaming between the first 28 The Telecommunications Board, the managing body of the NRA, is related to and controlled by the Minister of Transportation. The Minister appoints the majority of the board members. In the period here defined as the first stage, there was a conflict of interest between the NRA and the Ministry. The Ministry, having the power to control Turk Telekom, which was a state-owned firm and subject to regulations of the NRA, could indirectly control the NRA as well. 29 The Turkish Competition Act was the first legislation that brought detailed administrative procedural rules that were adopted from EU Council Regulations on the implementation of the rules on competition of the Treaty. The telecommunications laws lack this kind of procedural rules that would provide a sort of commitment or limitation to the NRA in its discretions. The most obvious difference is that while the NRA can issue decisions with only a page long reasoning, the NCA should lay out its detailed grounds and that sometimes reaches more than a hundred pages. 30 NCA Decision dated 2.10.2002 and no. 02-60/755-305. 31 This might have been a good benchmark in its later investigations over state-owned enterprises. Unfortunately, the NCA could not establish a tradition for not discriminating between private and public companies in this respect. 32 NCA Decision dated 23.7.2007 and no. 03-53/601-267.

ARTICLE IN PRESS  S. Ardıyok, F.Oguz / Telecommunications Policy 34 (2010) 233–243

240

two GSM operators Turkcell and Telsim.33 The NCA pursued the investigation and found no evidence of concerted interconnection practices that could be deemed as an infringement of the Competition Act. The national roaming case was the hardest experience for the NRA, because this case was the first implication of socalled faulty design of regulatory structure in telecommunications.34 The authorization provision of the Telecommunications Act requires the NRA to sign a long-term concession agreement with a licensee when a limited resource like frequency is to be allocated. Normally this agreement comprises some sort of regulatory provisions. Additionally, parties to this agreement have the right to bring any disputes in front of international arbitration. This means that if the needs in the market require a new regulation, it can immediately create a controversy with the provisions of the concession agreement.35 Depending on this asymmetry between two regulations governing the same issue, market players might realize the rights of international arbitration given to the operator in its concession agreements. Besides, the arbitration procedure relatively slows down the execution of the new regulation in any case. A controversy arose in the national roaming case. After signing a concession agreement with GSM 900 operators Turkcell and Telsim, the NRA issued a regulation which delivered newcomer GSM 1800 operators (_IS- T_IM and Aycell) a right to use incumbent Turkcell and Telsim’s existing networks or infrastructure under certain conditions (national roaming obligation). Incumbent operators refused to comply with the regulation claiming it was not explicitly required in their concession agreements. They brought the dispute to international arbitration and before that they applied to local courts in order to get an interim injunction to stop the execution of the NRA regulation. Local courts issued verdicts without indepth analysis in favor of the plaintiffs. _IS- T_IM complained to the NCA on the grounds that Telsim and Turkcell collectively abused their dominance by rejecting _IS- T_IM’s request to use their networks which were essential facilities for _IS- T_IM to penetrate the market. The case was whether incumbent operators should provide access to roaming for new entrants.36 The NCA found Turkcell and Telsim, acted against the competition law,37 following the essential facility doctrine.38 This case was an excellent application of the complementary nature of the NCA in telecommunications. The NRA had been blocked from remedying the market failure. Then, the NCA took the lead and investigated the dispute. Later on, the international arbitration court decided in favor of the NRA and the NRA applied the provisions of the national roaming regulation to the parties aforementioned. The NRA imposed a fine as high as the NCA’s. However, the appeal body repudiated the decision due to the technical infeasibility of the regulation. Consequently, the NRA’s decision, which was based on technical features of the NRA’s regulation, could not get approval at the appeal process. On the other hand, the presence of market failure due to the reluctance of incumbents to provide national roaming was tested with further activities in the market. Newcomer GSM operators failed to enter the market efficiently and had to merge in order not to become insolvent. The NCA easily approved their merger taking ‘‘failing firm defense’’ as the primary argument.39 The tension between the two agencies regarding their authorities of responsibility reached its peak during another investigation into Turkcell and Telsim following a complaint by _IS- T_IM.40 The latter GSM operator was an affiliate of Telecom Italia in Turkey and entered the market 8 years after Turkcell and Telsim. _IS- T_IM claimed that these two operators concertedly lowered their tariffs for on-net calls (calls in their respective networks) and increased their interconnection rates (off-net tariffs) between them. As required by legislation, the NCA took the opinions of the NRA in several phases of investigation. The NRA maintained that it would handle the case and bring necessary remedies to correct any market failure. At the end of the investigation, the NCA did not find clear evidence to find that Turkcell and Telsim had offended, but laid down its foundation for a complementary relation between the NRA and the NCA in telecommunications.41 In addition, the NCA criticized the NRA opinion as follows: ‘‘We aimed not to get the NRA’s arguments whether defendants conducts were violating competition rules but to happen to know whether there was any special condition for telecommunications industry surrounding the case at hand’’. The NCA’s extended explanation was also a response to the defendants’ arguments that they had informed the NRA of their tariffs to so they did not have a standing as the NRA did not take action on these filings. The NCA elaborated its reasoning about the differences between the tariffs that were approved or determined by the NRA and those notified to the NRA. This was the focal point where, in the second period, the NCA changed its position. This will be examined below. In another case, both the NRA and the NCA investigated Turk Telekom’s actions in the cable TV market asserted by internet service providers (ISSs) to be anti-competitive. In this case,42 the NCA stated that, the NRA’s regulation, ex-ante in nature, covering access issues in broadband internet services, could not prevent the anti-competitive behavior of Turk Telekom. Therefore, the NCA stepped in to investigate this behavior in order to remedy harms created. The NCA, by implying its ex-post role in the industry, actually defined its border of jurisdiction and showed that its ex-post inferences 33 34 35 36 37 38 39 40 41 42

NCA Decision dated 1.5.2003 and no. 03-28/348-148. See Shelanski (2007) for an extensive discussion on the causes and consequences of misconstructed regulatory frameworks. OECD, ‘‘Competition Policy and Concessions’’, Policy Brief, May 2007. NCA Decision dated 9.6.2003 and no. 03-40/432-186. Major legal grounds of the case can be found in Ardiyok (2006a, b). The amount of the fine imposed on Turkcell and Telsim ($15.4 million and $6.1 million, respectively) was the highest in all cases handled in Turkey. NCA Decision dated 18.12.2003 and no. 03-81/970-399. NCA Decision dated 25.10.2004 and no. 04-68/976-236. This explanation took 21 pages out of 40 pages in total of reasoned decision. NCA Decision dated 10.2.2005 and no. 05-10/81-30.

ARTICLE IN PRESS  S. Ardıyok, F.Oguz / Telecommunications Policy 34 (2010) 233–243

241

would complement the NRA’s ex-ante regulations. The NCA, in this case, found that Turk Telekom violated its dominant position but did not render a penalty. It was an exceptional decision with a finding of infringement but no penalty.43 The Competition Act requires that the NCA has to complete its investigations within a certain period of time. However, in telecommunications it is obliged to take into account NRA opinions before rendering its final verdict. In several incidents, the NCA experienced problems in complying with time limits due to the NRA’s late replies. The idea of signing a protocol indeed was an outgrowth of the NCA’s searches to settle this problem. Otherwise, any action taken would provide defendants with grounds for appeal. For example, in a dispute between Avea and Turkcell,44 the NRA did not render its opinion in time and the NCA was left no choice but to complete the investigation without the NRA’s opinion. Not all of the NCA’s cases have been problematic with regard to relations with the NRA. In cases where it was obvious that the NCA had the jurisdiction to investigate, such as all merger, negative clearance and individual exemption applications, the NRA provided the NCA with its opinions. In addition, in competition investigations, which were away from the scope of the NRA’s regulations, the NCA concluded these cases on time.45 In one instance, the NCA found Turkcell, the dominant GSM operator, having exclusive deals with certain phone producers (e.g., Ericsson).46 These exclusive dealings distorted competition in the market and led to an NCA case by the rival operator, Telsim. The NCA endorsed the claim and found Turkcell in infringement of the Competition Act. It imposed a record penalty. To sum up, with minor exceptions, the first period of historical development in the relations between the NCA and the NRA shows that, the NRA insisted on the view that the NCA was interfering in the jurisdiction of the NRA. It also claimed that the authority of both ex-ante and ex-post intervention in telecommunications markets were vested in the NRA. On the other side, the NCA played an active role during this period when the NRA left room for players having market power and tried to conceptualize the notion that the NCA’s role was complementary and ex-post to the role of the NRA. In the second period, the NCA adopted a passive role in the enforcement of competition rules in telecommunications. The main reason for this lane-change appears to be a vigorous discussion that occurred between the President of the NCA and the members of the committee in the parliament during a session on the draft bill of the Electronic Communications Act. The bill was drafted by the NRA and Ministry of Transportation, and as expected, preferred a framework with a limited or no role for the NCA.47 Consequently, it can be argued that the drafters’ reasoning accepted the NCA’s role as that of a substitute not a complement. From that point on, the NCA has generally left the cases to the NRA when the NRA shows even a little resistance to the NCA’s jurisdiction in the case. The NCA also began to reduce its complementary role in telecommunications. For example, if an ex-ante rule fails to recover or remedy an evident market failure and the injured party applies to the NCA, this institution prefers to refer the case to the NRA again. The NRA usually fills the gap by amending its regulation. This amendment is effective from its creation, so the injured party would not be satisfied legally as a result. Taking a closer look at the cases in the second period, the case48 concerning Turk Telekom initiated by a complaint by an alternative ISS, Superonline, is exemplary. Superonline claimed that Turk Telekom abused its dominant position in internet services by applying a price squeeze in its marketing campaign for teachers. The NCA took the NRA’s legal opinion, which stated that the case was in its jurisdiction. By adopting this opinion as its final decision, the NCA referred the case to the NRA. In a similar case between GSM operators Turkcell and Telsim,49 the NCA decided that the case was out of its jurisdiction, relying on the NRA’s opinion that the parties had also applied to the NRA, which was handling the case. A case of raising rival’s cost between long-distance operators and their supplier and rival Turk Telekom also indicated the NCA’s reluctance to be involved in telecommunications.50 This case could be deemed to be a perfect application of complimentarity between ex-ante regulations of the NRA and the ex-post role of the NCA in the industry. Although the NRA took every measure for new entries of long-distance operators after issuing their licenses, Turk Telekom found new strategies to stall these firms. Those strategies were comprehensively outlined in the petition of the complainants, but the NCA preferred to reject all of them based on the view that the NRA could have solved them. In sum, according to the NCA those practices which raised long-distance operators’ costs in the market could be remedied by the NRA’s further regulations in the future. Thus, the NCA implicitly leaves competition rules enforcement to the NRA in telecommunications. With this decision, the NCA created a time in which an anti-competitive injury was created but not remedied legally. Interestingly, in 2006, a private company, Eser Telekom, filed a case with the NCA complaining about Turk Telekom’s behavior favoring OTE Globe (an affiliate of the Greek incumbent OTE) in an auction for high capacity transmission lines between Turkey and Greece.51 In this case, the NRA unexpectedly claimed that the authority to deal with the case belonged 43

Eight months later, Turk Telekom was privatized but the cable TV business was divested from Turk Telekom as required by the NCA in the process. NCA Decision dated 14.4.2005 and no. 05-24/281-76. An example to this situation might be NCA Decision dated 8.7.2005 and no. 05-44/634-166 on Turk Telekoms application in order to receive individual exemption for its agreement with ADSL solution providers. 46 NCA Decision dated 20.7.2001 and no. 01-35/347-95. 47 This bill was enacted on 05.11.2008 comprising a much more dominant role for the NRA in telecommunications. 48 NCA Decision dated 8.9.2005 and no. 05-55/833-226. 49 NCA Decision dated 10.11.2005 and no. 05-78/1052-294. 50 NCA Decision dated 14.4.2006 and no. 06-27/331-81. 51 NCA Decision dated 16.2.2006 and no. 06-13/155-39. 44 45

ARTICLE IN PRESS 242

 S. Ardıyok, F.Oguz / Telecommunications Policy 34 (2010) 233–243

to the NCA. Surprisingly the NCA decided that there was not an infringement of competition rules in spite of several pieces of clear evidence of infringement. The objective of this part of the study is to question the substitution and complimentarity issue in the Turkish telecommunications industry. Therefore, the discussion concentrates on the disputes where the NCA and the NRA’s authority were discussed or their role in the case could be seen as complementary. In some other cases, these issues were not primary. For example, mergers, negative clearance or exemption applications are always concluded by the NCA without major objections from the NRA. On the other hand, cases with competition allegations from the unregulated portions of the telecommunications industry are normally solved by the NCA as well.52 The only problem worth mention here might be the NRA’s unpunctuality with requests for legal opinion from the NCA. Unfortunately, the protocol drafted between the NRA and the NCA has not been able to remedy this trouble.

3.2. Discussion Examples are plenty and the point is clear. The NCA and the NRA have had conflicting views on the authority over competition rules enforcement. The lack of a well-defined jurisdictional boundary between these institutions has created room for political maneuvering by the NRA. NRAs are, in general, political institutions. They are usually established to be a restraint over administrative expropriation, yet become important tools in the political rent-seeking game. The NRA in the Turkish telecommunications industry is no exception. The tension between the NRA and the NCA can also be seen from this angle. The NCA, being a general institution, is less likely to be captured by specific industries. It emphasizes the economic nature of competition and remains outside political opportunism most of the time. The NRA’s regulations and practices that resulted in protection of the market positions of Turkcell and Telsim for a long time, its non-critical acceptance of Turk Telekom’s actions, and other examples show that it plays the role of the intermediary in wealth transfers of monopoly rents between demanders and suppliers.53 The NCA is reluctant to face politicians or regulators. In the second period, it has chosen to stay away from confrontation. More generally, the NCA does not actively pursue the role of competition advocacy, even though it is in its law. The absence of a legal standard leaves the issue up to the appeal body, Danistay (the Council of State) and the Constitutional Courts. Neither of them has addressed the issue conclusively so far. The Constitutional Court has the authority to define the boundaries between the NRAs and the NCA. It may choose the interpretation that the competition law applies to all restraints of trade as the US Supreme Court did, or may favor the NRAs and substantially limit the role of the NCA.54 To summarize, regulation and competition rules offer alternative ways to implement competition in regulated industries. Both the NRAs and the NCAs have advantages and disadvantages. The NRAs are more prone to lobbying and political pressure, yet they are better equipped in price-setting and market regulation. the NCAs are mostly immune from rent-seeking, yet they are weak in dealing with pricing and other sector-specific issues. This creates room for gains from trade. NRAs can deal with government expropriation and opportunism with less cost than competition authorities can. This is related to their being mostly political-institutional agencies rather than making purely economic decisions. With respect to the issues of competition, an NCA may prove efficiency enhancing in controlling the competition-related aspects of sectorspecific regulations. In the end, regulation and competition rules can become foes more easily than friends. While complementarity may increase social welfare, substitution remains politically more feasible. References Ardiyok, S. (2006a). California energy crisis and critics of Turkish electricity deregulation process. International Journal of Business, Management and Economics, 1(2), 45–71. Ardiyok, S. (2006b). Comparative analysis of collective dominance. Journal of Yeditepe University Law School, 3, 1. Armentano, D. (1982). Antitrust and monopoly: Anatomy of a policy failure. New York: Wiley. Atiyas, I. (2006). Competition and regulation in the Turkish telecommunications market. Ankara: Tepav. Atiyas, I., & Dogan, P. (2007). When good intentions are not enough: Sequential entry and competition in the Turkish mobile industry. Telecommunications Policy, 31, 502–523. Beato, P., & Laffont, J. (2002). Competition in public utilities in developing countries (Sustainable Developing Department Technical Papers Series). Washington, DC: Inter-American Development Bank. Becker, G. (1983). A theory of competition among pressure groups for political influence. Quarterly Journal of Economics, 98, 371–400.

52 In the NC Decision dated 1.3.2007 and no. 07-18/165-55, ADSL modem retailers brought a case to the NCA complaining about Turk Teleko procurement auction on 500,000 ADSL modems. The NRA in its legal opinion stated that the case was in the authority of the NCA. 53 See Becker (1983) and Peltzman et al. (1989) for a discussion of the political economy of regulation. 54 The issue came to the US Supreme Court more than a hundred years ago. In that case, the Supreme Court decided that antitrust law should be applied in regulated industries as well (Carlton and Picker, 2006).

ARTICLE IN PRESS  S. Ardıyok, F.Oguz / Telecommunications Policy 34 (2010) 233–243

243

Benson, B. (2005). Regulation, more regulation, partial deregulation, and reregulation: The dynamics of a rent-seeking society. Advances in Austrian Economics, 8, 107–146. Buigies, P. A. (2003). The competition policy approach. In P. A. Buigies, & P. Rey (Eds.), The economics of antitrust and regulation in telecommunications—Perspectives for the new European regulatory framework (pp. 9–26). Cheltenham, UK: Edward Elgar. Burnham, J. (2007). Telecommunications policy in Turkey: Dismantling barriers to growth. Telecommunications Policy, 31, 197–208. Carlton, D., & Picker, R. (2006). Antitrust and regulation (John M. Olin Law & Economics Working Paper No. 312). The Law School, University of Chicago. C - etin, T., & Oguz, F. (2007). The politics of regulation in the Turkish electricity market. Energy Policy, 35, 1761–1770. DiLorenzo, T. (1996). The myth of natural monopoly. The Review of Austrian Economics, 9(2), 43–58. Economides, N. (2006). Competition policy in network industries: An introduction. In D. Jansen (Ed.), New economy and beyond (pp. 96–121). New York: Texas A&M University. Economides, N. (2005). Telecommunications regulation: An introduction. In R. Nelson (Ed.), The limits and complexity of organizations (pp. 48–76). New York: Russell Sage Foundation Press. Geradin, D., & Kerf, M. (2003). Controlling market power in telecommunications: Antitrust vs. sector-specific regulation. Oxford: Oxford University Press. Hocepied, C. (2002, May 20). The new EU regulatory framework for electronic communications: From sector specific regulation to competition law. Presentation at IBA/ABA communications and competition: Developments at the crossroad, 20–21 May, Washington, DC. Keegan, S. (2004). The new regulatory framework for electronic communications networks and services. DG Information Society. Knieps, G. (1997, November). Costing and pricing of interconnection services in a liberalized European telecommunications market. Institut fur Verkehrswissenschaft and Regionalpolitik, Albert-Ludwigs-Universitat Frieburgr Breisgau. Discussion Paper-Nr. 39. MacAvoy, P. W. (1998). Testing for competitiveness of markets for long distance telephone services: Competition finally?. Reviev of Industrial Organization, 13(3), 295–319. McChesney, F., & Shughart, W. F. (Eds.). (1995). The causes and consequences of antitrust. Chicago: University of Chicago Press. McCubbins, M. D., Noll, R., & Weingast, B. (1987). Administrative procedures as instruments of political control. Journal of Law, Economics, and Organization, 3, 243–277. McCubbins, M. D., Noll, R, & Weingast, B. (1989). Structure and process, politics and policy: Administrative arrangements and the political control. Virginia Law Review, 75, 431–482. McCubbins, M. D., Noll, R., & Weingast, B. (2007). Political economy of law. In M. Polinsky, & S. Shavell (Eds.), Handbook of law and economics, Vol. 2 (pp. 1651–1738). London: Elsevier. Newbery, D. (2006). The relationship between regulation and competition policy for network industries. Electricity Policy Research Group working paper, CWPE 0631, University of Cambridge. Newbery, D. M. (2003). Privatising network utilities. Mimeo, Cambridge (paper for CESifo conference on privatization, Cadenabbia, Oct 31–Nov 3, 2003). Parker, D., Kirkpatrick, C., & Zhang, Y. -F. (2005). Price and profit regulation in developing and transition economies: A survey of the regulators. Public Money and Management, 25(2), 99–106. Peltzman, S., Levine, M. E., & Noll, R. G. (1989). The economic theory of regulation after a decade of deregulation. Brookings papers on economic activity: Microeconomics, Vol. 1989. Washington, DC: The Brookings Institution pp. 1–41. Shelanski, H. A. (2002). From sector-specific regulation to antitrust law for US telecommunications: The prospects for transition. Telecommunications Policy, 26, 335–355. Shelanski, H. A. (2007). Adjusting regulation to competition: Toward a new model for US telecommunications policy. Yale Journal of Regulation, 24(1), 55– 105. Spiller, P. (1996). Institutions and commitment. Industrial and Corporate Change, 5(2), 421–452. Spiller, P., & Levy, B. (1994). The institutional foundations of regulatory commitment: A comparative analysis of five country studies of telecommunications regulation. Journal of Law, Economics and Organization, 10(2), 201–246. Spiller, P., & Tomassi, M. (2005). The institutions of regulation: An application to public utilities. In C. Menard, & M. Shirley (Eds.), Handbook of new institutional economics (pp. 515–543). Dordrecht, The Netherlands: Springer. Streel, A. D. (2003). The integration of competition law principles in the new European regulatory framework for electronic communications. World Competition, 26(3), 489–514. Zhang, Y., Parker, D., & Kirkpatrick, C. (2005). Competition, regulation and privatisation of electricity generation in developing countries: Does the sequencing of the reforms matter?. Quarterly Review of Economics and Finance, 45, 358–379.