Image

Competition Policies in Emerging Economies

Lessons and Challenges from Central America and Mexico

This page intentionally left blank

Competition Policies in Emerging Economies

Lessons and Challenges from Central America and Mexico

Claudia Schatan • Eugenio Rivera
Editors

Project of Canada's International Development Research Center (IDRC) and the Economic Commission for Latin America and the Caribbean (ECLAC)

"Strengthening Competition in Latin American Isthmus: National Policies and Institutions, Regional Coordination and Participation in International Negotiations"

Image

Foreword

The challenges faced by market competition have been more widely recognized in the Latin American region over the last few years. There has been renewed interest in antitrust policies, in modernizing various regulations and achieving greater transparency in the way firms operate. The relevance this topic has acquired has grown precisely at a time when the concentration of wealth has deepened regionally and globally. The lack of appropriate pro-competition legal and institutional frameworks during the privatization process of large public enterprises in the 1980s and 1990s and a great number of ensuing mergers and acquisitions have made possible frequent anti-competitive practices, adversely affecting consumers and the competitiveness of producers.

In a number of public utility services essential to the economy, large privatized firms, formerly under public ownership, often act as monopolies. These practices have spread internationally. Economic liberalization and digitalization have made it easier to invest capital in foreign markets, but little has been done to curb abuse of market power in many developing countries where they operate.

This book addresses competition policies in Central America and Mexico, particularly in the banking and telecommunications sectors, in which market distortions have led to low levels of efficiency and competitiveness. In the cases of both of these sectors, access to credit and a modern telecommunications system is vital for the constant innovation and efficiency of their services. On the other hand, access to these services has become part of the population's basic well-being. The arrival of foreign direct investment (both regional and international) in the banking system and in the telecommunications sector has not produced improvement in quality or more competitive prices of these services in most of the countries studied. An effectively enforced competition policy can go a long way towards strengthening these sectors, among others, especially in small economies where this policy faces many obstacles set up by strong economic and political interests.

The experiences of the seven countries studied in this volume are a valuable point of reference for competition policy officials, who are either adapting their laws in order to strengthen and adjust them to their own realities, or else are enforcing newly enacted competition laws for the first time. The studies included in this book also provide a wide range of experiences in the difficult relationship that usually exists between competition policy officials and sector regulators and suggest ways of improving their cooperation.

This volume contains the most important results of the project "Strengthening Competition in the Central American Isthmus: National Policies and Institutions, Regional Coordination and Participation in International Negotiations", funded by International Development Research Center (IDRC) and implemented by Economic Commission for Latin America and the Caribbean (ECLAC). Within this project 18 national studies were carried out: three in each country of the Central American Isthmus covering general competition conditions, competition and regulation policy in the banking sector, and competition and regulation policy in the telecommunications sector. These documents, together with three similar studies on Mexico, nurtured three chapters of comparative analysis in this book. Another two chapters show a country case study of competition policy application (Costa Rica) and a sector case study – competition within the banking system – in Mexico. Finally, the conclusions of the book are presented in an international perspective, where the experiences of third developing countries are brought up and enrich the findings of the book.

The efforts to improve competition conditions in the region have increased, but there is a long way to go. We can expect knowledge and experience to develop much more in the coming years.

José Luis Machinea
Executive Secretary
ECLAC

Acknowledgements

Our sincere gratitude to IDRC for having financed the project "Strengthening Competition in the Central American Isthmus: National Policies and Institutions, Regional Coordination and Participation in International Negotiations" which was executed by ECLAC between 2004 and 2006. This book is the second one published by ECLAC on competition policy which has been funded by IDRC. The first one was Competition Conditions and Policies in the Central America and Caribbean Small Economies (written in Spanish), coordinated by Claudia Schatan and Marcos Avalos (published by Fondo de Cultura Económica and ECLAC in Mexico in 2006). This book reflects the findings of more than 25 documents produced in these endeavours.

We deeply appreciate the work by all the consultants who participated in the IDRC/CEPAL project and made a direct or indirect contribution to this book. Besides Marcos Avalos, Fausto Hernández, Adolfo Rodríguez and Pamela Sittenfeld, who appear as authors, we would like to thank the background papers written by Claudio Ansorena, Pedro Antonio Argumedo, Edgar Balsells, Simon Evenett, Marco Fernández, Ricardo González, Greivin Hernández, Mauricio Herrera, Maribel Macías, Judith Mariscal, Francisco Molina, Gustavo Paredes, Diego Petrecolla, Antonio Romero, Marlon Tábora, Carmen Urízar, Leiner Vargas and Marlon Yong.

We are sincerely grateful to Fernando Ramírez and Juan Luis Ordaz for their excellent technical support to produce the comparative Chapters 2, 3 and 4.

This volume could not have been written without the cooperation of the competition authorities of the Central American Isthmus and Mexico, as well as other specialists and public officers of the countries of this region. Among them, special recognition is given to Isaura Guillén, Executive Director of Costa Rican Antitrust Authority (Commission to Promote Competition); Celina Escolán, Superintendent of Competition in El Salvador; Edgar Reyes, Director of Competition, Ministry of Economics of Guatemala; Santiago Herrera, General Coordinator, National Program of Competitiveness UCP-FIDE, Honduras; Julio Bendaña, General Director of Competition, Ministry of Promotion, Industry and Trade, Nicaragua; Gustavo Paredes, former President and Commissioner of the Free Competition Consumer Affairs Commission in Panama; Óscar García, Head of Analysis and Market Survey, Authority for the Protection of Consumers and Defense of Competition in Panama; and, finally, Eduardo Pérez Motta, President of the Federal Competition Commission of México. All of them contributed to the discussion in workshops and seminars which enriched the documents that have provided the main inputs for this book.

We also acknowledge the very valuable contribution of Andres Rius and Susan Joekes from IDRC to the substantive part of the IDRC/CEPAL project, which helped conceive solid studies, as well as Taimoon Stewart, who has done deep reflection on competition in small developing economies, for her thoughtful comments. Observations from Sebastián Sáez and Verónica Silva, from the International Trade Division of ECLAC, Santiago, Chile, were also very helpful. Other organizations which provided support and made an important contribution to the discussion of the background papers for this book were the Central American Bank for Economic Integration in Honduras; the Economic Policy International Center for Sustainable Development, National University of Costa Rica; the Economic and Social Research Institute, Rafael Landívar University in Guatemala; and the Investment and Development for Exports Foundation in Honduras.

We also want to thank Rebeca Grynspan for her support to the IDRC/CEPAL project and the development of the studies contained in this volume. In the same way, we are indebted to our colleagues Fernando Cuevas for generously sharing his knowledge on competition policies; Hugo Enrique Sáez, who supported the Spanish edition of the book; and Pedro Cote, for his great efforts in the communications area, which helped us have important feedback on the papers and discussion on which this book is based. We also thank Laura Gutiérrez for her efficient administrative support throughout the project, which made it possible for many people to collaborate in the IDRC/ECLAC project and the production of this book. We also thank Nicholas M.A. Papworth for the translation from Spanish into English.

Contents

Foreword

v

Acknowledgements

vii

Contributors

xi

1 Introduction

1

Claudia Schatan and Eugenio Rivera

 

2 Markets in Central America and Mexico: What Is Happening with Competition?

7

Eugenio Rivera and Claudia Schatan

 

3 Models of Privatization and Development of Competition in Telecommunications in Central America and Mexico

49

Eugenio Rivera

 

4 Competition and Regulation in the Banking Systems of Central America and Mexico: A Comparative Study

97

Eugenio Rivera and Adolfo Rodríguez

 

5 Advantages and Limitations of Costa Rica's Experience in Competition Policy. A Benchmark for the Rest of the Countries in the Central American Region

145

Pamela Sittenfeld

 

6 Banking Competition in Mexico

189

Marcos Avalos and Fausto Hernández Trillo

 

7 Findings in an International Perspective

223

Simon J. Evenett and Claudia Schatan

 

Index

235

This page intentionally left blank

Contributors

Marcos Avalos Bracho, Ph.D. in Economics, Essex University, England. He also has postgraduate studies in education teaching techniques on economics regulation at the JFK School of Government, Harvard University. He has been Economic Affairs Officer at the Economic Commission for Latin America and the Caribbean (ECLAC) in Mexico, professor at the Instituto Tecnológico de Estudios de Monterrey-Campus Ciudad de Mexico (ITESM), Centro de Investigación y Docencia Económica (CIDE), Instituto Tecnológico Autónomo de México (ITAM). Currently, he is full-time profesor at the Centro de Alta Dirección en Economía y Negocios (CADEN) at the Universidad Anáhuac. Among his more recent publications are: Condiciones y políticas de competencia en economías pequeñas de Centroamérica y el Caribe; Claudia Schatan and Marcos Avalos, Editors, FCE and ECLAC, 2006, Regulación en aerolíneas en México; Marcos Avalos and Víctor Valdes, CIDAC, 2006, Política de competencia en México: las fusiones; Marcos Avalos and Camilo Vidal, CIDAC, 2007.

Adolfo Rodriguez Herrera, Ph.D. in Economics, Catholic University of Louvain, Belgium. He worked at the University of Louvain as researcher and visiting professor (1989–1994). He has held different executive positions in the public sector of Costa Rica, such as Superintendent of the Securities Commission for eight years and was also adviser to the Presidency of the Republic (1995–1998), and more recently to the President of the Social Security Institute and to the General Manager of the National Bank of Costa Rica. He has worked as a consultant for IDB, The World Bank, UNDP and ECLAC. He has recently finished all the coursework of a postgraduate degree in psychoanalytic theory at the University of Leon and the Institute of Neurosciences and Mental Health in Barcelona (Spain). Currently, he is a member of the Board of Directors at the Service Regulatory Authority of Costa Rica.

Claudia Schatan, MA in Economics, University of Cambridge, England. She is Head of the Trade and Industry Unit at the Economic Commission for Latin America and the Caribbean (ECLAC), United Nations, in México. She has been professor and researcher at the Centro de Investigación y Docencia Económica (CIDE) in Mexico, and visiting researcher of different universities in the United States (Harvard Institute of International Development (HIID), Harvard University, Center for United States-Mexican Studies, California University in San Diego, California, among others). Among her more recent publications are: Condiciones y políticas de competencia en economías pequeñas de Centroamérica y el Caribe; Claudia Schatan and Marcos Avalos, Editors, Fondo de Cultura Económica and ECLAC, 2006; El medio ambiente y la maquila en México: un problema ineludible; Jorge Carrillo and Claudia Schatan, Editors, Libros de la CEPAL No. 83, México D.F. 2005.

Pamela Sittenfeld, MSc in Regulation, with emphasis in Utilities Regulation and Competition Law from the London School of Economics and Political Science. She holds a Law degree from the Universidad de Costa Rica. She was the ex-director of Costa Rican Antitrust Authority (Commission to Promote Competition) and worked as a consultant for the Economic Commission for Latin America and the Caribbean (ECLAC) in Mexico. Currently she is a member of the National Council for the Supervision of the Financial System (CONASSIF), a member of the Board of Directors for the Regulatory Authority of Public Services of Costa Rica (ARESEP) and a consultant at Weinstok in Costa Rica.

Fausto Hernández Trillo, Ph.D. in Economics, The Ohio State University. He is a professor of economics at the Centro de Investigación y Docencia Económicas. He has been Budget Director in the Finance Ministry in Mexico, and a member of the Research Department at the Mexican Stock Exchange. He also is the author of three books and numerous articles in academic journals. His research interests include fiscal issues and financial markets. He has also been a consultant for the Mexican and Guatemalan governments and several international organizations such as the IMF, the World Bank and the IADB. Currently, he is also the Director of the prestigious academic journal El Trimestre Economico.

Eugenio Rivera, Ph.D. in Economics and Social Sciences, Free University of Berlin. He has been professor and researcher at the Centro de Investigación y Docencia Económica (CIDE) in Mexico, Coordinator of the IDRC/ECLAC Research Project on "Strengthening competition in the Central American Isthmus: National policies and Institutions, Regional Coordination and Participation in International Negotiations", Head of the Regulation Division of the Ministry of Economics in Chile, and Chairman of the Preventive Antimonopoly Commission. Currently he is participating in the creation of the Regulatory Commission of Public Works in Chile. Among his more recent publications are: Teorías de la Regulación en la Perspectiva de las Políticas Públicas; New trends in the Latin American telecommunications market: Telefonica & Telmex.

Chapter 1
Introduction

Claudia Schatan and Eugenio Rivera

The need for an efficient functioning of markets has been gaining progressive recognition in the developing countries as a mechanism to supply goods and services at reasonable prices for consumers, inputs at competitive prices for producers and for levelling the playing field for potential competing companies, including small and medium-sized ones.

Between the 1950s and the 1980s, the control of prices, interest rates and imports in Latin America guaranteed the development of the national industry, as well as the access of population and producers to essential services and to banking credits at preferential rates. This situation became impossible to sustain after the mid-1980s. The over-indebtedness, the enormous fiscal deficits and a growing lag in the competitiveness of the productive sector, among other factors, ended the aforementioned policies. Therefore, an effort to promote the good functioning of markets was, and continues to be, one of the routes that can help these developing economies to ensure a suitable behaviour of economic actors. However, that has not happened in a spontaneous way despite the macroeconomic policies of opening and deregulation. That is why the competition policy becomes indispensable to eliminate distortions in the markets.

This book focuses on the development and the challenges that the competition policy faces in Latin American countries, with a special interest in the small economies of Central America. Most of the latter countries have incorporated competition policies with a considerable lag in their governmental agendas. Thus, the main aim of this book is to analyse the market distortions in that region, the legal and institutional competition instruments governments rely on and those that could be developed to face such distortions.

The competition policy – understanding by that, mainly, the antitrust law and the competition agency in charge of applying it – has faced serious difficulties in the Central American Isthmus. There is a fertile land for anti-competitive practices in this region given its numerous national markets of reduced size and its concentrated productive structure, especially in the non-traded product areas. These markets often shelter horizontal monopolistic practices (e.g. agreements among producers of goods or similar services to fix prices, to fragment markets geographically), as well as vertical practices (e.g. conditioning of input purchasing or input sales; establishing contracts of exclusivity in the distribution and sale of goods). It is particularly important for the competition agencies to distinguish concentrations that respond to a necessity of a greater firm size required by a technological innovation which will improve industry efficiency from the practices that search the increase in profit margins through monopolistic power.1

The scarce resources of competition agencies in countries with little human and financial capital call for a special effort to adapt the competition policies commonly used in developed countries to their own realities. Social conditions lagging behind, the great number of small and medium-sized firms and the exposure of competition authorities to be "captured" by groups of interest demand a careful design of competition policies in those countries (see Schatan and Avalos, 2006).2

Besides making a comparative and critical analysis of the Central American countries' experience (Costa Rica, El Salvador, Guatemala, Honduras, Nicaragua and Panama), additional reference points are taken into account in this book, especially from Mexico. Among the Latin American countries, Mexico is the closest to Central America, not only geographically but also in competition policy subjects. In fact, Mexico has had a strong influence on the laws of this region, so that it is a forced reference when studying the countries of the Isthmus. Although the latter nation differs in size from the former ones, they share a similar cultural, legal and institutional background. The competition authorities in Mexico have had to find ingenious solutions, not always successfully, to solve difficult cases in a relatively weak institutional framework. A helpful aspect of the Mexican competition policy is the cumulative experience regarding the notification to the competition authority obligation prior to the merger.

It is important to point out that the combination of supports and resistances faced in order to engender the competition policy explains the degree of speed upon which such policies were introduced in the studied countries. Between 2004 and 2006, progress has been fast, since three countries enacted their competition law for the first time and two others revised them. The sequence of law enactment in the seven countries under study is the following: Mexico (1992, and amended in 2006), Costa Rica (1995), Panama (1996, and amended in 2006), El Salvador (2004), Honduras (February 2006) and Nicaragua (October 2006).3 Guatemala was still in the process of discussing a law project at the end of 2006. The experience gathered by the first countries that enacted competition laws has been very useful for the rest of the countries.

This book also analyses the specific conditions and policies of competition in two strategic sectors – telecommunications and banking – in six Central American countries and Mexico. These two sectors are crucial as much for the possibilities of

1 For greater details on restrictive competition practices, see UNCTAD (2004).

2 On the economic groups formation and the role played by them in Central America, see Segovia (2006).

3 The studies contained in this book were concluded in September 2006, i.e. before the Nicaraguan competition law enactment (October 2006); hence, this legal change is not taken into account in the chapters of this book.

advancing economies more quickly in the technological innovation and competitiveness as for increasing population's well-being. This is important since intercommunication has become a basic need for society, as has the access to credit and banking services. The Central American region, to a greater or lesser degree, still suffers from a lag in the access to these services. This situation is, partly, a consequence of an incapacity of oligopolies and monopolies (until recently public) to respond to the increasing demand of population and companies to these services. These two strategic sectors have also been highly regulated. However, this regulation has neglected the consequences of maintaining a highly concentrated market and has not demanded a competitive behaviour from companies.

In the telephony sector, the countries under study show different trajectories in privatization and liberalization, as well as in their market performance. In all of these countries, however, this service has improved substantially since the incorporation of the mobile telephony. This technology has allowed the expansion of telephone connections at an accelerated rate and has been able to respond to a long-time dissatisfied demand. Meanwhile, in spite of its expansion, the fixed telephony shows that it is still lagging behind. As will be seen in this book, the behaviour of agents in the market has shown different results in different market segments (mobile telephony and fixed telephony). Private monopolies that emerged from the privatization have had comparatively a much greater capacity to remain as the only company in the fixed telephony market (specially those companies to which the government gave an exclusive concession for several years) as compared to the mobile telephony sector, where there has been more space for the entrance of new companies. In any case, the companies that managed to hold on as monopolies at a national level have tended to dominate not only national markets but also regional and international ones. From this reinforced position, and helped by the convergence of networks and services, these monopolies also have progressively been positioning themselves in the market of mobile telephony, while increasing the competition in this market. Nevertheless, the increasing presence of regional duopolies in telecommunications and the competition policy's institutional weakness may frustrate the possibility to improve competition further. In a context of high barriers to the entrance and weak competition policies, the strong and generalized dominion of companies such as Telefónica and Telmex-América Móvil may encourage large companies to distribute markets among themselves and hence dominate most of them in the telecommunication sector.

Like in other countries and regions, the banking sector is one of the most State-regulated markets. This is because its inadequate operation can cause potential economic imbalances. In fact, this sector's lack of solidness led to diverse insolvency crises, especially during the first half of the 1990s. Since then, the countries of the region have subscribed the Basel Protocol, which has led to the introduction of international norms of financial supervision and to other mechanisms in order to guarantee the sector's stability. These measures have had the aim of improving the capitalization and financial soundness of the banking system. On the other hand, this has led to a greater market concentration. Nevertheless, the mere concentration in the market structure is not an indicator of abuse of market power. In fact, the information technology, the offshore operations and the greater facility for the free flow of international capitals have introduced a greater contestability in the banking system of Central America and Mexico. But important anti-competitive practices persist in this sector, and the Mexican case, studied in this book, is quite illustrative of this behaviour.

Costa Rica has been the country that accounts for the richest experience in the field of competition policy in Central America. This country not only was one of the first countries in signing an antitrust law and in creating a competition authority (Comisión de la Promoción de la Competencia, CPC), but its legal and institutional design also made the processing of hundreds of cases possible throughout ten years or more. This experience contrasts with the case of Panama, which has been limited by its unavoidable bond with the judicial system. The study on Costa Rica provides a detailed analysis of the most important cases taken by the CPC, from its creation, obstacles, the useful instruments available for their solution, to the learned lessons and the best international practices.

Although it goes beyond the objective of this book, it is important to point out that the obstacles faced by small economies to have an effective legal and institutional competition framework could be lessened through regional and international cooperation. Such cooperation has already begun to take place in trade agreements both among Latin American countries and with third countries. Nevertheless, a solid competition policy is needed at a national level. The competition clauses can include a wide spectrum, from cooperation involving consultations between competition authorities to the joint resolution of anti-competitive practices affecting countries in the region. On these lines, an important case is the agreement reached by the Andean Community, which has a Competition Commission with supranational powers. Countries in the region lacking an antitrust law of their own can use such an agreement.4 Nevertheless, the latter regional administrative entity has not reached the expected results, which has led to a reform in the agreement in 2005. Among the most influential changes carried out by the Andean Community are the greater competencies given to national agencies of competition. Therefore, strong national institutions are needed to support the growing mutual support in competition issues in a context of economic integration (Silva and Alvarez, 2006). Likewise, cooperation between developed and developing countries – key to approach anti-competitive practices of multinational companies – requires a greater institutional soundness from the latter countries. Otherwise, the competition policy institutions in the developed countries would prevent developing countries from the access to their information because they would fear that this could be handled in an inadequate way (Stewart, 2005).

The three following chapters have been elaborated based on seven national studies (six in Central America and Mexico). Chapter 2 analyses the competition policy laws in these seven countries. Chapter 3 presents a comparative analysis in the telecommunications sector for the same group of nations. There is also a

4 Bolivia and Ecuador.

regional comparative analysis for the banking sector in Chapter 4. Chapter 5 reviews the main case studies in Costa Rica, the country with the richest experience in competition policies in Central America. It emphasizes the obstacles faced as well as the lessons learned by the competition agency in that country. Chapter 6 contains a study on competition policy in the Mexican banking system which, notwithstanding its greater openness to foreign investment over the last years, is still considerably inefficient. This is a valuable point of reference for the Central American region where foreign capital is quickly starting to make its way into the financial market. Finally, Chapter 7 summarizes and interprets the main findings for Central America and for Mexico of the research conducted for this volume and relates those findings to the experiences of nascent competition agencies in other developing countries.

Bibliography

Schatan, C., Avalos, M. (2006), Condiciones y políticas de competencia. Economías pequeñas de Centroamérica y el Caribe, FCE and ECLAC, Mexico.

UNCTAD (2004), Model Law on Competition. http://www.r0.unctad.org/en/subsites/cpolicy/docs/Modelaw04.pdf.

Segovia, A. (2006), Integratión real y grupos centroamericanos de poder económico. Implicaciones para la democracia y desarrollo regional, ECA: Estudios Centroamericanos, ISSN 0014-1445, N°. 691–692, pp. 517–582.

Silva, V., Alvarez, A.M. (2006), Cooperación en políticas de competencia y acuerdos comerciales de América Latina y El Caribe: desarrollo y perspectivas, serie Comercio Internacional, N° 73, CEPAL, Santiago, Chile.

Stewart, T. (2005), Special cooperation provisions on competition law and policy: the case of small economies, In: P Brusick et al (eds.), Competition provisions in regional trade agreements: How to assure development gains UNCTAD, New York, Geneva.

This page intentionally left blank

Chapter 2
Markets in Central America and Mexico:
What Is Happening with Competition?

Eugenio Rivera and Claudia Schatan

Introduction1

Given their lack of competition, market functioning in Central America2 has become a topic of growing importance and attention. In recent years competition policy has evolved rapidly throughout Latin America, and has been included in the policy agenda of governments throughout the region.3 The purpose of this text is to evaluate the means by which some Central American countries have ratified competition laws, to analyse the characteristics of those laws or of the bills that are currently the subject of congressional debate and to identify the principal problems they experience while exploring possible solutions. Some existing laws or bills in Central America were influenced by those of Mexico, a country that has been applying such legislation for more than a decade (since 1992). That precedent justifies the inclusion of an analysis of that country as a reference point for much of the research carried out in this book.

This work assesses the reasons why those countries that have had such laws in place for a decade or more (Costa Rica, Panama and Mexico) have had to make changes to their competition laws in order to strengthen competition agencies. In the case of those nations that have only recently approved this legal and institutional framework, which is to say El Salvador (2004) and Honduras (2006), this study depicts the challenges they have faced in overcoming deep-seated opposition to approving such laws, as well as the extent to which they contain advances compared to their predecessors. Lastly, we point out the obstacles to approving such legislation in the two countries that have yet to adopt a competition law, Guatemala and Nicaragua. As was the case in El Salvador and Honduras, the

1 This chapter was concluded in September 2006. Nicaraguan competition law was enacted in October, 2006, so it is only considered as a law project in the analysis of this chapter.

2 For the purpose of this book, the Central American Isthumus includes the following countries: Costa Rica, El Salvador, Guatemala, Honduras, Nicaragua and Panama. Belize is not included because not enough information on competition was available in that country.

3 Competition law is a legal instrument for broad application that is aimed at promoting and protecting market competition.

resistance in these last two countries emerged from an optimistic view of the market's self-regulating capabilities as well as a deep-seated opposition from within the business world. With the exception of Guatemala, however, a conviction developed throughout the region that the lack of a competition policy constitutes an insurmountable obstacle to achieving proper resource allocation and efficiency at the level of production. Within this analytical context, this chapter reflects on the experience and the most adequate competition framework for small economies and developing ones such as those of Central America. This reflection also seeks mechanisms for strengthening legal bodies that attend to competition problems, as well as the forms that help to politically value the issue of competition in a way that allows it to become a true priority in governmental agendas.

In synthesis, competition has recently assumed great importance and become the subject of considerable activity in the countries under study. Whether at a swift or plodding pace, countries have elaborated proposals, approved laws or sought to present their legislatures with major reform bills in an effort to improve competition conditions. This process has taken place at a time when considerable progress has been achieved on the level of competition policy worldwide. By 2006, there were 14 countries in the Americas with competition laws and a competition authority,4 approximately half of which had been adopted sometime around the middle of the 1990s. All other developing countries have since made progress on this front. Nevertheless, this remains a controversial subject for such economies as was apparent in 2001, when it proved impossible to include the Singapore issues, including that of competition, in the Doha Round of the World Trade Organization (WTO). Although Doha took up the question of competition (in response to a proposal from the European Union), it did so in the specific hope of keeping the benefits of trade opening from falling under the control of major corporations that might ally in cartels or adopt other anti-competitive practices to assure themselves of monopoly-scale profit margins. Many developing countries have expressed the need to expand their ability to implement competition policies by assuming multilateral commitments under which they could apply trade sanctions whenever competition rules are not respected.

There exists an undeniable need for competition policy in Central American countries.5 As with most developing countries, during the decade of the 1980s the region experienced a foreign-debt crisis and a dramatic upswing in oil prices that led to a scaling back of the economic intervention capabilities of governments, while leaving the market to assume a much larger role in the resource allocation within the economy. The deep distortions markets experienced following the privatization of large-scale publicly owned companies and the deregulation of prices and trade, to name a couple of key developments that got under way in the

4 International Competition Network (ICN) web site [online] (www.internationalcompetitionnetwork.org/).

5 While many types of economic policies affect the conditions of competition, such as trade policy, we will not focus on such matters because they are beyond the scope of this study.

mid-1980s, show that such markets have failed to spontaneously operate in a healthy manner, thereby making it necessary to regulate the behaviour of industry protagonists. Owing to a series of conflicts of interest between various groups within each country, several governments failed to intervene as needed.

The circumstances described above, their impact on economic growth, as well as their adverse effect on distribution and competitiveness dispelled optimism as to markets' self-regulating capabilities. The result was to make competition policy an indispensable aspect of a successful economic reform.

In what follows, Section I develops a conceptual framework for facilitating the analysis of competition policy in small developing economies (SDC). Section II studies the processes that led to ratification of competition law in Costa Rica, El Salvador, Honduras and Panama, as well as the obstacles to approving such legislation in the remaining countries in the region, including those that only recently managed to get such laws passed. We also analyse the factors that are weighing either in favour of, or against approval of, such bills that are currently under discussion. Section III offers a detailed analysis of the characteristics of the legal frameworks that are already in place or which are currently being considered. This section also takes up the institutional characteristics of the defence of competition in Mexico, Costa Rica, El Salvador, Honduras and Panama as well as the institutional environment envisioned in bills that are pending approval by some countries in the region. Section IV considers the role of the judicial system in applying such laws while exploring the reasons underpinning the grave problems encountered in trying to achieve an effective defence of competition. In Section V, we explore competition conditions prevailing in some markets in the region, as well as some pertinent cases that have been resolved in Costa Rica and Panama. This chapter ends with a series of conclusions in Section VI.

I. Competition Policy: A Necessity in Small and Developing Economies?

1. General Context

Two decades after most countries in the region introduced a series of economic reforms, several of which were directed towards improving market functioning in order to optimize allocated and productive efficiency, the results have generally proven to be frustrating on the level of competition conditions, although some signs are encouraging. The persistence of barriers to entry, especially those related to the ability of firms and business associations to influence public policy, leads to a variety of distortions that affect the development of market competition and which have proven impervious to change despite trade openings and other pro-free trade economic reforms.

At the outset of the economic reform, doubts emerged as to the need for, or usefulness of, including competition laws as part of the reform. Powerfully influenced by analytical focuses such as unrestricted market access theory (Baumol et al., 1982), some analysts maintained that to the extent that tariff and non-tariff trade barriers, as well as legal entry barriers to various industries, could be swept away by an active deregulation process, markets would function better and even optimally. Some writers holding a different perspective questioned the convenience or possibility of applying a competition policy in a context marked by incompetent governmental institutions or, worse yet, that could be dominated by private interests. From these analysts' perspective, a maximalist application of competition law could inhibit corporate freedom and restrict the rewards that firms should otherwise derive from the functioning of markets (Khemani and Dutz, 1996).

International experience has been categorical in showing that even when the liberalization of international trade introduces important competitive pressures to domestic markets, it offers no direct guarantee that these will function properly. In fact, while products more easily enter markets, there is nothing to keep importers and distributors from taking unfair advantage of their market power when marketing their products. In this way trade liberalization in and of itself does nothing to eliminate the propensity of firms to adopt anti-competitive practices. Similarly, privatization in no way assures that companies will behave better than their public sector predecessors as they are capable of erecting private anti-competitive barriers in the place of state ones (Cernat and Holmes, 2004; Schatan and Avalos, 2006).

It is worth noting that even when large and developed countries enjoy conditions propitious for market competition, some argue the existence of innate imperfections derived, for example, from the existence of change costs, especially for non-standardized products or commodities such as sugar, flour or other goods with characteristics that are difficult to differentiate. As Klemperer (1995) has pointed out, the cost of change arises out of the need for compatibility between newly acquired items and existing equipment in order to avoid the transaction costs that arise from having to change the supplier. Other costs that derive from change are learning how to use new brands, those related to uncertainties as to the quality of unfamiliar brands, the loss of discount coupons and similar mechanisms and, lastly, the psychological costs that accompany change or the abandoning of brand loyalty for non-economic reasons. All of these costs tend to raise prices and to provoke oligopoly-related deadweight losses, thereby discouraging the entry of new companies and reducing market competition.6

6 From a different perspective, some analysts substantiate the need for competition legislation on the basis of three factors. First, there is no proper controversy resolution mechanism owing to a lack of the litigation legislation needed to assure a perfect adjudication. Second, the costs for obtaining information or the transaction costs that may be incurred by those affected by a monopoly practice prove to be greater than the costs of those who engage in such practices and, third, it is less costly to adjudicate through a specialized body than by a regular civil court (Sánchez Ugarte, 2004).

The anomalies described above tend to occur in economies that enjoy competition-conducive conditions, which is to say large and developed countries, but the irregularities are much greater in the markets of developing economies and, even more so, in those which are small.

2. Small Economies: An Elusive Concept

Proposals that tend to formulate and win approval for competition promotion legislation have sparked a series of questions and debates among SDC. These include: (i) whether such economies need competition legislation or, in contrast, whether trade liberalization and deregulation suffice to generate a self-regulating market economy; (ii) what SDC means when speaking of market competition; (iii) whether the specific characteristics of such countries demand particular types of competition laws and policies or require an adaptation of the existing ones in the way in which they should be applied; and (iv) in light of the characteristics of such nations should competition policy add to its list of goals helping to achieve social objectives?

This study focuses primarily on Central American countries, which are small and developing economies. There is a limited body of literature analysing small economies and competition, and even less dealing with the topic in relation to SDC. Further complicating matters, the authors of such literature employ different concepts of what constitutes a small economy, and some even regard this as an unnecessary analytical category. In our view, the Central American economies are characterized by specific features that demand that the competition policies applied in large developed economies – which are by no means homogeneous – undergo a certain degree of adaptation.

In practice there is no specific view of competition for SDC. An initial focus for dealing with small economies that tend to be better developed was provided by Gal (2001), who defines a small economy as one that in the case of most industries is capable of sustaining only a reduced number of competitors in the face of limited domestic demand. The factors that help determine the scope of the market include the size of the population, how it is distributed and trade openness. According to this logic, a reduced market affects the three principal indicators of social well-being: efficiency in resource allocation, efficiency in production and dynamic efficiency. According to this outlook, three characteristics of small economies make competition in their markets more complicated: high degrees of concentration, diverse entry barriers and suboptimum production levels, which is to say below the minimum efficient scale (MES) that imposes upon them elevated production costs (Gal, 2003). The principal consequence of these considerations for competition policy in small economies is that they must make economic efficiency their principal objective because they are in no position to sacrifice it in favour of achieving broader objectives such as social goals. In this regard, Gal proposes that small economies, unlike the large ones, approve concentrations that expand the size of corporations so as to broaden dynamic efficiency, even as such transactions tend to curb competition. In this regard, De León (2006) points out how difficult it is for a competition agency in a small developing country to determine up to which point the actions of companies that expand their size favour dynamic efficiency, an issue tightly linked to technological innovation and to what extent actions such as mergers and acquisitions undermine competitors.

Gal has her critics, including Simon Evenett (n.d.), who believe that Gal's concept of small economy fails to offer characteristics that truly distinguish them from the largest economies. The alleged peculiarity that these countries display on the level of firm size (MES) applies to very specific cases in which imposing barriers to entry exist. From our point of view, Gal's analysis is mainly based on the situation of countries that are highly protected and to a certain extent isolated from the international economy in ways many countries experienced before the economic reforms that began in the 1980s became generalized. Nevertheless, her analysis remains pertinent for markets of non-tradable services.

On the other hand, Gal's concept of small economies (2001) is too skewed if we recall that this category extends to include Australia, Canada, Israel and New Zealand. In such a context it is difficult to analyse through the prism of competition conditions the sort of SDC that exist in Central America, especially the smallest of these such as the Caribbean nations studied by Stewart (2006).

It is interesting to note the opposite side of small economies from the one that Gal contemplates, which is to say those of Caribbean countries that Stewart (2006) studies. In reality each of these two groups suffers very distinct market problems. Stewart argues that for many countries of the Caribbean Community (Caricom), the problem resides in the extent to which such small economies make it difficult for competitive firms to exist, thereby limiting their capacity for technological progress. Rather than generate their own technology, countries on these islands import it, so they invest practically nothing in their own research and development, thereby limiting their ability to generate dynamic efficiency. She adds that there are "sub-optimal levels of production, since a considerably large fraction of all output is produced in sub-optimal volumes and sub-optimal plants, leading firms to be inefficient and internationally uncompetitive"7 (Stewart, 2006).

Furthermore, Stewart holds that the insertion of small economies into the global economy and international trade exposes them to competition from multinational corporations, which are capable of displacing local companies and, in the process, gravely undermining employment and the provisioning of food in such economies. According to the author, in such instances it would be totally justifiable if the objectives of competition policy were to be subordinated to social concerns.

We may deduce from this that the problem that the tiny countries of the Caribbean encounter from trade opening is the substitution of the domestic production of basic goods by international sources and its grave impact on employment

7 Quoted from the author's original text in English (unpublished in such language), p. 5.

with no possibility of generating additional economic activities, thus Stewart's conclusion on the need to subordinate competition policy to social objectives.

Central American nations stand somewhere between the two extremes studied by Gal (2001, 2003) and Stewart (2006). Trade liberalization has not produced a displacement of local production in these countries on the scale that Stewart describes. Within such nations there exists a wide range of products that can be produced by companies in the category of medium-sized enterprises (MSE), or the same effect can be achieved through exports to the rest of the region or countries beyond.

For SDC on the scale of Central American countries, the elimination of entry barriers to new products may expand economic activity. There are greater similarities between Central American and Caribbean countries in the extent to which economic and political power is concentrated into a few hands, a consolidation which facilitates collusion and other anti-competitive practices. In relation to such features, Gal (2004) emphasizes these societies' lack of a competitive culture, the political influences exerted on competition authorities and the interests of groups linked to major corporations among others characteristics8 that greatly limit the possibility of applying an effective competition policy. In such a panorama, Gal suggests making a priority of competition promotion and competition advocacy9 as a means for discouraging anticompetitive practices and postponing the formal establishment of legal and institutional competition channels. From another perspective, however, she advises that a competition authority that monitors and acts in cases of anti-competitive practices may also serve as a means of competition advocacy combined with significant effects designed to dissuade such behaviour.

In practice, economic reforms in developing countries have been – or are in the process of being – accompanied by the creation of indispensable regulatory and competition institutions. Nevertheless, these bodies at first tend to be characterized as congenitally weak and are suspected of tending towards an incorrigible tendency towards arbitrariness. These bodies' weakness is accentuated when they are dependent on judicial-deficient mechanisms to review their rulings, frequently neutralizing their effectiveness.10 In regulatory spheres, this situation has frequently translated into excessively generous tariff policies that at least partially explain the possibility for incumbents to expand their companies more than they would have been able to under genuine competition conditions, for example by means of an

8 It is noteworthy that the author points out that in developing countries the intention of achieving economic efficiency and competitiveness cannot be isolated goals and that competition must at times be subordinated to certain social objectives, especially those which concern distributive matters. These contrasts with the outlook this writer expressed in earlier works: 'Therefore, social objectives should have little or no weight in the formulation of competition policies in small economies' (Gal, 2001, p. 1452).

9 Competition advocacy generally refers to two types of activity: those that consist of promoting competition culture and the issuing of opinions on the impact of competition laws and norms.

10 This situation contrasts with that of developed countries. In England, for example, jurisdictional authorities generally assume that the regulatory authority serves the public interest and, as a consequence, firms are reluctant to carry their disputes before such judicial authorities.

aggressive policy of takeovers.11 In the specific field of competition, the panorama described above has contributed to undercutting the effectiveness of competition policy that in some countries has led to a sense of generalized impunity in this field. A case in point is the extent to which such violators excessively recur to an amparo (in the region's legal systems, an amparo is a frequently used appeal alleging a violation of constitutional rights that can act as a stay or injunction against government decisions, in both regulatory and judicial proceedings). In this way institutional deficiencies, particularly on the level of competition, can obstruct the development process by deterring economic agents from adopting salutary practices. Notwithstanding these problems, as we shall see further in this text, the countries under study have made progress towards overcoming such problems. But while the Central American countries, as well as Mexico, remain far from achieving a truly effective competition policy, it is encouraging to see the extent to which they have strengthened the process both as they have managed to draw on the experience accumulated over at least a decade into greater efficiency and in the extent to which the judicial system has grown more skilful (Costa Rica is a good example of this).

3. Some Additional Peculiarities of Small Developing Economies

Small and developing economies are highly susceptible to anti-competitive practices due to a series of issues that we dealt with in Section I of this chapter. Under these circumstances, some policy instruments assume particular importance, especially regulation. In effect, whenever several markets traditionally characterized by the presence of natural monopolies have been progressively undercut by competition the possibility has been posed of eliminating that sector's regulatory authorities and leaving supervision of such industries exclusively in the hands of competition agencies. New Zealand is the paradigmatic example in this regard. Nevertheless, it has become clear that over an extended period of time aspects peculiar to natural monopolies coexist with those of a competitive environment, thereby assuring the need to maintain regulatory policies. In effect, it is necessary to sustain tariffication in many of these industries and anti-competitive behaviour acquires forms that demand the intervention of agencies that specialize in the sector. Such situations assume even greater significance in the countries that we analysed and which face international operators in these types of sectors.12

In addition, developing countries and, even more so, the small ones in this category are in need of regulations that afford foreign investors security. In the case

11 For the case of telecommunications, see Mariscal and Rivera (2005a, b, 2007).

12 In this regard see the chapters of this book that analyse competition in banking and telecommunications.

of the countries we deal with in this study, Levy and Spiller (1996) offer an analysis of the advisability of defending investors against the risk of expropriation that should be taken into account. The regulatory framework should stimulate private investors, especially those from abroad, by employing two complementary mechanisms to restrict arbitrary administrative actions: first, by imposing substantial restrictions on discretional administrative action and, second, by putting into place formal and informal restrictions on making changes to regulatory systems and institutions that are designed to reinforce such restrictions. In the countries under study, the job of guaranteeing the rights of foreign investors is further backed by investment accords between countries. Such agreements are designed to provide foreign investors protection and non-discriminatory treatment, as well as to establish clear rules on earnings repatriation, controversy resolution mechanisms and the rules under which compensation is to be determined in the case of an expropriation, to list a few. While these concerns may appear reasonable in principle, a doubt emerges as to whether one is exaggerating the importance of investor protections to the detriment of consumers, which can explain the excessively high rates charged by major, privatized corporations for publicly necessary services (e.g. telephony services in some of the countries under study). A similar risk involving competition is to be found in some sectors. in an effort to attract greater investment, policies on competition and concentrations are left lax. It would appear indispensable to understand that only strict policy can assure the desired benefits from foreign investment.

A very important feature of competition conditions in Central America is the influence exerted by international concentration processes on the level of production (telecommunications and banks, among others). This phenomenon has emerged out of the massive wave of mergers and takeovers that got under way in the 1990s and continues unabated. The technological advances that allows various sectors to achieve greater flexibility in production should help open the market to a greater number of competitors, but while that has been the case to a certain extent (wireless phone service and electric power distribution, to cite two examples among others), it has failed to adequately offset the aforementioned concentration process, and frequently it is the same major corporations that take advantage of such technologies to diversify production and in this way further strengthen their conglomerates. While this is by no means an exclusively SDC phenomenon, and is more of a generalized process, such countries require a special effort as they lack the sufficiently powerful competition-policy instruments needed to limit market power abuses by major corporations within their countries. Singh (2004), for example, suggests that the defence of competition in developing countries would best be conducted on a regional level in order to counteract the power of the aforementioned multinational firms, and argues that national authorities are too weak to apply competition principles inside each country in an independent manner. Singh proposes the need for a supranational body to regulate the behaviour of this type of firms and limit their ability to expand through mergers and acquisitions, the effects of which are predominantly anti-competitive. In this sense, he coincides to a certain extent with Adhikari and Knight (2004), who propose regional rather than national competition policies. Both positions deserve greater study.

In synthesis, the characteristics of SDC such as those of Central America that demand special attention in the formulation or reformulation of competition policies are as follows:

• The public's lack of competition culture and the heavy weight of the corporate organization of production on consumer mentality.

• Concentration of economic and political power in a few families, thereby complicating the independence of competition authorities and facilitating the capture of public entities by powerful economic interests. This is a characteristic closely associated with a precarious rule of law.

• Scarcity of financial and human resources for the creation of institutional structures strong enough to combat anti-competitive practices.

• An abundance of small and medium-sized firms and their lack of national and global competitiveness. They may achieve a more viable entry into the international market by forming alliances among themselves – export cartels – in which case competition law would have to be applied only to the anti-competitive practices of truly large-scale firms.

• Given the small scale of these markets, regulation is especially important for achieving a more proper behaviour on the part of the companies that operate within them. This is especially valid in the case of markets for non-tradable public utility services.

• The judicial system of the countries under study is extremely precarious and lacks the capacity necessary for efficiently managing cases that competition authorities might tackle.

• Each country's abilities to combat anti-competitive practices are weak in the face of the growing international trend towards a concentration of production (in telecommunications, banking and others), so regional cooperation is indispensable in this regard.

II. Economic Reform and the Rise of Competition Law in Central American Countries

Central American countries, as in the case of many others, promoted broad programmes of economic reform as far back as the 1980s, and with particular vigour beginning in the 1990s. The reform programmes in some of these nations lagged owing to serious problems arising out of acute political crises, which in various instances escalated into civil wars. Nevertheless, things began to regain a sense of normalcy beginning early in the 1990s and significant efforts to achieve economic reform got under way. During an initial phase, these policies came in response to the problems that the traditional model of development began to experience in recent decades, the onset of a new intellectual mood both locally and internationally and the pressures exerted by international financial bodies amid the foreign-debt crisis that erupted at the beginning of the 1980s. Such developments prompted countries in the region to try and find a new way to insert themselves into the international economy. Some of the reforms from this period were conducted in the absence of the sort of legal and institutional framework essential to their success. Those that were aimed at leaving resource allocation in the hands of the market and relieving the state of that burden did so without creating conditions that would have allowed these markets to function properly once the deregulations, privatizations and trade and financial openings occurred. Such weaknesses have heightened concern within all of the countries under study for the need to establish new measures for overcoming the limitations of the initial economic reforms. One such measure was the introduction of competition policies in the 1990s or in the middle of the current decade, which is to say less than ten years since the first wave of reforms were implemented.

As the current decade began, the countries of the region were at varying stages of progress on competition matters and offered a rich amalgam of situations (Table 2.1). In 2005, some of them (Costa Rica, Panama and Mexico) were already revising laws that had been put into place roughly ten years earlier, others (El Salvador and Honduras) had just adopted a law that drew on some of the accumulated experience of the first group of countries, and the last two countries, Guatemala and Nicaragua, were continuing to debate either a draft of a law or a final bill, respectively. Most of the laws and drafts of competition law built on Mexico's experience and legislation.

A series of elements, such as the insistence and support of international bodies, served as an impetus to competition policies in the region (see Schatan and Avalos, 2006). The heterogeneous response to the stimuli and/or demand from such international agencies depended on specific circumstances in each country including the power wielded by major corporations and the extent to which governments could function independently of power groups. Even in the countries that have competition laws and authorities, their characteristics (their strengths and weaknesses) depend on the degree of resistance exerted nationally by various sectors when the draft laws were being negotiated with the various groups. We can describe at least three situations in this regard within Central America:

• Countries that have implemented extremely broad economic reforms, including massive privatizations, and which have adopted competition laws and regulatory bodies. Panama has been in this situation since 1994, while El Salvador and Honduras joined this group in 2006. Honduras' privatizations have not been as extensive as in other countries, owing to difficulties encountered in passing control of some companies to the private sector, but the country remains firmly committed to that goal (Tábora, 2007). Panama's law was revised in 2006.

• Countries with privatization and trade liberalization, with weak regulations and lacking a competition law, such as Guatemala and Nicaragua. These countries are considering a draft of a law or a final competition law bill, respectively, with Nicaragua closer to adopting it than Guatemala as it is already before Congress (since July 2005) and lawmakers have already established some points of consensus on the proposal.

• Countries that have introduced economic reforms that are more limited on the level of privatizations, but which have adopted a competition law and agency. This is clearly the case of Costa Rica, which has a competition law, and where the

Table 2.1 Legal and institutional framework: competition and regulation (August 2006) (Based on official information)

 

 

Costa

 

 

 

 

 

 

Mexico

Rica

El Salvador

Guatemala

Honduras

Nicaragua

Panama

Competition

×

×

×

 

×

 

×

Law

 

 

 

 

 

 

 

Competition

×

×

×

 

×

 

×

Commission

 

 

 

 

 

 

 

Regulation by

×

 

×

×

×

 

×

industry

 

 

 

 

 

 

 

Unified regula

 

×

 

 

 

×

×

tion from a

 

 

 

 

 

 

 

single body

 

 

 

 

 

 

 

state maintains a considerable presence on the level of banking, infrastructure, electric power and telecommunications, among other public utility enterprises. Its competition law, however, exempts public sector firms from its enforcement.

Costa Rica is the country that maintains the strongest public sector industry in Central America, one that includes a series of natural monopolies. Despite that, very early on the country embarked on an effort to reorient its growth model towards exports by introducing a programme for deregulating prices and the financial sector. Beginning in 1994, Costa Rica began to emphasize the search for free trade agreements.13 Approval of the country's competition law was motivated in part by interest in achieving the Free Trade Agreement with Mexico, which contained a clause on competition, and by the need to negotiate the Third Structural Adjustment Program (SAM III), whose conditions included having a competition law. When Costa Rica drew up its law, it patterned it after that of Mexico (Yong, 2005; Sittenfeld, 2006 and her chapter in this book). The extent to which the government enjoyed a greater degree of independence from the private sector compared to those of other countries in the region explains the relative ease with which the law was approved. However, the problem with Costa Rica's competition policy is that it contains numerous exceptions aimed at protecting public sector monopolies.

Different factors led to Panama adopting a competition law and establishing a Commission on Free Competition and Consumer Affairs (CLICAC). One of the principal motives underpinning this initiative was to achieve the institutionalism needed to support its joining the WTO in 1997. The Competition law (Law N° 29) enacted in 199614 contains the instruments for combating unfair trade practices

13 In 1994, after joining the General Agreement on Tariffs and Trade (GATT), tariffs were lowered to no greater than 55%, except on dairy products, poultry and their derivatives. Costa Rica adopted trade and reciprocal investment agreements with Mexico in 1994, Dominican Republic in 1998, Chile in 1999 and Canada in April 2001, the Caricom in March 2004 and is currently discussing a free trade agreement with the USA (DR-CAFTA). Costa Rica sustains reciprocal investment promotion and protection accords with more than 20 countries (Yong, 2005).

14 The Competition Law was enacted the same year as the Law on Industrial Property (Law 35), both of which are complementary.

internationally such as dumping as well as undertaking consumer protection. The law contains rules on four issues: monopolies, consumer protection, unfair trade practices and safeguard measures. The law emphasizes the effect of anti-competitive conduct on consumer well-being. In fact, the function of the CLICAC is frequently confused with that of setting prices favourable to consumers. The Panamanian government received considerable support from international financial bodies in preparing to deal with competition policies. This support came from the US Agency for International Development (USAID), which financed the initial studies for the draft of the competition law, and later the World Bank conducted a broad training effort. The new institutional arrangement that is expressed in a series of laws and the establishment of new bodies includes the 1996 establishment of the Public Services Regulatory Body (Ente Regulador), whose tasks include setting rates on public services that are in the process of being privatized (Fernández, 2005). Significant changes were made in 2006 to the competition law and the CLICAC was dissolved by decree to be replaced by the National Authority for Consumer Protection and Defense of Competition (ANPCDC). The changes made to the law included extending its application to all entities and types of economic activity with the sole exception of those that are of the unique purview of the state, and a major restructuring of the competition agency, which changed its character to a certain extent, as we shall see later in this text.

The case of El Salvador is different from the other countries we have mentioned. While proposed versions of a competition law were produced even before they first emerged in Costa Rica and Panama – such as a draft bill drawn up by the Salvadoran Foundation for Economic and Social Development (Fusades) in 1993–opposition from business groups and political parties in the Legislative Assembly assured that such bills would end up being filed away in the stacks of the Economic Commission. Interest in competition law revived in 1996, when the Center for the Defense of the Consumer and the Foundation for National Development (Funde) presented the Legislative Assembly with a second bill, and when in 1997, El Salvador took out a World Bank loan for implementing its National Competitiveness Program (PNC), which stipulated a strengthening of the institutional framework for the promotion of competition defence and consumer protection. USAID and the European Union were among the other bodies that worked with academic institutions in bolstering local capabilities in designing and analysing competition policies.15 However, not until evidence emerged that the economic model was stagnating did a possibility emerge for approving a competition law. In the wake of the major setback the governing

15 In 1998, USAID financed a project run by the Programa Académico de la Fundación Doctor Guillermo Manuel Ungo (Fundaungo) on strengthening competition conditions and the consumer protection system in El Salvador which identified good international practices in the design and application of competition policies as well as generating consensuses among key protagonists on the need for a competition law. In 2000 the European Union provided funding for the Friedrich Ebert Foundation to execute a programme for strengthening Salvadoran democracy that included training and technical assistance to the Legislative Assembly's Commission on Free Competition in an effort to assist in the process of designing and discussing a competition law.

Alianza Republicana Nacionalista (ARENA) Party suffered in the 2003 presidential, municipal and legislative elections, internal and external pressure began to build in favour of a recomposition of government economic strategy, and they ultimately achieved the necessary consensus on the content of the competition legislation,16 which was approved in late 2004 (Molina, 2007).

In Honduras, the task of introducing competition policy and, more specifically, a competition law has been made more difficult than in the other three mentioned countries owing, above all, to that country's significant degree of economic concentration and the intimate working relationship that exists between the government and the business sector, which rests on a web of deep personal and family ties that criss-cross sectors with economic and political power and ultimately influence many political and economic decisions (Tábora, 2007). Nevertheless, the country needed to make progress on a number of fronts before it would be capable of applying a competition law, especially in administering justice. For example, Honduras lacked a public prosecutor until 1993 when the Law on Public Prosecutors was enacted. It was impossible to effectively implement the Consumer Protection Law approved in 1989 until the 1997 establishment of the Special Consumers' Prosecutor's Office (attached to the Public Prosecutor) for prosecuting cases. Another series of laws necessary for applying competition law was recently enacted in Honduras, including an investment promotion law, the Law for the Promotion and Development of Infrastructure and Public Works (1998), a revised version of the Law on Industrial Property (2000), among others (Hernández and Schatan, 2006). Each of these measures helped to establish the principles under which agents could exercise their economic freedom in the broadest possible sense. But major resistance to the competition law remained, and the consensus necessary for approval failed to emerge despite the first efforts to draw up a version of the law, dating back to 1994 when the Office of the Presidency prepared a discussion document (Guidelines for formulating and implementing a law for promoting economic competition), and in 1996 when the first draft was published.

Renewed support for the idea emerged when the idea was listed as one of the commitments in a Poverty Reduction Strategy accord signed with the International Monetary Fund (IMF), and again among the 2004 political goals assumed under the Poverty Reduction Strategy for Honduras. In 2002, the National Competitiveness Commission was formed, whose functions included furthering the development of the PNC. The latter's tasks included improving competition and lowering administrative barriers, thereby giving added impetus to approval of a Law for Promoting the Defense of Competition. These endeavours enjoyed special funding from the World Bank. The competition law was finally enacted in February 2006 and the agency was established in August 2006.

16 The main points of consensus that emerged included the need for an independent competition agency and on sanctions that would discourage anti-competitive conduct, but which would not be confiscatory for firms.

In Nicaragua, the government remained wedded to promoting a strong public sector economy until 1989. When we add to this hurdle the economic crisis and the worsening social conditions with which the country entered the 1990s, we can appreciate just how much more of an effort would be needed to generate conditions favourable for achieving conditions of effective competition than in other countries in the region. A number of key reforms were enacted once the country emerged from a period of hyperinflation by 1991 and growth resumed in 1994. Cooperation programmes were resumed in 1992 with a series of donors – the United Nations Development Program (UNDP), German Technical Cooperation (GTZ, for its German-language initial abbreviation), the UK's Department for International Development (DFID), among others – when a support and advisory programme was created to assist with the competitive transformation of the country. In addition to macroeconomic issues, the programme anticipated a modernization of the legal framework governing the private sector as well as the institutional support upon which it could draw. Significantly, in Nicaragua it has been the forums for negotiating international trade agreements that have pushed the country to commit to, and enforce, competition laws and norms. Added urgency to approve a competition law has arisen from membership in Dominican Republic – Central America Free Trade Agreement (DR-CAFTA), which contains no explicit clauses in this regard but indirectly takes up competition in the case of the banking and telecommunications industries (Ansorena, 2007). The Nicaraguan Congress was debating the law in mid-2006 and the small and medium-size enterprises were actively supporting the law project.

Guatemala has expressed the least interest in enacting a competition law of any country in the region. The need for constitutional and legal norms in defence of competition has not been a topic of significant debate in forums within civil society or business circles; nor has it become an important item in the agendas of either political currents or the government (Romero and González, 2006). Since the second half of 2001 a draft version of a law has languished in the Economy Ministry's Department for the Promotion of Competition, but it took years to emerge as a bill. That lack of progress apparently reflects the extent to which the individuals who alternated as ministers of the economy from 2000 to 2004 were unconvinced that the proposal might win the necessary support either in the executive branch or in Congress (Romero and González, 2006). Furthermore, some leading government officials were also businessmen with direct interest in some of the most protected segments of the economy, another factor that further inhibited officials at the Economy Ministry from more enthusiastically promoting the draft law. One major source of resistance to even discussing competition law in Guatemala, as in the rest of the countries under study, is the extent to which monopolies or oligopolies control domestic markets for a number of strategic goods and services.

In synthesis, a long list of factors have led to a protracted process in approving competition law (El Salvador and Honduras) or kept such legislation from yet being approved by mid-2006 (Guatemala and Nicaragua). Those obstacles include resistance on the part of business interests that have long been protected or which have enjoyed exclusive concessions or acquired state-owned companies at very favourable terms; the influence such sectors wield within government; the very marginal extent to which this matter has been incorporated into the economic agendas of governments; a lack of local technical capacity to seriously and thoroughly explore the issue; and a paucity of competition culture in these countries. Even those nations that enacted competition law and established competition authorities in the mid-1990s and which have managed to implement reforms that restrict market power abuses continue to face hurdles. They continue to run up against the resistance posed by the most reticent power groups to effective competition policy; they feel pressured to draft their laws with too great a haste or pattern their legislation after those of other countries while failing to account for the peculiarities of Central American nations; or they lack the human capital and necessary funding to carry through on this task, or any combination of the above factors.

III. Competition Norms and Institutions in the Laws and Draft Bills Under Discussion in Central America and Mexico: A Critical Evaluation

1. Characteristics of the Competition Laws in Central America and Mexico

Concern regarding the eventual negative effects from monopolies can be found in all national constitutions in the region and, while they vary in form, all of them prohibit monopoly and anti-competitive practices. However, for an extended period of time they lacked corresponding regulatory laws, meaning that such constitutional principles were dead letter in practice. However, the existence of such nominal rules aided governments or other interested parties in promoting a legal and institutional competition framework. As we have already indicated, most Central American countries already have competition laws and regulatory bodies, and those that do not have drawn up competition bills and created offices in charge of promoting their adoption and implementation. After Honduras approved its law in early 2006, the formation of a Competition Authority was promoted by the PNC, which in turn formed part of the Foundation of Export Research and Development (FIDE, for its Spanish-language initial abbreviation). In Nicaragua, the Ministry of Promotion of Industry and Commerce (MIFIC) has a General Department of Competition that directly oversees market regulatory issues, especially those related to competition. Guatemala's Economy Ministry has a Deputy Ministry for Investment and Competition (which includes a department on competition), which is drawing up the law and norms designed to favour free competition and the elimination of protectionist practices (Romero and González, 2006). Nevertheless, the agencies in Guatemala and Nicaragua that have been assigned the task of promoting competition law lack the decision-making power needed to define market rights. As a result, they are limited to generating information and lack the ability to embark on an active policy of competition supervision.

In the following analysis, we compare the existing laws and bills under discussion. We seek to identify the various national perspectives and both the similarities and differences between countries on the level of competition. In addition to the six countries of Central America, we take up the case of Mexico, whose laws inspired those of several countries in the region.

An initial, general overview appears in Table 2.2, which lists the dates on which competition laws were approved, the name of the law, the name and date on which the respective government agency began operations, and whether the law is or is not under review.

Most of the existing laws or proposed competition laws in Central America define their main purpose as promoting competition as a means to improve economic efficiency, and in this way to raise consumer well-being. Something different happened in Guatemala and Mexico; in the former country the argument is that efficiency will benefit both producers and consumers, while Mexican laws make no explicit reference to benefits for either producers or consumers. Countries in the region tended to adopt the European model of establishing norms for the anticompetitive conduct of economic agents more than impeding the emergence of monopolies or oligopolies in domestic markets. This perspective calls for a legal framework along with a body of norms that are universally applicable to all who participate in the market for goods and services.

The classification of anti-competitive practices in competition laws differs from country to country, but the concepts are quite similar. Following the example of Mexico, Panama and Costa Rica distinguish between absolute monopolistic practices, relative monopoly practices, concentrations and unfair competition. In El Salvador, the law lists competitors, agreements between competitors, agreements between non-competitors and concentrations. However, the concept of agreements between competitors is fully applicable to absolute monopolistic practices, while agreements between non-competitors are typified as relative monopolistic practices. Guatemala's typification identifies prohibited behaviour, abuse of a dominant position and concentrations. An interesting concept in the Guatemalan draft law is that of functional competition, which refers to the difference between real competition, conditionally restricted practices and concentrations. The Honduran law distinguishes between practices restricted by their nature and those that are prohibited due to their effects. Although the terminology differs, this is another instance in which the law is speaking of absolute and relative monopolistic practices. In Nicaragua, the typification is that of monopolistic practices (concerted practices and anti-competitive agreements) and abuse of a dominant position as well as unfair competition. Concerted practices and anti-competitive practices refer to absolute monopolistic practices while abuse of a dominant position refers to relative ones.

Three countries under study – El Salvador, Honduras and Mexico – have defined specific antitrust thresholds beyond which competition is harmed and at which point official notification is required before any mergers or takeovers are entered into. The other countries lack any such notification rules but companies may voluntarily give advanced notice and officials are free to retroactively review such

Table 2.2 Laws, draft bills and competition commissions in Central America and Mexico 1992–2006 (Authors, based on official data)

 

Mexico

Costa Rica

El Salvador

Honduras

Guatemala

Nicaragua

Panama

Date

December

December

December

February 2006

 

 

February 1996

approved

1992

1995

2004

 

 

 

 

Revised

June 2006

Draft of

 

 

 

 

Law revised 2006

law

 

revised

 

 

 

 

 

 

 

law, 2006

 

 

 

 

 

Name

Federal

Law for the

Competition Law

Law for the Defense

Competition

Competition

Law 29, February 1,

 

Economic

Promotion

 

and Promotion

Protection

Promotion

1996 on Defense

 

Competition

of Competition

 

of Competition

Law

Law

of Competition

 

Law

and Effective

 

 

 

 

 

 

 

Consumer

 

 

 

 

 

 

 

Defense

 

 

 

 

 

Competition

June 1993

August 1995

January

August 2006

 

 

January 2007

agency

 

 

2006

 

 

 

 

begins

 

 

 

 

 

 

 

to operate

 

 

 

 

 

 

 

Name of

Federal

Commission

Superin-

Commission for

Intendency

Nicara-guan

National Consumer

competition

Competition

for the

tendency on

the Defense and

for the

Consumer and

Protection

agency

Commission

Promotion

Competition

Promotion of

Protection of

Competition

Authority

 

 

of Competition

 

Competition

Competition

Defense

(prior to 2006,

 

 

 

 

 

 

Commission

CLICAC)

concentrations and order a break-up of the firms in question ex post when they deem such measures are needed (see Table 2.3).

There are significant differences between the laws in terms of the areas that each addresses and of the responsibilities and functions assigned to competition authorities. The Panamanian law and the Nicaraguan draft, for example, list numerous tasks in addition to the defence of competition. Consumer protection, unfair practices in international trade and, therefore, antidumping and import safeguards are also included in the Panamanian law.17 Under the Nicaraguan bill, the competition authority must deal with market competition, consumer protection, intellectual property, simplifying filing and application processes and unfair competition (in the sense of misleading or deceiving consumers). In the other countries competition laws and draft versions exclusively focus on competition, except in Costa Rica, where consumer protection is also mentioned, but two separate entities were created for each subject: the Competition Promotion Commission and the National Consumer Commission.

Both existing laws and pending legislation generally extend competition law to all economic agents whether public or private. However, exceptions abound. In Mexico, Costa Rica and Panama, the laws do not apply to monopolies that were legally created for industries that are constitutionally reserved for the state. Many of the exceptions that were once contained in the Panamanian law were eliminated during the 2006 revision, leaving only providers of public utilities (electricity, water, etc.) under the scrutiny of the Competition Authority. Guatemala's draft law and Nicaragua's bill, as well as existing law in Costa Rica and Mexico, allow for export cartels with some conditions: that they neither reduce the value of exports nor exclude other agents involved in the same type of economic activity. Costa Rica's competition law also does not apply to service providers whose activities are regulated by concession contracts. The pending Guatemalan and Nicaraguan laws would exempt their application to intellectual property rights. In Mexico, Honduras, Nicaragua and Panama, competition law does not apply to collective bargaining agreements. The Costa Rican law contains the greatest number of exceptions, but the recently designed laws or draft versions of such laws in other countries tend to be much more broadly applicable. Practically no exceptions are to be found in the laws of El Salvador and Honduras, for example, and in the latter of these two countries the law extends even to economic sectors regulated by special sector laws. The Honduran law is the only one in the region that even applies to foreign-based professional and business associations whose actions may impact the local market; not even Mexican law contains such clauses.

Some limits on spontaneous market functioning arise out of the government's authority to set some prices. Mexico's law leaves open the possibility of setting prices of products that are regarded as essential or basic consumer items. In Costa

17 The 2006 revised version of Panama's competition law also created separate offices for consumer and competition protection.

Table 2.3 Thresholds for analysing concentrations (The competition laws and draft bills of each country)

Country

Threshold

Costa Rica

Unspecified

El Salvador

Article 33. When a merger is planned whose total assets surpass 50,000 urban minimum wages in effect in the industry or when the merged combined revenues would exceed 60,000 urban minimum wages in effect in the industry in question, the parties to the merger must request prior authorization from the superintendency

Guatemala

Article 21. Scope of application

 

Whenever any corporate merger affects or has the potential to affect the Guatemalan market and leads to an absolute dominant industry position, the Economy Ministry may call for the Superior Council for the Defense of Competition to report on:

 

(a) When the merged firm would expand its market share by 25% or more of the domestic market, or a substantial percentage of said market for any specific good or service or

 

(b) When for the most recent year the combined sales volume within Guatemala of the merger participants surpasses $30 million

Honduras

The commission will decide which concentrations are subject to regulatory approval

Mexico

Article 20. The commission must be notified before the concentration takes place if:

 

I. The transaction involves a sum greater than 12 million times the daily, general minimum wage in effect in the Federal District (Mexico City)

 

II. The transaction implies the accumulation of 35% or more of the assets or shares of an economic agent whose assets or sales are greater than 12 million times the daily, general minimum wage in effect in the Federal District; or ...

 

III. If the parties to a proposed merger have separate or combined assets or annual sales totaling more than 48 million times the daily, general minimum wage in effect in the Federal District, and the merger would imply an additional accumulation of assets or shareholder equity greater than the equivalent of 4.8 million times the general minimum wage in effect in the Federal District

Nicaragua

Whenever the merger implies an acquisition of or an expansion in market share of 25% or more, or when the combined gross revenues of the economic agents that are to merge surpasses an average of 200,000 times the annual minimum wage

Panama

The commission is to decide which concentrations have to be verified. The Guide for Control of Economic Concentration (2001) lays out critical values for some concentration indicators (Herfindhal-Hirschman and Dominance), and whenever a case surpasses those reference points it merits greater scrutiny

Rica the law allows for regulation of rates on public utilities that are still provided by state-owned companies, as well as to ration import and export licences for some products on behalf of public interest, but this mechanism can be employed only on a temporary basis and the arguments cited in its application are subject to review every six months. The Guatemalan draft law states that it does not take precedence over other laws or their accompanying rules and regulations (Article 5) that are aimed at objectives such as adjusting supply, promoting exports and dealing with idle capacity or social issues.

One of the most difficult tasks facing competition authorities is that of market definition for antitrust purposes. Five out of the seven countries in our study use the same criteria for defining a "relevant market" (Panama uses the term "pertinent market"). The relevant market normally refers to the product market and the geographical market. The former refers to the market for goods that respond to a certain specific demand that can be satisfied with similar, interchangeable and comparably priced products, while the second market refers to the area in which these goods may be obtained on favourable terms (e.g. at a distance that would not imply an additional charge for transportation costs). When it comes to sanctions, the statutory maximum fine in Mexico for absolute monopolistic practices can be as high as $1.55 million and the one for relative monopolistic practices as great as $935,000.18 In 2006, the highest potential fines in Central America were in Panama, at $1 million, followed by El Salvador, where they can reach $800,000, depending on the seriousness of the infraction. The other countries contemplate fines that are much smaller in absolute terms. Fines defined as a percentage of earnings or capital might prove severe, but these practically have never been imposed. This type of fine is contained in the laws of Costa Rica and Mexico. In this last country, such sanctions can be particularly severe as they can be as great as 10% of the annual revenues the guilty company took in during the preceding fiscal year or up to 10% of its asset value, whichever is higher. The draft law in Guatemala would impose fines of up to 100% of sales from the preceding year in extreme cases. The recently approved competition law in Honduras also leaves open the possibility that companies can appeal such rulings by filing a writ of amparo.

There are many obstacles to imposing the penalties contemplated in competition laws. In Mexico, companies accused of monopolistic practices or abuse of market power can easily appeal such rulings, thereby postponing the levying of fines or other penalties for a long period of time (Avalos, 2006). In a similar vein, there are a number of means in Panama for avoiding fines (only 10% are paid) or other punishment, at least under the law as it read prior to the revision of 2006.

As for non-monetary penalties, all of the laws or proposed laws in this study contemplate the possibility that competition authorities may order that the prohibited activity be ceased or corrected, including the partial or total break up of a business that has been improperly merged. Except in Nicaragua, all such laws allow for the possibility of filing criminal charges whenever the authority has exhausted all administrative recourse or such action is warranted by the seriousness of the accused party's conduct. Nicaragua would only impose fines and those are to be calculated based on what is deemed necessary to compensate for, or repay, the damages suffered by the affected parties. While El Salvador imposes sanctions

18 These would be greatly increased under the revised version of the competition law that the Mexican Congress was considering in mid-2006.

for procedures that contravene the law or regulations, the penalty may be lifted if the company cooperates in correcting the problem under a procedure we can generically refer to as a leniency resource. This option provides authorities with means for obtaining essential information on anti-competitive practices that they might otherwise have a hard or impossible time obtaining. The proposal to reform Mexico's competition law also introduces this concept, which is already widely used in many other countries including the USA, where companies that self-report violations or offer early cooperation with criminal or regulatory investigations may avoid criminal prosecution and fines.

Competition advocacy is a crucial issue. It is conducted in various ways in a number of countries. In Mexico, advocacy extends both to the promotion of a competition culture and the issuing of opinions – either on request from the federal government or by the authority's own initiative – regarding new laws or norms, and if the revised version of the law is approved such opinions would be binding on such rules. Costa Rica also allows the competition commission to consider whether application processes, requirements and regulations on economic activity obstruct, impede or distort either domestic or international market transactions. The Nicaraguan draft law contains a section on competition culture and refers to the simplification of filing procedures, reviews of judicial rulings and the treatment of government assistance. This is the only group of countries covered in this study that textually confers on the competition commission the power to specifically engage in competition advocacy. The laws of El Salvador and Honduras and the draft competition law in Guatemala make no mention of this subject.

2. Competition Agencies

Although there are substantial differences between the competition authorities of the countries covered by this study, something of an international consensus has formed as to the characteristics a competition agency needs in order to function efficiently and effectively. The most significant of these features include the need for autonomy from the executive branch of government; stable funding; that commissioners meet certain ideal standards (that they be competition specialists, that their nominations be widely accepted, that they serve a clearly defined term of office and that they be subject to removal for reasons that are clearly spelt out in the law). It is also recommended that the process of choosing and replacing commissioners be staggered so as to better assure the possibility of consistent rulings and an independent outlook.

As Table 2.4 shows, under the laws of Costa Rica and Mexico, the competition authority is attached to the Economy Ministry, which draws up its budget. The competition agencies of El Salvador, Honduras and Panama send either to Congress or to the executive branch their own budget proposals, which are delivered by a ministry, usually the Finance Ministry.

Table 2.4 Institutional characteristics of competition authorities in Central America and Mexico (Competition laws and law projects of Central American countries)

Characteristic/

 

 

 

 

 

 

 

Country

Mexico

Costa Rica

El Salvador

Honduras

Guatemala

Nicaragua

Panama

Name of the

Federal

Commission for the

Superintendency on

Commission for

Superior Council

Competition

National Consumer

competition

Competition

Promotion of

Competition

the Defense

for the

Law

Protection and

authority

Commission

Competition

 

and Promotion

Oversight of

Application

Competition

 

(CFC)

(CPC)

 

of Competi-

Competition

Authority

Defense

 

 

 

 

tion

(in proposal

(draft)

Authority (2006;

 

 

 

 

 

stage)

 

previously

 

 

 

 

 

 

 

CLICAC)

Legal

De-concentrated

De-concentrated

Autonomous

Autonomous

Not defined under

De-concentrated

Decentralized

structure

 

 

 

 

the current

 

 

 

 

 

 

 

proposal

 

 

Budget

Prepared by

Prepared by the

Prepared by the

Prepares its own

Undefined

Prepares its own

Prepares its own

 

the Ministry of

Ministry of the

Governing Council

budget, which

 

budget but it

budget and

 

the Economy

Economy

of the

the Finance

 

may not

submits it to

 

 

 

Superintendency

Secretariat

 

request

the Executive

 

 

 

 

proposes to

 

it directly

Branch, through

 

 

 

 

Congress

 

 

the Ministry of

 

 

 

 

 

 

 

Commerce and

 

 

 

 

 

 

 

Industry

Administrative

Five member

Commission of five

Directive Council

Five member

Commission

Not yet

Administrator, two

nature of

commission

members and

(formed by a

commission

consisting of

determined

directors and

the

 

five alternates,

Superintendent

 

a president

 

a five mem-

authority

 

with a Technical

and two directors;

 

and six

 

ber Authority

 

 

Support Unit

there are three

 

members

 

Advisory

 

 

 

alternate

 

 

 

Council

 

 

 

directors)

 

 

 

 

Designation

Executive

Executive branch

Executive branch

National

Presidential and

Executive branch

Executive branch

of authority

branch

 

 

Congress

academic

with

with ratifica-

members

 

 

 

 

commission

participation

tion by National

 

 

 

 

 

proposes 25

by Congress

Assembly

 

 

 

 

 

candidates,

 

 

 

 

 

 

 

appointments

 

 

 

 

 

 

 

are decided by

 

 

 

 

 

 

 

Congress

 

 

Commitment

Full-time, must be

Part-time for

Full time for

Full time

Dedication

Not defined

Full time. Must be

and

Mexican; have

commission

Superintendent and

 

requirements

 

Panamanian,

require-

noteworthy

members

Directors. They

 

are not

 

have at least five

ments

experience in

 

must be Salvadoran.

 

stipulated.

 

years experience

 

professional

 

There must be an

 

Commission

 

in public

 

public service

 

economist, a lawyer

 

should include

 

administra-

 

or academia

 

and others must

 

lawyers,

 

tion and not be

 

substantially

 

be professionals

 

economists

 

related by

 

related to issues

 

in related areas.

 

and other

 

family to high

 

covered by the

 

Members should

 

professionals

 

government

 

competition law

 

not be from the

 

 

 

officials

 

 

 

same family

 

 

 

 

Term

Ten years and stag-

Four years; does not

Five years.

Seven years and

Six years. The

Six years for the

Seven years and

 

gered appoint-

coincide with

Superintendent and

staggered

Nominating

first period

the authority

 

ments

political terms

directors may be

appointments

Commission

and renewal

can be re-

 

 

 

re-elected

 

may propose

takes place

elected for one

 

 

 

 

 

re-election of

in a staggered

period. Does

 

 

 

 

 

one of the

manner every

not coincide

 

 

 

 

 

members

three years

with political

 

 

 

 

 

 

 

terms

Grounds for

Reasons must be

Inefficiency,

Engage in activities

Outright

Found guilty of

Found guilty of

Permanent

removal of

considered

consistent

incompatible with

negligence in

premeditated

premeditated

incapacity

members

grave and duly

negligence that

commission duties;

carrying out

crimes;

crimes;

to carry out

 

justified

delays the

grave non-fulfil-

duties; has

permanent

incompatibility

functions;

 

 

commission's

ment of obligations

been arrested

incapacity;

with duties;

bankruptcy or

 

 

activities, guilty

and functions of the

or has mental

termination on

bankruptcy or

manifest

 

 

of premeditated

position; negligence

or physical

grounds of

manifest

insolvency;

 

 

crimes, failing to

or incompetence in

incapacity

grave non-

insolvency;

found guilty

 

 

attend three

carrying out duties;

 

fulfilment of

non-fulfilment

of a crime;

 

 

sessions during a

conduct that threatens

 

duties

of duties; fully

negligence

 

 

calendar month

the seriousness and

 

(proposed by

proven

in fulfilling

 

 

or absence from

impartiality of the

 

three fourths

incompetence

duties

 

 

the country for

commission

 

of the Superior

 

 

 

 

more than three

 

 

Council)

 

 

 

 

months without

 

 

 

 

 

 

 

the authorization

 

 

 

 

 

 

 

of the

 

 

 

 

 

 

 

commission,

 

 

 

 

 

 

 

physical or

 

 

 

 

 

 

 

mental

 

 

 

 

 

 

 

incapacity

 

 

 

 

 

In the countries under study most laws or bills initially opted for a commission rather than a single commissioner except El Salvador, where the authority is a superintendency presided over by a single director. In Panama's 2006 reform of Law 29, the three-member CLICAC commission was replaced by the National Authority for Consumer Protection and the Defense of Competition presided over by a single administrator.19

The commission formula offers the advantage that decisions are not made by a single individual but instead are the product of collective deliberations among various commissioners. A multi-member body may also prove less vulnerable to control by private economic agents. Nonetheless, the commission may be at a disadvantage in some respects compared to a single director, who might handle cases more expeditiously, for example.

The manner in which commissioners are chosen is another factor that determines the commission's degree of autonomy. A selection process, in which both the Presidency and Congress are involved, for example, may bolster a competition agency's autonomy and legitimacy by limiting the chances that commissioners could be arbitrarily removed. Panama's Law on Competition, Guatemala's draft law and Nicaragua's bill all contemplate such a selection process. In any event, commission autonomy must be accompanied by a proper mechanism for rendering accounts.

Due to the predominance of political systems that concentrate power in the Presidency, some countries have preferred to leave the faculty for designating competition authorities strictly with the president: Mexico, El Salvador and also Costa Rica, which added the step of having the Ministry of the Economy, propose a group of candidates from which the president chooses. In other countries, proposals by the president or a commission to designate candidates must be ratified by the Congress or Legislative Assembly (Guatemala, Honduras, Nicaragua and Panama). In almost all countries the designation of competition authorities is staggered (when there are appointments) or does not coincide with presidential periods, in order to isolate them from political pressures. Finally, in some countries these authorities may be re-elected and in others they may not (see Table 2.4).

Some legislation establishes norms aimed at ensuring that the individuals appointed to competition commissions have specialized knowledge (see Table 2.4). Although the small size of most of the countries considered in this study makes it difficult, at present, to demand professional preparation and experience in the field of economic competition, the laws require that the commissions include people with a background in economics together with professional lawyers. This is true of the recently approved Salvadoran law and the proposal under study in Guatemala. The case of Mexico is, without a doubt, the most advanced in this field, requiring

19 The new agency has two departments: the National Department on Free Competition and the National Consumer Protection Department. The administrator maintains a close working relationship with the five-member Council of Advisors, which is made up of the Minister of Trade and Industry, who presides over the council; the Economy and Finance Minister; the Minister of Health; a representative of the consultative council of the consumer associations; and a representative from the associations of business owners, retailers and/or industrialists.

that the professionals participating in the commission have experience in the area covered by the law.

In all the countries studied it has been made very difficult to remove competition authorities from office in order to prevent political interest groups from influencing the competition body. In Mexico, members of the commission can be removed only for well-founded causes, and other countries go one step further by specifying the only acceptable causes for removal (see Table 2.4).

Despite the multiple mechanisms that contribute to greater independence of competition authorities, budget limitations, political regimes and other factors in developing countries make it difficult to achieve total independence for the commissions and their members.

Last, it is important to note that competition legislation has been changing, or is in the process thereof, in the countries that pioneered competition policy in the region (Costa Rica, Mexico and Panama). These revisions respond largely, but not exclusively, to the difficulties encountered during the first decade of existence of the legislation.

Panama's Law on Competition was modified in February 2006. This new vision expands the law's reach to include companies that provide publicly useful services, as well as substantially altering the commission's internal structure in order to expedite case resolution, and raising fines and facilitating their collection. The new law also offers greater protection for competition authorities, who previously could face prosecution as a result of litigation from existing anti-competition cases. The new law, however, has become a source of controversy as the agency administrator has lost some of the autonomy previously enjoyed by CLICAC commissioners.

Costa Rica, for its part, was in the process of revising its competition law in 2006, but by mid-year a proposal for an alternative law was yet to emerge. The changes being debated were: improved typification of anti-competitive practices; possible requirement of prior notification of concentrations, with adequate establishment of limits; the incorporation of "clemency appeals;" the elimination of various exceptions to the application of the law; and the strengthening of institutions linked to competition policy, among others.

In Mexico, the latest version of the law (approved by Congress in mid-2006, but yet to be signed by the president as of this writing) grants greater powers to the Federal Competition Commission (FCC) to break up companies, including those in the telecommunications, media and financial services sectors, that engage in anticompetitive practices and fail to abide by earlier commission rulings. The new law imposes considerably higher fines than those imposed under previous rules in an effort to more seriously dissuade potential violators of the competition law. For example, the fine for instances of absolute monopoly practices would reach as high as 1.5 million times the daily minimum wage in effect at the time in Mexico City. Additionally, the "appeal for pardon" would be introduced to help the FCC bring together evidence to identify any anti-competitive practices by companies. Collection of the fines, which had been minimal (approximately 14% of the total levied), would be carried out by the Tax Administration System, which would be likely to increase the effectiveness of collections.

Within Central America, Costa Rica and Panama are the countries that have the most experience in resolving cases filed before competition authorities, given that in the other countries the competition commissions are very new (El Salvador) or do not yet exist. In Costa Rica, between 1995 and 2004, the competition authority resolved 537 cases, or 89% of those brought before the commission. Among the cases resolved during this period, there were 145 general consultations, 203 suits, 92 formal investigations, 47 consultations to the Technical Unit, 9 rulings on licences, 37 opinions, 2 instances of price setting and 2 mergers. Fines have tended to increase from around $2,400 in 1998 and to surpassing $400,000 in 2002. Nonetheless, the overall amount of fines applied is still quite limited. Through mid-2006, no penalty had yet been issued and paid that was equivalent to 10% of the value of annual sales or 10% of the value of a firm's assets (Yong, 2005; Sittenfeld, 2006, and her chapter in this book).

In Panama, CLICAC (whose name was changed at the outset of 2006) handled few cases of anti-competitive practices. In fact, between 1998 and 2004, approximately 30 cases of monopolistic practices were fully processed including both those brought by private companies or CLICAC.20 As was discussed earlier, the competition authority only has the faculty to investigate the existence of indications of anti-competitive practices, and if they appear to exist, the authority must file suit in the courts and wait for them to resolve the case. Since the CLICAC was created, 14 cases were settled (through 2005) without the need to file suit in the courts, either because it was decided to handle the case through competition advocacy or because evidence of wrongdoing was not sufficient to file suit. The rest of the cases were taken to court and by mid-2006 decisions had been issued at the initial and appeals levels; these cases involved meat, flour and medical oxygen, among other products. It should be noted that once a case is taken before the courts, and the judge rules in favour of the competition authority, the accused may appeal before the Superior Court and other courts, such that a case may be prolonged significantly (this has changed somewhat since the Competition Law was changed in 2006). It is interesting to note that in cases of absolute monopolistic practices brought by CLICAC that have been resolved, the courts have generally supported the position of CLICAC.

The majority of the 1,450 cases processed between 1993 and 2004 by Mexico's FCC (Avalos, 2006) were related to mergers or acquisitions, which can be explained by the requirement that the companies provide advance notification of these operations. This suggests that from the time this prior notification was required, the competition commission has had to devote considerable resources to resolving these cases. It should also be noted that the majority of the cases of monopolistic practices processed in Mexico have been of the "vertical" type (Avalos, 2006), in contrast to those handled in Costa Rica which have been principally "horizontal," partly because it has been harder to prove "vertical" practices.

20 Some cases are investigated confidentially and the number of such cases is unknown.

IV. The Competition Law and the Judicial System

International experience shows that in various developing countries, two principal mechanisms have served to debilitate the institutionalism of competition defence: the creation of weak competition bodies, and the establishment of weak or nonfunctioning judicial mechanisms that instead of assisting the commissions in applying competition law effectively neutralizes and weakens the law even more.

Application of competition policy obligates the state to intervene in transactions between private parties (or between private parties and the public sector) to preserve the competitive conditions of markets. Although diverse institutional models exist, the application of competition law, including the application of monetary sanctions, generally has been, or is to be undertaken by, an administrative body, either within the executive branch or in the form of an autonomous agency. In Central America, the exception is Panama, where the competition authority (ANDCC) may not apply sanctions until there is a court ruling. Other countries resolve cases through administrative channels, or they will once the law is passed. However, it will be possible to continue in this mode only if the parties found in violation of the law comply with their penalties. If not, the competition authority will resort to judicial channels once administrative options have been exhausted. Furthermore, if the sanctioned company feels that the penalty it has received is unfair, it may appeal or seek an injunction before the judicial system in Mexico and Honduras, thus overlapping the administrative and judicial systems. In some countries, once the competition authority makes its ruling, it may be appealed exclusively within the administrative system. At that point there are no more administrative channels and any successive litigation is handled by the court for administrative disputes. When the competition authority conducts its investigation, it may solicit information from the companies involved, and the majority of competition laws, or proposals for laws, grant the commission the power to demand this information; some even give the competition authority the power to carry out inspections in situ. Information provided by the companies is expected to be trustworthy, and if a company is found to falsify or omit any requested information, it may be subject to fines according to the laws of some countries. If these fines are not paid, competition authorities may seek redress through the judicial system. El Salvador imposes the lightest fines for this type of infraction, but it does impose them. In Mexico the provision of information is voluntary, but as the law is revised this will probably change, and in Panama, as discussed, the judicial branch intervenes from the start, issuing the instructions to provide information.

To the degree that competition authorities depend on judicial tribunals to carry out their functions, they face various obstacles to the effective resolution of their cases. Noteworthy among these are the following: (i) generally, judges are not prepared to handle cases of great technical complexity; (ii) the existence of administrative bodies that can issue legal rulings creates different options for the resolution of cases. In international practice at least three methods may be identified. (i) Perhaps the most common one holds that the administrative authority is just one party on an equal footing with the private party in the case, and in which the object of the dispute receives no special treatment. In consequence, cases are presented by the administrative body and are subject to all ruling bodies to which other legal cases would be subject. (ii) A second method recognizes the administrative authority's faculty to issue rulings, and in its capacity as a representative of the public interest, on certain differences with respect to the private party involved in the case. In this context, the role of the judicial system's review is reduced to a review of the due process carried out by the high court. (iii) A third method distinguishes between the administrative faculties of government bodies and autonomous agencies and the courts' faculties for issuing rulings, but recognizes the special characteristics of competition cases, and in consequence, creates special tribunals for them. The basis for this method is that the judges can become specialized in the issues relevant to competition cases, allowing them to manage the technical complexities involved. At the same time, this guarantees greater coherency and consistency in decisions and presumably allows for swifter resolution of cases. The countries in the study group with competition legislation and authorities use various methods. In Costa Rica, El Salvador and Honduras, cases brought before the competition authority are resolved by that same body, which has the faculty to apply penalties. As previously noted, parties have recourse to an appeal or reversal of the authority's final resolution. As such, once administrative channels are exhausted, the legality of the final resolutions may be disputed directly before the courts of administrative disputes.

In Panama, the National Authority for Consumer Protection and the Promotion of Competition (previously CLICAC) has the characteristic of being an investigative body that does not issue binding rulings. This agency only reports on evidence of any anti-competitive practices; if an investigation leads it to conclude that there are grounds for proving anti-competitive practices, it has the option of presenting its case before the courts. For this purpose, Panama created the Civil Circuit Courts,21 which are responsible for resolving cases resulting from the application or interpretation of the law regarding monopolies, consumer protection and bad-faith business practices. These courts were intended to overcome the difficulties faced by the ordinary courts in resolving highly technical cases and, at the same time, to ensure swift resolution. Few cases have been brought before these courts, however, and resolution of these cases has been extremely slow.22

21 In the 1970s, once new laws for business were defined, Panama debated whether these laws should be interpreted and ruled upon within the administrative system, in the traditional courts or in specialized courts. International organizations and the executive branch preferred that these issues be resolved administratively, as was taking place in the commercial arbitration tribunals of the Chamber of Commerce. In 1978, however, a decision by the Supreme Court regarding commercial exclusivity contracts determined that commercial issues should be taken up by the ordinary courts and should not be resolved through administrative channels. Following passage of the Industrial Property Law and the Competition Law, the volume of business justified the creation of the Specialized Commercial Courts (Fernández, 2005).

22 In part, this situation can be explained by the fact that during the period 1998–2002, these courts received 2,511 cases, of which 93% were suits regarding intellectual property rights (Fernández, 2005).

The Third Superior Court of Justice of the First Judicial District was also created and made specifically responsible for handling appeals of rulings handed down by the circuit courts.23 In practice, the authority brings a suit before the court either based on a complaint by third parties or ex officio in a summary oral hearing. The case is brought before the respective Commercial Court, which sets a preliminary hearing. Subsequently, an in-depth hearing is set for the oral presentation of evidence by any of the parties. Once the judge issues a decision, the decision may be appealed to the Third Superior Court. In the case of rulings above 750,000 balboas, the affected parties may seek a higher appeal before the First Hall of the Supreme Court of Justice, but when penalties are below this amount, no appeal is allowed (Fernández, 2005).

In Mexico, the Federal Competition Law allows for an appeal for reconsideration before the competition commission (Article 39). This appeal suspends execution of the resolution while the commission considers the appeal request (within 60 days). This first phase extends the period of action taken by the commission, leading to a second stage that begins once fines are issued. The competition law establishes fines that appear to be sufficiently significant to dissuade or punish parties engaged in anti-competitive practices, but the fines ultimately are practically unenforceable due to the many defence mechanisms available to companies. The principal problem results from the use of the writ of amparo, originally designed to protect individual guarantees as a means of resolving disputes within administrative law. Through April 2004, which marked the tenth year of the Federal Competition Commission, 6,666 cases were resolved, against which 636 amparos and 90 rulings of fiscal nullification were applied. Among these, as of the date indicated, 375 amparos had been resolved, with 260 still pending. Of these injunctions that were resolved, 35% was decided within one year; 76.5% was in less than two years and the remaining 23.5% in a period of three to five years. Of all the amparo filings, 284 were sought against procedural movements to protect companies from requests for information or allegations of responsibility for anti-competitive practices (Sánchez Ugarte, 2004).

The amparo presents various problems when applied to companies. It does not seem reasonable for a legal recourse aimed at protecting the life and safety of individuals to be used in commercial suits. Even more so when the accused in question are alleged by the Commission to be engaged in prohibited practices that gravely affect smaller companies, as is very frequently the case. Furthermore, the suit of amparo does not explicitly take into account the protection of the public interest. A second problem arises because of the very nature of the amparo, because the case becomes focused on the protection of guarantees with little attention paid to resolving the original dispute. We can cite a series of additional complications: the possibility that each and every subsidiary of the company obtaining the amparo also seek protection,

23 Panama also has municipal courts that handle only cases involving claims of 3,000 balboas or less on the part of the consumer. In the provinces or districts that do not have one of these courts, cases involving free competition and consumer issues are handled by the civil or mixed circuit and municipal courts.

and given that such processes are not cumulative, each of the subsidiaries' cases may be brought before different judges and courts and result in different and even contradictory rulings. Such complications are further magnified by the absence of any time limits for courts to issue decisions on such cases (Sánchez Ugarte, 2004).

How can these problems be addressed given that they severely limit the possibilities for effectively defending competition? In Mexico various measures have been proposed that tend to restrict the use of the amparo. These include combining all the cases brought over the same issue; the introduction of greater requirements for the granting of such an injunction24; that ruling on amparo requests take into account all the presumed misdeeds and issues of both form and substance25; requiring the courts to rule on such cases within a specific period of time; and assuring that the sentence be applicable to all economic entities. Several of these aspects have been incorporated into the proposal for revision of the competition law approved by Congress (mid-2006) but pending approval by the president.

During the discussion surrounding the elaboration and approval of the competition law in Honduras, one bill proposed that once administrative channels are exhausted, a case may proceed to the courts on the condition that the accused must already have paid the fines levied. The logic behind this proposal was to counteract actions meant to prolong the process, thereby undermining the effectiveness of the law. Nonetheless, this element was removed from the latest version of the proposal. The bill now stipulates in its Article 48 that once administrative appeals against the decisions of the commission have been exhausted, the only further recourse will be to seek an amparo before the Supreme Court of Justice, which could generate problems of the type mentioned in the case of Mexico.

The case of Chile is illustrative. Based on experience accumulated by the traditional system,26 Chile created the Court of Defense of Free Competition as a special and independent ruling body, the function of which is to prevent, correct and sanction actions deemed prejudicial to free competition. The Supreme Court is in charge of the directive, correctional and economic superintendency of this court. In this way, the court assures the existence of personnel who have been trained to

24 On this point it has been proposed that judges not grant an amparo when the ruling in question is against a monopolistic practice that substantially affects the public interest, and that in cases where an amparo might be issued, such a ruling could only come after the party seeking the injunction assures future payment of damages to the harmed party through a bond or other adequate guarantee (Sánchez Ugarte, 2004).

25 If this proposal was to be approved, as reasonable as it seems, it would definitively transform the entire concept of an amparo.

26 Until the reform of November 2003, the Chilean anti-monopoly system consisted of the National Competitiveness Enforcement Department (Fiscalía Nacional Económica), an administrative body and both prevention and resolution commissions. The system recognized the need for specialized personnel (in this regard, it was not constituted by judges but rather included economics professionals and government officials) but even as the Resolution Commission was presided over by a Supreme Court judge, its rulings were only subject to review by the country's high court, ensuring agile dispatch of cases. The Court of Defense of Free Competition recently replaced the two commissions mentioned.

tackle the difficult technical problems of competition legislation, consistency in rulings and, at the same time, swift resolution of cases, given that decisions may be appealed only before the Supreme Court. A formal complaint may be filed in the case of definitive rulings on measures included in Article 17 K,27 and such a request can be made either by the National Competitiveness Enforcement Department (Fiscalía Nacional Económica) or by any of the parties involved within a period of ten working days from the time of notification. In order to ensure swift processing, the law establishes that this claim be given preference over other issues. Furthermore, to assure termination of the anti-competitive conduct, the law establishes that the filing of a formal complaint will not suspend compliance with the decision except for the payment of fines.28 However, at the specific request of one of the parties and via a well-founded resolution, the court in which the claim is filed may totally or partially suspend the effects of the sentence.

V. Market Distortions and Cases Resolved by Competition Agencies

The objective of this section is to use examples to illustrate market distortions that occur throughout Central America and Mexico. National studies carried out as part of the project "Reinforcing Competition in Central America" by the International Development Research Centre (IDRC) and Economic Commission for Latin America and the Caribbean (ECLAC) reveal that in certain markets, similar anti-competitive practices repeatedly occur. Some of these practices are part of regional or even international business strategies, and others are exclusively domestic in scope.

Setting aside the telecommunications and banking sectors, which are developed in other chapters of this volume, sectors such as cement, rice, milling, beer and other alcoholic beverages, air transport, the poultry industry, slaughterhouses and beef are particularly anti-competitive, although some of these have recently undergone a positive evolution. In the following section we will analyse three markets: the milling industry, in which market distortions occur at the national level; the abuse of market

27 The measures included in the Article are the following: (i) to modify or put an end to the acts, contracts, pacts, systems or agreements that are contrary to the dispositions of the present law; (ii) to order the modification or dissolution of associations, corporations and other private corporate entities that may have intervened in the acts, contracts, pacts, systems or agreements referred to under the previous letter; (iii) to apply fines payable to the public fund of a sum equivalent to up to 20,000 annual tax units. The fines may be imposed upon the corresponding corporate entity, or its directors, administrators and anyone who may have participated in the conduct of the respective act. In the case of fines applied to corporate entities, their directors, administrators and anyone else who has benefited from the respective acts must be responsible for payment, as long as they have participated in the conduct of these acts' (own translation; Republic of Chile, 2005).

28 To file a formal complaint in cases in which a fine is imposed, the guilty party is required to post a sum equivalent to 10% of the fine.

power by the airlines at the regional level; and anti-competitive practices in the soft drink industry in which global companies are involved.

1. The Wheat Flour Market

The flour market is of great concern to governments for its wide-ranging social impact on the economy (representing 7% of the basic food basket in Panama; Fernández, 2005).

This sector has been analysed in the cases of El Salvador (Molina, 2007), Nicaragua (Ansorena, 2007) and Panama (Fernández, 2005), but the problem also exists in countries such as Costa Rica and probably in the rest of the region. In the first half of the decade of the 2000s, the problem of collusion among flour millers became more evident and severe as the price of wheat rose on international markets, affecting the price of flour and subsequently of bread at the national level. In Nicaragua, for example, between 2001 and 2004 the price of flour rose 55% (in fact, there was a bread "crisis" in 2001 in this country).

Generally, the flour market has a monopolistic or oligopolistic structure in which the key players also often retain control over flour imports and distribution (creating distortions such as market segmentation). Some countries even depend on imported wheat for domestic flour milling.

In El Salvador the milling industry is duopolistic: Molinos de El Salvador (MOLSA) and Harinas de El Salvador (HARISA) control the three phases discussed. In Nicaragua there are only three flour millers, and they maintain very close ties to the two flour importers Proharina and Fhacasa.

In Panama, the problem of the millers became one of the first cases to be investigated and prosecuted by the competition authority (CLICAC) in the mid-1990s. The authority brought suit before the courts against four companies in the country for collusion in fixing the price of wheat flour and in maintaining their respective market shares for wheat flour sales, for which they exchanged relevant information between November 1996 and September 1997. The case was recently decided in favour of CLICAC.29

Additionally, some countries erect entry barriers on flour imports by requiring compliance with specific regulations or norms. Such is the case of Resolution 94-2002 approved by the Council of Ministers for Economic Integration (COMIECO) in October 2002 for El Salvador, Guatemala, Honduras and Nicaragua, which accepted the technical norm for fortified wheat flour, Decree No. 30809, published

29 The sentence determined that on March 8, 1994 the millers had signed an agreement to set wheat prices and divide up the market estimated at that time at 120,000 quintales (1 quintal = 46 kg) per month. The defendants presented a high court appeal before the Supreme Court of Justice which was rejected as inadmissible; they have subsequently presented a second appeal which is awaiting disposition by the court (Fernández, 2005).

in Gazette No. 216 on November 8, 2002. This counteracted to a certain extent the effects of the liberalization of trade in wheat flour decreed by the Executive Committee of Economic Integration (CEIE) in October of the same year. Regardless, this latter measure represented an advance in the regional free trade of wheat flour reflected in expanded trading volumes for this product in Central America. Nicaragua, the country that most protected its milling industry with a 35% tariff until 2004, was one of the most favoured by this agreement. Nonetheless, there is a certain degree of differentiation in this product area, and even if imported flour is cheaper in Nicaragua, it is also of inferior quality, which has segmented the market. A significant price drop was by no means felt in all countries (Molina, 2007).

Given that the largest consumers of flour are the bread bakers, and the relatively high price of this input raises the price of this basic foodstuff in the region, some governments have removed import tariffs on flour from third-party countries such as Mexico. This has provided producers with a source of flour at competitive prices. Nicaragua's MIFIC agreed to liberalize flour imports by bakers and even extended to them a preferential line of credit for such imports. Nonetheless, in El Salvador as well as Nicaragua, it is difficult for bakers to import their inputs directly due to their small size and the lack of an established custom of taking advantage of external supply sources. As a result, the majority of their bakers continue to source flour in their domestic markets at monopolistic prices. One promising sign is that in July 2006 a new association of bakers in Nicaragua made its first direct import of flour.

2. The Air Transportation Market in Central America30

There has been little competition for passengers in the market for intra-Central American air transportation due to the extent to which the market is concentrated in the hands of Transportes Aéreos del Continente Americano (Air Transport of the American Continent; TACA) airlines since the early 1980s. This Salvadoran-owned company formed Grupo TACA, a business alliance that includes the participation of various Central America airlines, including Líneas Aéreas Costarricenses, S.A. (LACSA); Taca International Airlines, S.A.; Taca de Honduras, S.A. de C.V.; Aviateca, S.A.; and Nicaragüense de Aviación, S.A. (NICA). As such, practically all the region's national airlines belong to this conglomerate except Panama's Compania Panamena de Aviacion (COPA).

Midway through 2006, TACA was operating flights to 19 cities in Central America and the USA via its operations hub in San Salvador. TACA's principal route is North America – Central America, but its profit margins are particularly high on intra-regional flights. Outside of the region, the competition is greater and the company is not able to apply the same strategy.

30 This section is partly based on Molina (2007).

Central American airspace is protected through an intra-regional "open skies" agreement, but operators from outside the region are not accorded the same treatment. As a result, US and South American airlines are not permitted to operate flights between points within the region. Put another way, no airline other than TACA can pick up passengers in a Central American country and take them to another destination within Central America. In contrast, the market for North America – Central America routes is highly competitive.

Beginning in 1998, the regional market became somewhat more competitive due to an open-skies agreement signed between some Central American countries and the USA, allowing for the entry of new competitors into the market. Nonetheless, Grupo TACA continues to engage in monopolistic practices. For example, Panama's COPA airline requested and was granted, in December 2004, permission to operate additional flights on its routes from Central American countries to destinations in South America, Panama and the Caribbean. TACA filed several legal challenges to COPA operating a second daily flight on these routes, and on several occasions has succeeded in denying the Panamanian airline permission to land in El Salvador.31 Such anti-competitive conduct has had adverse effects across the region for consumers and airlines alike.

3. The Market for Carbonated Soft Drinks

Anti-competitive practices occur frequently in this market and have been challenged before competition commissions in Costa Rica and Mexico. However, these are practices that some multinational corporations, principally Coca Cola, engage in throughout many countries, including within the European Union.32 Such conduct consists of requiring both large and small-scale retailers, and even stadiums, amusement parks and other types of businesses agree to sell only Coca Cola brand beverages. This barrier also leads to the brand being sold at prices higher than those of alternative brands of carbonated soft drinks, as was the case when Big Cola began to compete with Coca Cola in Mexico.

In order to supply their product to retail resellers, Coca Cola frequently demands the exclusivity described above. This requirement may be applied in direct or indirect forms. For example, the Coca Cola Company or an affiliated bottler may provide the retailer with a small refrigerator in exchange for the commitment to stock it only with Coca Cola products. If the retail location is small enough that no other refrigerator of this type can fit, this guarantees that no other brand of cold beverages will be sold there. It is important to note that the small family stores that face these monopolistic practices are the channel for 75% of Coca Cola sales in Mexico, while

31 [online] skyscrapercity.com/archive/index.php/t-295767.html

32 Ibid.

restaurants, schools, clubs, hotels and entertainment facilities represent 24% of sales, and supermarkets only 1%.33

Consumption of this type of beverage in Mexico is enormous. By the end of the 1990s, 10% of the Coca Cola Company's global earnings was being generated in Mexico and, in 2004, this country was the sales location for 11% of the 19,800 million unit cases of soft drinks sold by the company. In addition, the Coca Cola Company regularly enters markets for related products seeking to replace domestic brands through large investments in advertising and promotions. This is the situation faced by firms producing fruit drinks, juices and nectars in Mexico. Coca Cola will use its Minute Maid brand to challenge Jumex, Jugos del Valle and the worker's co-operative Pascual Boing for control of this market.34 Special attention is needed to assure that the practices that shaped the carbonated soft drink market do not extend worldwide to that of juice, fruit drinks and nectars.35

Charges of monopoly practices in the sale of certain products were filed before competition officials in Costa Rica in 2001,36 and Mexico's FCC received several such complaints between 2000 and 2003. In both instances heavy fines were levied on the company accused of conducting these practices.

VI. Conclusions

There were no instruments for promoting market competition in Central American countries at the time when governments in the region carried out most of the economic deregulation process and privatizations of large public companies, which in many cases began to operate as private monopolies. Such monopolies and other market distortions that accumulated since the economic reforms posed a major challenge for the competition agencies once they were formed (in five of the seven countries studied, including Mexico, by the end of 2006), especially in the smaller economies. The countries that still need to create the legal and institutional framework for competition will face these problems perhaps to an even greater degree, but they also will be able to learn from the mistakes and successes of their neighbours.

Competition is now on the agenda of governments of most countries in the region. Competition policy has greatly expanded in recent years and between 2004 and 2006 not only have two new laws been passed (in El Salvador and Honduras),

33 [online] www.rel-uita.org/companias/coca-cola/cosecha-para-femsa.htm

34 [online] www.crain.com.mx/Snews/news_display.php?story_id=633

35 Ibid.

36 The suit in Costa Rica was against Coca Cola and Embotelladora Panamco Tica S.A. on the grounds that these companies engaged in relative monopolistic practices. In June 2001, the company Embotelladora Centroamericana S.A. brought suit against Embotelladora Panamco Tica S.A. for the same reasons. The latter company, one of the largest in Latin America, commonly imposes prices on the businesses that sell the products they distribute (Sittenfeld, 2006 and her chapter in this book).

but the pioneers in competition policy (Costa Rica, Panama and Mexico) have reformed or are in the process of reforming their laws, as we have previously noted. The trend is towards expanding the coverage or applicability of competition law by reducing the number of exceptions, incorporating the "clemency appeal" in order to have more channels for access to information on monopolistic practices and increasing fines to give them more dissuasive influence over potential transgressors. In addition to modifying the laws, these countries have also made efforts to improve training for judges and seek closer relationships and collaboration between competition agencies and the judicial systems. The countries that have recently passed their laws – El Salvador and Honduras – have already incorporated some of these improvements.

After at least ten years of experience in competition policy in three of the countries under study, we can see that the legislative model based on that of the USA, in which the courts play a major role in case resolution, does not appear to be the most adequate for countries such as those of Central America which lack expertise on both the judicial and competition-policy arenas. This is the situation in Panama, which even created special courts for cases that are often handled in a very protracted and unsatisfactory manner. It seems that a European-style system, in which cases could be handled through administrative channels, would be much more effective. Furthermore, there are numerous ways companies engaging in anti-competitive practices may avoid or delay the payment of fines, and such loopholes for legal manoeuvering must be eliminated to make competition laws more effective.

The lack of culture of competition is a prominent problem in Central American countries and a concerted effort to address this issue is indispensable. An effective application of competition law to anti-competitive practices, however, is an important means of legitimizing and promoting the knowledge and respect for the institutional framework needed for guaranteeing the healthy functioning of markets.

The relationship between regulation and competition is very important in this type of economy, and they should complement each other so that goods and services, especially of types previously provided by the state and now dominated by large corporations, can be provided in a technically adequate manner.

In small economies, the competition agency faces a special challenge in the case of market concentrations, given that these may be necessary in order to achieve important technological innovations and increased efficiency. A major challenge for the countries under study is how to distinguish such objectives from that of increasing profit margins through monopolistic practices. In this regard the competition agencies of Mexico, on the one hand, and Costa Rica and Panama, on the other, vary in their capacity to deal with this issue. In Mexico, the majority of cases are related to corporate mergers and takeovers, transactions for which participating companies must give regulatory officials prior notification. In the other two countries, where notification of merger or acquisition plans is voluntary, competition agency efforts are focused largely on other types of practices (generally horizontal monopolistic practices). El Salvador is the first SDC in Central America to require prior notification of mergers, and its competition commission will face challenges in carrying out its work, considering the scarce resources available to it.

As we have seen, competition policy has advanced considerably in recent years in the countries under study, but at the same time the further advances have been registered in competition policy at the international level, thereby posing new challenges for the region. For example, the local trend has been to apply the "rule of reason" instead of prohibiting certain practices per se, even when the practices are horizontal in nature. Even if this way of treating cases is undoubtedly more reflexive and probably leads to fairer resolutions, it poses the need to provide SDC with greater resources. It is also important to maintain a broad view of the changes experienced by these countries as a result of their insertion into global markets. They all have strengthened commercial ties through free trade agreements (FTAs) between each other or with third-party countries and this can open the door for large multinational corporations to take advantage by exerting their market power in these countries, or for other types of distortions to occur although improvements in competition also may take place in some markets as a result of these agreements. Hence, many FTAs include competition clauses.

In this context, regional and international cooperation on competition is crucial within Central America, and between the region and those countries with which it has strong economic ties such as the USA and Mexico. The possibility of collaborating on information sharing and coordinating actions to avoid anti-competitive practices that simultaneously affect various markets of the region (e.g. the cement, beverage, air transportation and banking industries, among others) would be of great help. This becomes even more significant when considering the limited resources available to all of the regions' competition agencies. It would also be very useful to make a great effort to harmonize competition rules at the regional level, create similar criteria for confidentiality and transparency of information and pursue cases of trans-border monopolistic practices. A seed has been sown for taking up this effort in the Guatemala Protocol of the General Treaty of Central American Economic Integration (1993), which includes the following stipulation: "In the trade sector, the states party to this agreement agree to adopt common dispositions to avoid monopolistic activities and to promote free competition in the countries of the region."

Bibliography

Adhikari, R. and J. Knight (2004), What type of competition policy and law should a developing country have? South Asia Economic Journal 5: 1–25.

Ansorena, C. (2007), Competencia y regulación en las telecomunicaciones: el caso de Nicaragua, (LC/MEX/L.785), serie Estudios y Perspectivas, No. 73, ECLAC/IDRC, July.

Avalos, M. (2006), "Condiciones generales de competencia: el caso de México", Serie Estudios y Perspectivas No 48, ECLAC/IDRC, Mexico, May.

Baumol, W.; J. Panzar and R. Willig (1982), Contestable Markets and the Theory of Market Structure, Harcourt Brace Javanovich, New York.

Cernat, L. and P. Holmes (2004), Introduction, competition, competitiveness and development: Lessons from developing countries, in P. Brusick, A. M. Álvarez, L. Cernat, and P. Holmes, United Nations Conference on Trade and Development (UNCTAD), New York and Geneva, June.

De León, I. (1998), Un enfoque alternativo de políticas para la promoción de la competencia en economías en transición, Boletín Latinoamericano de Competencia, N° 5 (third part), August, pp. 50–72 [online] (http://europa.eu.int/comm/competition/international/others/latin_america/boletin/boletin_5_3_es.pdf).

De León, I. (2000), The role of competition policy in the promotion of competitiveness and development in Latin America, Boletín Latinoamericano de Competencia. N° 11. December, pp. 57–72. [online] (http://europa.eu.int/comm/competition/international/others/latin_america/boletin/boletin_5_3_es.pdf).

De León, I. (2006), La promoción de la competencia en Centroamérica. Dificultades y posibilidades, in C. Schatan and M. Avalos (Coords.), Condiciones y políticas de competencia. Economías pequeñas de Centroamérica y el Caribe, FCE and ECLAC, Mexico D. F.

Evenett, S. (n.d.), "Competition policy in small developing countries: reflections on current literature", Power Point presentation.

Fernández, M. (2005), Condiciones generales de competencia en Panamá, Serie Estudios y Perspectivas, N° 37, ECLAC/IDRC, Mexico, September.

Gal, M. (2001), Site does matter: The effects of market site on optimal competition policy, Southern California Law Review, vol. 74, Los Angeles, CA.

Gal, M. (2003), Competition Policy for Small Market Economies, Harvard University Press, Cambridge, MA.

Gal, M. (2004), Competition: Prerequisites for development-oriented competition policy implementation, competitiveness and development: Lessons from developing countries, in Brusick (ed.), United Nations Conference on Trade and Development (UNCTAD), New York and Geneva.

Hernández, R. and C. Schatan (2006), Políticas de competencia y de regulación en el Istmo Centroamericano, in C. Schatan y M. Avalos (Coords), Condiciones y políticas de competencia. Economías pequeñas de Centroamérica y el Caribe, FCE y ECLAC, Mexico.

Khemani, R. S. (1997), Competition policy and economic development, Policy Options, October [online] (http://www.irpp.org/po/archive/oct97/khemani.pdf).

Khemani, R. S. and M. Dutz (1995), The instruments of competition policy and their relevance for economic development, in Frischtak (ed.), Regulatory Policies and Reform: A Comparative Perspective, (Pre-publication Edition), World Bank, Washington, DC.

Klemperer, P. (1995), Competition when consumers have switching cost: An overview with applications to industrial organization, macroeconomics, and international trade, Review of Economics Studies 62, 515–539.

Levy, S. (2005), Ensayos sobre desarrollo económico mexicano, Fondo de Cultura Económica, Mexico.

Levy, B. and P. Spiller (1996), Regulations, Institutions and Commitment, Comparative Studies of Telecommunications, Cambridge University Press.

Mariscal, J. and E. Rivera (2005a), New trends in the Latin American telecommunications market: Telefonica & Telmex, Telecommunications Policy, N° 29, pp. 757–777.

Mariscal, J. and E. Rivera (2005b), Organización, industrial y competencia en las telecomunicaciones en América Latina: las estrategias empresariales, serie Desarrollo Productivo N° 169 (LC/L.2423-P), ECLAC, Santiago, Chile.

Mariscal, J. and E. Rivera (2007), Regulación y competencia en las telecomunicaciones mexicanas, serie Estudios y Perspectivas N° 83, ECLAC/IDRC, June, Mexico.

Molina, F. (2007), Condiciones generales de competencia en países centroamericanos: el caso de El Salvador, Serie Estudios y Perspectivas N° 67, ECLAC/IDRC, Mexico, January.

Romero, A. and C. E. González (2006), Condiciones generales de competencia en Guatemala, serie Estudios y Perspectivas, N° 52, ECLAC/IDRC, Mexico, May.

Sánchez Ugarte, F. (2004), Análisis económico de la legislación federal de competencia económica en Competencia Económica F., Editorial Porrúa, Mexico.

Sánchez Ugarte, F.; P. García Alba Iduñate; J. Aguilar; F. Heftye; G. Moguer and S. Sarmiento (2004), La primera década de la Comisión Federal de Competencia, Comisión Federal de Competencia (CFC), Mexico.

Schatan, C. and M. Avalos (2006), Condiciones y políticas de competencia. Economías pequeñas de Centroamérica y el Caribe, FCE and ECLAC, Mexico.

Singh, A. (2004), Multilateral competition policy and economic development. A developing country perspective on the European community proposal, Series on Issues on Competition Law and Policy (UNCTAD/DITC/CPL/2003/10), Geneva.

Sittenfeld, P. (2006), La puesta en marcha de la agencia de competencia; algunas lecciones aprendidas del caso costarricense, PowerPoint presentation [online] (http://www.eclac.cl/mexico/competencia/index.htm).

Stewart, T. (2006), Evaluación de la relevancia de las leyes de competencia en países pequeños: el caso del Caricom, in C. Schatan and M. Avalos (Coords.), Condiciones y políticas de competencia en economías pequeñas de Centroamérica y el Caribe, FCE and ECLAC, Mexico.

Tábora, M. R. (2007), Condiciones generales de competencia en Honduras, serie Estudios y Perspectivas No. 82, ECLAC/IDRC, Mexico, May.

Yong, M. (2005), Condiciones generales de competencia en Costa Rica, ECLAC/IDRC, Mexico, unpublished.

This page intentionally left blank

Chapter 3
Models of Privatization and Development
of Competition in Telecommunications
in Central America and Mexico

Eugenio Rivera

Introduction

Telecommunications in Central America and Mexico have undergone unprecedented changes. The coverage attained in fixed telephony, and particularly in mobile telephony, was unimaginable just a few years ago. Despite this general trend, the countries of the region have taken varied paths in the privatization and liberalization of telecommunications, which have resulted in different levels of competition and performance in the industry.

The purpose of this chapter is to analyse, based on national studies on competition and regulation in telecommunications, how the different forms of privatization and liberalization of this industry or, conversely, the preservation of public enterprise have affected the intensity of competition in the industry, the coverage and the quality and price of services.

For this purpose, first we analyse the different forms of privatization and liberalization of telecommunications or their rejection, seeking to identify the causes leading to the application of the various models. Three courses of development can be pieced together in the region.

The first concerns the case of Costa Rica, in which the citizens rejected privatization and opted for keeping the Costa Rican Electricty Institute (ICE), also in charge of telecommunications, as a public enterprise. Honduras shows significant similarities with this course of development. The second involves the cases of Mexico, Nicaragua and Panama, where the strategy was to privatize and provide the enterprise with a period of exclusivity. This policy aimed at strengthening the enterprise prior to the liberalization of the market. The third approach was that adopted by El Salvador and Guatemala which was characterized by the simultaneous privatization and liberalization of the industry and also by the privatization of radio spectrum management.

Among the major options mentioned, privatization is considered by the literature an important factor for the sector's performance.1 Whether the monopolistic public enterprise was privatized as a vertically and/or horizontally integrated monopoly, or

1 On this subject see Rivera, 2006.

whether measures were taken so as to limit its monopolistic power through restructuring, or if certain precautionary steps were taken to limit its dominant position seems to have been crucial for the enterprise's behaviour and this will be looked into in this chapter. These different paths are the result of specific political configurations and coalitions and diverse types of historical performance by the public enterprise. The international context in which the different processes took place also had a strong influence. Thus, the Mexican reform can only be understood within the context of approval by the North American Free Trade Agreement (NAFTA), and the final destiny of ICE in Costa Rica will probably be determined by the Dominican Republic – Central America Free Trade Agreement (DR-CAFTA).

Together with the forms of privatization and liberalization, the industry's performance is associated with the existence of a legal framework favouring competition and effective conditions for its application. This latter aspect refers to the institutional capacity of both the sector regulator and the competition agency to effectively perform regulatory tasks and supervise competition, and their respective capabilities for coordination. It also refers to the ways in which the overall institutional framework operates, and particularly the administration of justice. The regulatory and competition agencies, in addition to undertaking tasks incumbent upon the Executive Branch such as administration of the radio spectrum and formulation of telecommunications and competition policies, also operate as administrative entities assisting in the administration of justice.2 Paradoxically, the Judiciary sometimes stands as a significant obstacle to the application of those policies, thus affecting competition and market efficiency.

In this context, Section I compares the different privatization and liberalization processes as well as their main determining factors. Section II analyses the main features of the regulatory and competition institutions, as well as their legal framework. The general business climate and its impact on a competitive atmosphere are also studied in this section. Section III studies the processes of opening up to competition, a crucial time for assessing the consistency of the regulatory framework and its application. Section IV focuses on the characteristics of the sector's industrial organization and the performance and development of competition in the industry's different segments. Section V analyses the evolution of rates in the countries of the region. Section VI identifies the trends influencing the future of telecommunications in the region and sets forth the main conclusions together with policy recommendations.

I. Privatization and Liberalization of Telecommunications in Central America and Mexico: The Diversity of Experiences

With the exception of Costa Rica, the seven countries covered by this study have furthered forms of privatization and deregulation of telecommunications. However, these forms of privatization and liberalization of the sector have varied.

2 We wish to thank Gustavo Paredes, while being a Commissioner of CLICAC, for having called our attention to this link.

1. The Case of Costa Rica

Costa Rica has a long experience in open trade practices and in the establishment of free trade agreements. Recently it took part in the negotiations leading to the signing of DR-CAFTA, although it is yet to be ratified by the Legislative Assembly. Even though the main political forces have expressed their support for this agreement, citizens' rejection persuaded the President of the Republic to postpone seeking its ratification by the said Assembly. The chief reason for this rejection is the commitments assumed by Costa Rica in telecommunications in Annex I.

In DR-CAFTA, Costa Rica pledges to promote the opening up of telecommunications services gradually and selectively, and strictly in accordance with the social objectives of universality and solidarity in the provision of such services. To that end, Costa Rica would have had to enact, by December 31, 2004, at the latest, a legal framework that would allow the strengthening and modernization of ICE.3 Likewise, by virtue of the Agreement it would have to allow other providers to supply telecommunications services under no less favourable terms and conditions than those established by, or granted in accordance with, its legislation in force to January 27, 2003. Costa Rica would not have permitted other telecommunications services providers (including international) on a non-discriminatory basis to compete in order to supply directly to the client, with the technology of its choice, the following telecommunications services: (i) private network services, at the latest by January 1, 2006; (ii) Internet services, at the latest by January 1, 2006; and (iii) mobile wireless services, at the latest by January 1, 2007. Finally, the country would have to apply the new regulatory framework for telecommunications services as of January 1, 2006, in accordance with the following provisions: universal service, independence of the regulatory authority, transparency, allocation and utilization of scarce resources, regulated interconnection, access to, and use of, information services supply networks, competition, undersea cables systems and flexibility in technology options.

The DR-CAFTA negotiations were accompanied by intense public debate and social conflict. The debate continued after its signing in January 2004. The situation was further complicated when proceedings were instituted against two ex-presidents involved in cases of corruption in connection with the Social Security Fund and ICE, probably the two leading institutions of the so-called Costa Rican model. These cases appeared to confirm all citizens suspicious who viewed the different initiatives for modernization of the public electricity and telecommunications enterprise only as attempts by privileged groups to appropriate, at a low price, the national wealth.4 The early start of the campaign to elect a new president in

3 In February 2006 presidential election, the candidate of the Citizen's Action Party actively opposed this Free Trade Agreement. Nevertheless, the winning candidate, Oscar Arias, supported such agreement all along.

4 For an analysis of the different proposals for reform of ICE, see Rivera (2006).

February 2006 led the authorities to postpone the parliamentary debate on the Agreement until after the elections.

Costa Rica is the only country among those studied5 that has rejected the privatization and liberalization of telecommunications. The main reasons for this stance are as follows. (i) The public electricity and telecommunications enterprise has been associated with the most positive aspects of the Costa Rican model, which enabled a relatively poor Central American country to emerge as the most developed in the region. (ii) Associated with the above, ICE took electricity and telecommunications services to every corner of the country and achieved higher levels of coverage than most Latin American countries. While in the rest of the region users had to wait years for access to a telephone, Costa Ricans were able to gain access so much more quickly. (iii) ICE has also played a crucial role in helping the country reach levels of competitiveness that have enabled it to participate favourably in the most modern international chains of production. National Telecommunications Institute's (INTEL) investments are one of the main examples of this situation. (iv) After the privatization and liberalization of telecommunications in the rest of Central America and Mexico, fixed telephony in Costa Rica remained at the head of the industry in terms of coverage, and was placed second in Latin America.6

2. Privatization with Exclusivity: The Cases of Mexico, Nicaragua and Panama

In the cases of Mexico, Nicaragua and Panama, the privatization process was accompanied by the right of an exclusivity period for the telecommunications enterprise, prior to the opening up of the market to competition. The similarities among these experiences, however, conceal significant differences in telecommunications public policy.

Mexico's case is undoubtedly a paradigmatic example of nationalism in the process of privatization and opening up of telecommunications. The enterprise was privatized as a national, vertically integrated monopoly and was granted a six-year exclusivity period. The new General Law of Telecommunications (LGT) was passed more than five years after privatization and the regulatory body was established shortly before the end of the exclusivity period. The privatization of telecommunications in Mexico formed part of a broader process of reform begun in 1986 with the accession to General Agreement on Tariffs and Trade (GATT) and the government's decision to establish a free trade agreement with the USA and Canada. Despite opposition from large social groups, the hope that this policy would contribute to reactivating the economy and to its

5 The group of countries includes Costa Rica, El Salvador, Guatemala, Honduras, Mexico, Nicaragua and Panama.

6 The only other country in Latin America that has rejected privatization of its telecommunications enterprise has been Uruguay.

long-term growth succeeded in rallying majority support in society. The poor performance of the public telecommunications enterprise constituted a forceful argument in favour of the reform and, at the same time, privatization became a fundamental component of the modernization of the economy (Mariscal and Rivera, 2005a).

Telecommunications policy in Mexico cannot be understood unless it is inserted into the context defined by the decision of the administration of Salinas de Gortari (1988–1994) to reorient Mexico's development from the import substitution model towards an export model based mainly on close association with the USA and Canada. Betting on NAFTA led the Mexican government to use the privatization of Teléfonos de México (TELMEX) as evidence of its commitment to approval of the Agreement. With the privatization of the enterprise the Salinas administration wanted to show its split with the national revolutionary tradition and its willingness to further private sector-driven capitalism. Government policy was also determined to privatize because of its fear that the telecommunications enterprise would suffer a rapid process of denationalization or become a victim of competition from major foreign operators, particularly from the USA. In this context, the enterprise was strengthened financially as much as possible so as to become attractive to domestic investors (Mariscal and Rivera, 2007). The decision was taken, moreover, to privatize the enterprise as a vertically and horizontally integrated monopoly and it was decided that the dominant share among the investors should be Mexican.7 The exclusivity period paved the way for an important set of conditions for the modernization of the enterprise, digitalization of its networks and financial strengthening which stood it in good stead to face the generalized opening up of the industry to competition as of 1997. Together with tax benefits, the tariff policy set in the Concession License was highly favourable to the enterprise, allowing high rates of return, which meant that in 1999 Mexico registered the Organisation for Economic Co-operation and Development's (OECD) highest prices for telecommunications services (OECD, 1999). This policy enabled the company to begin its international expansion in 1998 and since then it has become, together with Telefónica España, one of the two dominant companies in the region.

If in Mexico's case the exclusivity period and other policies applied in favour of TELMEX can be explained by the aim of contributing to shape a national champion capable of vying with US companies for the position of privilege within Mexican borders, in the cases of Panama and Nicaragua this policy is harder to understand.

In Panama, on July 14, 1992, the Legislative Assembly, controlled at the time by the Christian Democrat Party, succeeded in passing a law "whereby the privatization process for state enterprises and services is established and regulated." However, the opposition introduced an Article which indicated that the law would not be applicable to the privatization of public utility enterprises, INTEL, Institute of Water Resources and Electrification (IRHE) and National Aqueducts and Sewerage

7 The concession licence obliges TELMEX to maintain its capital structure and Board of Directors, so that the power to determine the company's administrative control and management should fall chiefly on the Mexican partners (Mariscal and Rivera, 2007).

Systems Institute (IDAAN), or to the services they provide, and that a special law to approve their privatization would be required. Paradoxically, one of the parties that authored the above-mentioned Article took office in government in 1994 and proposed a bill, passed in 1995, paving the way for the privatization of INTEL.8 The immediate causes of privatization appear to have been problems of supply in fixed telephony, the non-existence of mobile services and the fiscal problems of the Panamanian State, which made it difficult for the state to carry out the necessary investments in this sector. (González, 2007).

The opening of envelopes containing the companies' offers took place on May 20, 1997. The company Cable & Wireless won the bid for the enterprise by offering $652 million as against General Telephone and Electronics' (GTE) $451 million. The benchmark price had been established at $500 million, so $152 million above that price was obtained. With regard to mobile telephony, in 1996 an invitation for tenders was issued for Band A for the provision of mobile services throughout the country. The winning bid was from Bellsouth, in alliance with Multiholding,9 which paid $72.6 million. Motorola, for its part, offered only $42 million. Band B was assigned to the winner of the bid for INTEL (which was Cable & Wireless, as mentioned), which would have to pay the same amount paid for Band A (González, 2007).

What are the reasons behind the Panamanian government's decision to grant an exclusivity period in fixed telephony, award a mobile licence as part of the sale of the fixed telephony enterprise and limit mobile telephony service to two operators? Three explanations have been put forward in the debate on the matter: (i) the presence of natural monopoly in fixed telephony; (ii) the government's objective of maximizing its revenues as owner of 49% of the shares led it to turn over the sector in exclusivity – The present value of cash flows is greater under that system; (iii) radio spectrum restrictions did not make the entry of more operators viable (González, 2007).

The first argument may be convincing to justify the sale of the enterprise as a monopolistic operator. It is not, however, the grounds for a period of exclusivity. Exclusivity is justified precisely as a period in which the enterprise renews itself and reaches a critical threshold that makes it possible for it to face the competition. But the characteristics of a natural monopoly rule out any attempt at entry of new firms, and therefore additional barriers seem unnecessary. The second argument is undoubtedly the relevant one. The concession of exclusive rights allowed the government to maximize its revenues at the time of sale and, as owners of 49% of the shares, profit from a greater flow of dividends. The fiscal reasons do not, however, seem sufficient to explain the continued public presence in the ownership of the enterprise. In this regard, the Democratic Revolutionary Party (PRD), the dominant group in the coalition that assumed the Presidency of the Republic in 1994 and was characterized by strong nationalistic tendencies, saw in public participation in the

8 Ernesto Pérez Balladares, of the United People Alliance, headed by the PRD, obtained 355,307 votes of a total of 1,066,844 valid votes. The PRD is the party founded by the revolutionary national leader Torrijos.

9 The shareholding was Bellsouth USA 43.7% and Multiholding Corporation 56.3%.

company INTEL a mechanism to ensure consistency between economic growth objectives and modernization of the enterprise and national interests. The Panamanian government's interest is reflected by the fact that eight years after 51% of the enterprise's shares were privatized, the Panamanian State continued to hold 49% of the shares. A further expression of this situation is the existence of ongoing negotiations between the government and the enterprise to obtain lower prices for public institutions than those paid by other users of telecommunications services (González, 2007). In this regard, Panamanian policy in privatization and liberalization of telecommunications may be interpreted as a version of the policy of national champions in the case of small countries.

Nicaragua's case also has its peculiarities. It is the region's poorest country, with long-standing high external indebtedness resulting from having to finance its current account deficits and a substantial percentage of its fiscal deficits with international aid or concessionary loans. This situation makes the country highly vulnerable to international pressures in favour of privatization. Privatization of the public telecommunications monopoly was put forward as a solution to the problem of demands for indemnity on properties confiscated by the Sandinista government. The funds obtained would be allocated to finance the bonds issued in favour of persons with unfairly confiscated properties. The proposal thus succeeded in building political consensus on privatization of the industry (Ansorena, 2007).

The privatization process was preceded by a significant investment effort. Between 1992 and 1994, Nicaragua achieved the highest rate of investment and modernization of telecommunications in the whole of Central America, so that the assets of the public operator were the most modern in the region at the end of that period. Investments during that time span totalled $103.9 million, 72% of which was earmarked for the acquisition of infrastructure and equipment. As in the two preceding cases, a horizontally and vertically integrated monopoly was privatized, with an exclusivity clause for a period of three years. The exclusive concession also included national and international long distance (LD) and the option to acquire a mobile telephony licence. In its part, the winning company had to pledge to meet network expansion goals, improvement of service quality and creation of a technological platform (essential equipment) to allow interconnection and thus increase the participation of other operators once the exclusivity period was over. The privatization model included the sale of 40% of the shares, with majority voting power equivalent to 57%; employees could have access to 11% and the remainder 49% would remain as property of the state for subsequent sale on the domestic or international stock market. A number of difficulties postponed the privatization until 2001.10 Finally, on December 12, 2001, after a public bidding process, 40% of the

10 Political difficulties and strong public opposition to the privatization of the only profitable public services enterprise postponed approval of the sale by the National Assembly until 1998. The first competitive bidding process begun in May 1999 was declared void when the two interested companies, TELMEX of Mexico and Telefónica of Spain, decided to withdraw. The second attempt, six months later, had the same outcome, when a consortium made up of MCI WorldCom, TELMEX and France Telecom withdrew from the process (Ansorena, 2007).

enterprise's shares were turned over to the Swedish-Honduran consortium Telia Swedtel Ab – MEGATEL EMCE for a period of 20 years. The price paid was $83 million, far below the benchmark price of $203 million (Ansorena, 2007). In 2004, in contrast to the above case, the government proceeded to sell the remaining shares, awarding them without a bidding process to the Mexican company América Móvil, part of Grupo Carso. In parallel, the same company acquired MEGATEL of Honduras, as a result of which América Móvil became the sole shareholder of Empresa Nicaraguense de Telecomunicaciones, S.A. (ENITEL). Furthermore, with the purchase of MEGATEL, América Móvil became an operator with two cellular telephony licences, one from ENITEL and the other from Servicios de Comunicaciones Personales Inalambrica (PCS SERCOM).

If in Panama's case the exclusivity policy expressed an interest in maintaining a public presence in the sector, in Nicaragua's case it represented only a concession to the investor. Protecting an international telecommunications operator does not appear to make sense from the point of view of the public interest. On the other hand, the sale of the enterprise to the Mexican operator, which enabled the latter to strengthen its position in the fixed telephony segment and also in national and international LD and, at the same time, made it easier for that same company to have majority control of mobile telephony, significantly reducing the intensity of competition, is not easy to explain. The severe political problems affecting Nicaragua between 2003 and 2006, the consequence of which has been, among others, the parallel and conflictive existence of two regulatory bodies, may contribute to understanding that situation. We will return to this topic further ahead.

3. Privatization without Exclusivity: The Cases of El Salvador and Guatemala

In the cases of Guatemala and El Salvador, privatization and opening up to competition took place simultaneously.

In El Salvador there appear to be two driving forces behind the privatization and reform of telecommunications. First, the decision of the 1989–1994 Administration to promote a programme of comprehensive reforms, drawn up with the support of a working group headed by Arnold Harberger and aimed at building a new economic model. This programme included initiatives to stabilize macroeconomic imbalances (inflation, fiscal deficit, current account, competitive exchange rate), a structural adjustment programme (privatization of the banks, liberalization of foreign trade and impetus for tax reform) and additional policies aimed at propelling an export-led economic growth (Argumedo, 2007). The second was the deep discontent of the main entrepreneurial groups with the situation of telecommunications.11

11 The export strategy was facing serious difficulties due to lack of infrastructure. Textile maquila, for example, needed modern telecommunications to participate in international "just in time"

In this context, as in the case of Nicaragua, the Administration decided to promote a wide-ranging programme of investments and a policy expressly designed to raise the enterprise's productivity. Thus, between 1990 and 1993, the enterprise called Administración Nacional de Telecomunicaciones (ANTEL) invested $133.1 million, i.e. an average of 0.6% of gross domestic product (GDP) annually. The number of lines was increased from 2.9 per 100 inhabitants in 1989 to 5.5 in 1993; the number of employees was reduced from 48 per 1,000 lines to 22 in 2003; and the institution's profits improved (Argumedo, 2007). The following Administration (1994–1999) decided to intensify the progress made, creating the Presidential Commission for the Modernization of the Public Sector (CPMSP) with the aim of decentralizing or privatizing public services such as electricity, telecommunications, water, public transport and ports, as well as analysing the reform of the Pensions System. As in other cases, the reform of telecommunications was assigned an exemplary role, which led to its being carefully prepared. Indeed, between 1994 and 1995, various studies were commissioned which had to include a diagnosis of the problems and the requirements for creating a competitive market in telecommunications. In 1996, the new Law of the sector was passed and the regulatory body was created. In 1997, the enterprise was restructured in such a way that a fixed telephony company was created and another one also to take charge of mobile operations. Finally, in 1998 both companies were privatized. France Telecom acquired 51% of the shares of Commonwealth Telephone Enterprises (CTE), S.A. de C.V., the fixed telephony company, for $275 million (the benchmark auction price was $268.8 million), and Telefónica España bought INTEL, the mobile company, for $41.0 million (the benchmark auction price was $11.9 million; Argumedo, 2007).

In Guatemala, privatizations began in the early 1990s. In 1993, a selective process to sell low-utilization and low-performance state assets began with the aim of financing public-benefit programmes. The most distinctive feature of economic reform until 1996 was the sale of state assets to finance the fiscal gap and obtain resources for activities and programmes on the social agenda. Most of the privatizations carried out in this period (mainly the sale of the national airline Empresa Guatemalteca de Aviacion, S.A. (AVIATECA) and the concession awarded to Comunicaciones Celulares (COMCEL) for Band A) were severely criticized for the

production strategies. In the early 1990s, the waiting time to obtain a new telephone line was more than two years, there were problems of quality in communications, the geographical coverage of the network was very limited, and the damage caused by the civil war was to demand resources to replace the infrastructure. This situation limited the State's possibilities to attract foreign investment. The high rates in long distance telephony, which subsidized local calls, hindered the competitiveness of exporting companies. A survey on obstacles to competitiveness carried out by the World Bank showed that over 58% of the companies interviewed stated that they had problems with the service, 71% of these said they faced the problem of overloaded lines, 14.5% pointed out crossed lines and interruption and more than 78% stated there were not enough lines to be able to acquire new ones. Other indicators showed low-quality service since 75% of faults took between four and 20 days to repair, whereas the rest took even longer. Around 35% of calls could not be completed due to high congestion (Argumedo, 2007).

procedures followed, which lacked transparency and left the population with a bad aftertaste regarding privatization processes. Towards 1996, after the signing of the Peace Accords, the political regime was becoming consolidated and although the growth rate was satisfactory with respect to the region, the country was facing a fiscal situation that threatened macroeconomic stability. In this context, a new government took office, headed by President Álvaro Arzú. Progress in the modernization of the state as a means to build a more efficient and competitive economy was the prime objective of the Government Program for the period 1996–2000. Special importance was placed on physical infrastructure, particularly in telecommunications (Urízar, 2007).

The government analysed three options: (i) keeping Empresa Guatemalteca de Telecomunicaciones (GUATEL) as a public enterprise and promoting initiatives aimed at strengthening it; (ii) selling GUATEL with a five-year period of exclusive concession; and (iii) selling GUATEL, thereby immediately opening up the market to competition and concentrating the government's activity on its regulatory role. The first option was discarded, as it did not adjust to the principles of subsidiarity promoted by the government. The second option proved attractive, particularly from the standpoint of public finances. However, the five-year concession of a monopoly was considered a risk, since a benefit of that nature would lead the private investor to do everything possible not to lose it. Thus, the reform would lead to nothing more than the passage from a public monopoly to a private one. To avoid that risk, the second option was also discarded. Despite the fiscal costs involved, the government opted for the third possibility. It was believed that a competitive atmosphere would contribute to attain more quickly a modern telecommunications services and at low prices (Urízar, 2007).

The government rejected the possibility of dividing the enterprise and preferred to sell it as a vertically and horizontally integrated operator. It was considered that its destructuring could hinder attracting investors and further affect the sale price. Efforts were made to neutralize the impact of this decision on competition through the approval of a Telecommunications Law. Political and legal difficulties of various kinds made it necessary to create a new enterprise, Telecomunicaciones de Guatemala, S.A. (TELGUA), to which all the assets and liabilities of GUATEL were transferred, leaving the latter with a minimal structure to enable it to continue serving community telephony. TELGUA took over the provision of local and LD telephone service and entered the market of the Personal Communications System (PCS). On October 1, 1998, the sale of 95% of TELGUA's shares to the company Luca, S.A. was finally concluded. The process was audited by Arthur Andersen and the value of the transaction was $977 million. Subsequently the owners of TELGUA established a partnership with TELMEX, an international operator, which acquired 45% of the shares. Prior to the Telecommunications Law, frequencies were obtained by authorization from the Executive Agency by means of concessions or authorizations. Chapter II (Articles 54–56) of the Telecommunications Law created the legal concept of "Frequency Usufruct Titles" (TUF), giving the spectrum the nature of an economic good rather than a public good. Thus, the category of regulated (free) Bands, which are allocated as TUF and can be freely transferred and utilized, within certain technical limitations, was created. TUFs detail the hours and area of operation, the maximum transmission power and the maximum interference permitted within the area of coverage (Urízar, 2007).

4. The "Honduran Way"

In the case of Honduras, with the enactment of the Law on Telecommunications (LMT) on October 31, 1995, Empresa Hondurena de Telecomunicaciones (HONDUTEL) became the responsibility of the Ministry of Public Works and Transport and later12 this responsibility was conveyed to the President of the Republic. Subsequently, these tasks were entrusted to the recently founded National Telecommunications Commission (CONATEL). Finally, the development and operation of telecommunications services were assigned to HONDUTEL and private companies. One fundamental feature of the LMT was the reform of the Organic Law of HONDUTEL, permitting association with private investors.

In this context the "capitalization process" arose for the purpose of attracting foreign private capital. At the same time, the enterprise was granted a concession in the following terms: "[F]or 10 years following the date of entry into effect of the Framework Law for the Telecommunications Sector, HONDUTEL shall provide, in an exclusive manner, national and international telephony services and telegraphy services in places where there is no other means of communication in the country, and such exclusivity shall be incumbent upon the company or companies established by HONDUTEL" (quoted by Tábora, 2007). In October 1998, the National Congress of the Republic approved Decree 244-98 which broadened the concession to 25 years and extended it to the non-exclusive operation of the carrier service. To modernize the sector, a mechanism was promoted that consisted in capitalizing a new enterprise with the assets of HONDUTEL. The new company, Compania Hondurena de Telefonos, S.A. (HONDUCOM), would be capitalized by means of a public bidding process in which 47% of the shares would be transferred to a technical partner who would take over the management of the enterprise, 2% would be sold to the workers and the state would keep 51% through HONDUTEL.

On signing a letter of intent with the International Monetary Fund (IMF) in 1997, a new distribution of shares was agreed upon, by which the state would keep 47% of the shares, 2% would be sold to the workers and 51% would be assigned to the strategic partner. A number of problems, among them Hurricane Mitch, delayed the process.

Finally, in 2000, seven companies showed an interest in acquiring the above-mentioned block of shares. However, only one of them, TELMEX, made an offer. The offer was only just over one third of the benchmark price ($300 million), as a result of which on October 16, 2000, the bid was declared void and in May 2001 it

12 On October 25, 1977, by virtue of Decree 118–97.

was decided to suspend the process until another arrangement could be found (Tábora, 2007).

The difficulties in capitalizing the enterprise led to the design of the programme "Telefonía para todos" (Telephony for all – TpT) with the aim of benefiting users and the national economy by addressing unsatisfied demand for telephone service and improving service quality and attention to customers, thus laying the foundations for the opening of the market as of 2005. To that end, competition would be introduced in fixed voice services, private participation in the sector and the supply of new services would be encouraged, thus stepping up the growth and modernization of the sector.13

The TpT programme envisaged broad access to the telecommunications market. In fact, all interested parties could sign a contract and receive equal treatment from HONDUTEL. Specifically, HONDUTEL should sign a contract with any party interested in establishing itself as a marketer of sub-operator type that complied with the following requirements: (i) being legally incorporated as a business enterprise in accordance with the country's laws; (ii) in the case of natural persons, having the legal capacity to contract and in the case of foreign companies, being authorized to carry out business transactions in Honduras; and (iii) submitting a sworn statement to the effect that neither the applicant nor any of the partners holding more than 10% of the capital stock are included in the cases provided for in Article 92 (Sections e to k of the General Regulations of the Framework Law for the Telecommunications Sector). Having complied with the above requirements, the interested parties would proceed to sign the Marketing Contract, after which they would obtain their registration and the licences for use of the radio spectrum (if necessary). Likewise, they would be assigned blocks of numbers and would have to pay the established rates.

The TpT programme introduced two important innovations. The first stems from the creation of the concept marketeer sub-operator, who is defined as "the one that receives from HONDUTEL an extension of the rights granted by Law, to enable it to directly provide the public telecommunications services authorized to HONDUTEL. To provide the services the sub-operators may make investments in infrastructure in order to expand the public telecommunications network, contributing both facilities and value-added services." Even more important is the implicit annulment of HONDUTEL's exclusivity (Tábora, 2007).14 Finally, the TpT decree

13 The programme "Telephony for All – Modernity for Honduras" (TpT) was approved by means of Executive Decree No. PCM 018–2003 dated September 23, 2003, and ratified by Legislative Decree No. 159–2003, published in the Official Gazette on October 24, 2003 (Tábora, 2007).

14 In this regard, Tábora points out that "On the other hand, the same TpT Decree, with the aim of permitting anyone wishing to invest in telecommunications to do so, eliminated all kinds of restrictions for those wanting to enter the market. This situation, due to the principle of egalitarian treatment and non-discrimination established by the LMT, left in contradiction and legal inconsistency all the constraints included in this matter in the concession contract of the mobile operator, CONATEL being obliged to amend the Concession Contract in these aspects, a situation which gave rise to dissent among the members of CONATEL, and which was interpreted by some as an example of inconsistency in sectoral policy on the part of the government, and as an irregular situation by others, even though the amendment was executed by a motion of the National Congress and was duly approved by Congress once it was made effective." (Tábora 2007).

ended segmentation in the provision of telecommunications services, since the sub-operator can use the same infrastructure to provide the different services, which also promotes technological convergence.

The impact of the "Honduran way" toward the liberalization and modernization of telecommunications seems promising. By September 2005, 40 companies had obtained registration as sub-operator15 of which 17 had started commercial operations and installed 63,000 telephone lines,16 two thirds of which belong to MULTIFON (Tábora, 2007).

II. Regulation and Competition in Telecommunications: Institutional Problems

1. Introductory Considerations

In the region's experience, with the exception of Costa Rica, public telecommunications enterprises traditionally have been unable to provide adequate services. Hence, privatization emerged as the possibility to overcome these problems by incorporating private management capacity into service provision.

However, the fact that telecommunications companies were natural monopolies made it necessary to create a regulatory agency to ensure consistency between private operation and public interests. International evidence was conclusive to the effect that a good law was not sufficient and that a regulatory agency was needed with the necessary resources to effectively apply the regulatory framework.

The problem is that the regulatory agencies created in the region began to operate either long after the privatization process (Mexico's case) or simultaneously with the privatization and the opening up to competition. Thus, these agencies are weak and face serious problems associated with asymmetry of information. Moreover, the regulatory agencies of small countries have to face transnational corporations which have major financial and technical backing. The technical capture is, therefore, a permanent possibility that prevents, or at least seriously hinders, effective regulation. The difficulties regulators face to obtain the required information seriously hinder their task and hence the need to promote competition as the main mechanism that can assure the industry of an efficient performance. However, the types of privatization and opening up to competition in the region have not been so favourable to the development of competition. Although the situation varies from one segment and from one country to another, the fact is that companies other than the incumbent effectively operating in the market are not sufficient in number or

15 The TpT Decree establishes that once HONDUTEL's exclusivity period is over, sub-operators may continue offering the services, whether under the same scheme of sub-operator or directly requesting their empowering title accrediting them as operators (Tábora, 2007).

16 Equivalent to one sixth of the total lines installed (Tábora, 2007).

in size to create the necessary competitive climate for the telecommunications market to operate efficiently.

In this context, the competition agency should play a crucial role in eliminating obstacles to entry and preventing anti-competitive conduct. Despite the progress made,17 the existing competition agencies in the region show serious difficulties in efficiently protecting competition in the telecommunications sector. In some cases the general law on telecommunications assigns the regulatory agency tasks designed to protect competition in the sector, but these are very general standards that are frequently almost inoperative and that definitely do not represent an effective means of promoting a competitive market.

In this context, the aim of this section is to identify the main problems regarding design of the regulatory framework for telecommunications and the institutional system for regulation and promotion of competition in this market.

2. Institutional Design and Competition18

Academic reflection and analysis of international experience regarding privatizations and the transition towards a competitive market in telecommunications tend to agree that the success of the process is associated with it being effectively planned; being preceded by the enactment of a Telecommunications Law that defines the rules of the game; and with the creation of a regulatory agency with the necessary authority and resources to efficiently apply the regulations. This institutional framework should be complemented with the prior establishment of the competition agency (Rivera, 2006). As we shall see, the experiences in the region are still far from this model.19

Although on different dates, Panama, Nicaragua, Guatemala, Honduras and El Salvador approved new Telecommunications Laws and created specialized regulatory bodies prior to the privatizations process. Up to the end of 2005 only three countries of the region had competition authorities: Costa Rica, Mexico and Panama. In Costa Rica's case, however, the issue is not important in view of the existence of the public monopoly in telecommunications. Furthermore, the Competition Law explicitly denies the Commission on Promotion of Competition (COPROCOM) any power regarding the sector. Further ahead we will return to the cases of Mexico and Panama after dealing with the cases of the countries that have lacked a competition authority.

17 The Superintendence of Competition of El Salvador began its operations on January 1, 2006, and Honduras was due to set up its agency six months after the Law was passed (early 2006).

18 For a detailed analysis of the Central American telecommunications regulatory framework, see Rivera (2006).

19 Costa Rica is once again an exception, since it lacks a specialized regulatory body. The state telecommunications and electric power enterprise still maintains important regulatory powers, particularly as regards spectrum management.

In El Salvador the Law creating the General Superintendence of Electricity and Telecommunications (SIGET) was passed by Legislative Decree No. 808 of September 12, 1996. The competition-related problems that could be faced by a telecommunications industry, particularly in a small country, led different sectors, as of the early 1990s, to put forward legal proposals to punish abuse of market power by one company to the detriment of its competitors. These proposals, however, were rejected, and therefore Article 111 of the Telecommunications Law granted SIGET the authority to prohibit anti-competitive practices. Unfortunately the law does not establish mechanisms for their study and sanction (Argumedo, 2007).

SIGET's highest authority is the board of directors, which is made up of one director appointed by the President of the Republic, who performs the duties of superintendent, one director elected by private sector trade associations legally established in the country (through the National Private Enterprise Association – ANEP) and one director appointed by the Supreme Court of Justice (CSJ). The directors are designated for a period of seven years in office and may be nominated again.20 However, in the first nine years of operation SIGET had four superintendents (average tenure two years and one quarter), which reflects the strong tensions to which the entity has been subject. The case of El Salvador is also noteworthy due to the presence, in SIGET, of a director elected by private sector trade associations, in particular the above-mentioned ANEP.

In general, in Latin America, there is much discussion regarding the possibility of the governments exerting pressure on the telecommunications regulatory body to favour short-term interests. Instead, in the Salvadoran case concern arises over the regulatory body being subject to the influence of the private sector. One would think that SIGET's director could represent the interests of the companies using telecommunications services and therefore that conflicts of interest do not exist. However, the public interest represented by SIGET could at times be inconsistent with the interests of business chambers.

The Competition Law was passed in late 2004 and entered into force on January 1, 2006, with the creation of the Competition Superintendence. One of the main challenges would be its coordination with SIGET.

In Guatemala's case, regulation of telecommunications is the responsibility of the Superintendence of Telecommunications (SIT). When the law creating the agency was under discussion, the possibility of granting the entity full autonomy was suggested. However, the support of two thirds of Congress could not be garnered for that purpose. It was therefore decided to make SIT dependent on the Ministry of Communications, Infrastructure and Housing (MICIVI), which, in turn, was to appoint and could remove the superintendent. This official's authority depends on the minister of MICIVI (Urízar, 2007). Subsequent amendments to the Law further undermined the authority of the Superintendent of Telecommunications,

20 A reform approved in December 1997, established the director from the private sector and the one from the CSJ could remain in office for five and three years, respectively.

first (by virtue of Decree 47–2002 Articles 13 and 15 were amended and 14 was eliminated), through a reduction in his financial resources and by conditioning his budget to a government decision; and second, by stripping him of the power to proceed judicially and administratively against transgressors of the law (modifications of Decree 15–2003 which amended numerals 2 and 3 of Article 81, relative to infringements and fines; Urízar, 2007).

In Honduras, the LMT Sector, enacted through Decree No. 185–95 of October 31, 1995, created CONATEL to regulate and oversee the development and operation of telecommunications. CONATEL has the authority, among others, to collaborate with the President of the Republic to formulate telecommunications policies and ensure their effective execution by means of regulation and coordination actions. It therefore assumes the duties entrusted in other countries to the Ministry of Communications. CONATEL's Commissioners are appointed by the President of the Republic, may remain in their posts for up to four years, and may be nominated for additional periods. Their terms of office coincide with that of the President of the Republic and they may be removed whenever the authority deems it appropriate.

The LMT establishes that it is the responsibility of CONATEL to promote competition in the telecommunications services. In this regard CONATEL must ensure that operators of telecommunications networks provide access, on equal conditions, to other operators and users in the same or similar circumstances. Nevertheless, despite the fact that these duties were entrusted to CONATEL, the LMT, on establishing the breaches to be sanctioned as serious, makes no reference to infractions against free competition (Tábora, 2007). The type of privatization process in Honduras, already described, which maintains the incumbent enterprise as public, has so far minimized the problems that this legal and institutional design could pose to the industry's development.

In 2005, the debate ended and the Competition Law was passed. The new legislation has pre-eminence over sector legislation, but promotes coordination with the sectoral regulatory body, especially when the markets are reasonably open to competition as in the case of telecommunications. The reasons that led to demanding coordination are the following: (i) the need to have technical knowledge of subjects related to the sector during the transition to, and consolidation of, the new competition agency; (ii) the need to define the setting of competition in advance and not be limited to applying sanctions retrospectively for conduct contrary to competition or to restructuring the sector; and (iii) the need to apply policies unrelated to competition but which the government considers important, for example, on universal service, national security and protection of users.

Finally, as in the other countries without a competition authority, in Nicaragua the Telecommunications Law provides for a competitive mechanism in the provision of services. However, Article 24 of that Law envisages the possibility of granting concessions with exclusivity for a period to an operator. The Law prohibits anticompetitive practices and establishes that operators are obliged to provide satisfactory access at competitive rates to the telephone network of service providers whose licences have been authorized by Instituto Nicaragüense de Telecomunicaciones y Correos (TELCOR). However, TELCOR's powers are limited to demanding information and adopting pertinent corrective measures, which are not specified (Ansorena, 2007). The agency is headed by a Director General appointed by the President of the Republic.

The weakness of the regulatory body was accentuated by the serious conflict between the President of the Republic (2001–2006) and the alliance of Sandinistas and Liberals.21 The latter used their parliamentary majority to introduce important reforms that affected the Supreme Electoral Council (CSE), the CSJ and the Office of the General Comptroller of the Republic. In this context, and with the opposition of the President of the Republic, the above-mentioned political alliance succeeded in passing the Law creating the Superintendence of Public Services (SISEP) on November 23, 2004, in the Legislative Assembly. This new entity had to bring together in one single government structure the regulatory bodies of the sectors providing basic services (water, electric power and telecommunications). This approach to the institutional regulatory framework could be positively assessed in the context of the international discussion, bearing in mind the shortage of human resources in the country. However, the fact that at least during 2006 both TELCOR, supported by the President of the Republic, and SISEP, supported by the National Assembly, were operating gave rise to an unsustainable situation that made promotion of competition in the telecommunications sector even more difficult.

The above analysis makes it clear that the countries of the region that have no competition authority suffer from an institutional weakness in their regulatory framework, which, in turn, is an obstacle to create favourable conditions for competition.

What has happened with Mexico and Panama, which faced the opening up of the industry to competition when they already had a competition authority?

From the point of view of the sequence, Mexico operated in a very different manner from that usually recommended. It privatized in 1990, organized the Federal Commission on Competition (CFC) in 1994, passed the LGT in 1995, created the Federal Telecommunications Commission (COFETEL) by means of a presidential decree in August 1996 and opened the market to competition in 1997.

If analysed in general terms, the LGT proves adequate. It defines a regulatory framework and includes standards that assume a competitive structure in the industry. Nonetheless, if analysed in greater detail important difficulties appear. First of all, the sequence followed by the process was wrong. The concession licence granted to TELMEX in 1990, which regulated the performance of the enterprise until the approval of the general law, was likely to collude with the latter and with the Competition Law. Indeed, the LGT provides that licencees could freely set the rates for telecommunications services, which should be non-discriminatory. Only if a company has substantial power in the relevant market according to the Federal Law on Economic Competition (LFCE) can the Ministry of Communications and

21 Enrique Bolaños formed part of the Liberal Party headed by ex-president Arnoldo Alemán (1996–2001). However, both representatives drifted apart when Bolaños's government joined the accusations of corruption against Alemán (who was imprisoned for that reason).

Transportation (SCT) establish specific obligations related to rates, service quality and information (Article 63). In 1997, the CFC declared that TELMEX had substantial power in five markets (local telephony, access, inter-urban transport, national LD and international LD) and consequently requested from COFETEL the establishment of specific obligations regarding rates. TELMEX complained and held that such statement violated its concession licence.

A second difficulty was that the LGT assigns the SCT the basic regulatory authority. In spite of this, in August 1996, President Ernesto Zedillo created COFETEL by presidential decree, which basically plays an advisory role to the SCT, to which the law assigns the regulatory functions. Thus, a highly ambiguous situation arises. Formally, the regulatory body is COFETEL, but due to its weak legal base, companies question the legitimacy of any decision adverse to them. In fact, on crucial issues the SCT, particularly the Under Ministry of Telecommunications, is the regulatory body. Like many Executive Branch agencies, the under ministry tends to privilege short-term growth of investment and coverage, even at the cost of the development of competition (Mariscal and Rivera, 2007).

A third difficulty was that even though the Telecommunications Law includes among its objectives fostering healthy competition among the different providers of this service, no institution is assigned this generic function. Indeed, among the functions of the SCT, no specific mention appears in this regard; nor does the Federal Law on Competition appear among the laws that should be applied if no express provision on competition appears in the Telecommunications Law. Only in the case of tenders does it indicate that the bases should include a favourable opinion from the CFC. It is the decree creating COFETEL that said agency is obliged to register the rates for telecommunications services and establish specific obligations with regard to rates, service quality and information for licences of public telecommunications networks that have substantial power in the relevant market. Naturally, an Executive Branch decree is a weak legal instrument to assign powers of such importance to an administrative agency.

The above difficulties are compounded by the non-existence of a fixed term of office for COFETEL commissioners and of the specific causes of removal from their posts. Particularly important is the generalized use of the remedy of amparo which allows the regulatory action to be suspended when those affected consider that the norm applied violates their constitutional rights. In a context in which justice operates slowly, these remedies can remain in process for as long as three years, which naturally leads to ineffective regulation (Mariscal and Rivera, 2007).

In Panama's case, even though nominations for commissioners take place in the manner indicated by the Law, it seems it has not been altogether possible to prevent them being carried out on the basis of short-term political interests. In fact, the predominance of the party in government can be accompanied also by its prevalence in the Legislative Branch and, in this context, make politically based nominations, not in line with the capability required by the post (González, 2007). A similar problem has been put forth in the case of Nicaragua (see Ansorena, 2007).

In Panama, the fact that the state retained 49% of INTEL's shares gives rise to a conflict of interests since the government expects high earnings for the operator who can provide higher public revenues, but this can hinder the regulator's mission of applying an appropriate tariff policy (González, 2007).

III. From Exclusivity to Competition: Models and Main Lessons

As we have seen throughout this chapter, the countries of the region that privatized and opened up the telecommunications industry to competition show dissimilar processes. The sphere of design of the institutional economic framework also differs among countries.

However, all the countries studied, whether or not they went through a period of exclusivity, have opened their doors to a competitive model. The aim of this section is, first, to analyse the principal characteristics of the opening up to competition, identify the positive aspects and problems of each experience and subsequently, in the fourth section, evaluate the industry's performance in each of the countries and telecommunications segments.

As we saw above, three countries, Nicaragua, Mexico and Panama opted for granting a period of exclusivity to the new owners of the privatized enterprises. These periods varied in length (six years in Mexico and three years in Nicaragua and Panama) but in all of them, exclusivity included fixed and LD telephony and gave the company that acquired the public enterprise the right to operate a mobile telephony licence.

Another two countries, Guatemala and El Salvador, opted for immediately opening up the industry to competition after the privatization without offering exclusivity. Before carrying out this operation, El Salvador separated fixed telephony from mobile telephony and sold them to different operators, whereas Guatemala transferred a single enterprise to the private sector in an integrated manner.

1. Exclusivity and Opening Up to Competition: The Cases of Mexico, Nicaragua and Panama

The literature has confirmed that privatization with exclusivity represents significant costs for society as regards both coverage and service prices. Scott Waltsten (2000) empirically demonstrates that a monopoly is more valuable to its owners than a company operating in a competitive market and concludes that a government can practically double the revenues obtained from privatization by guaranteeing exclusivity rights. Coinciding with some of the findings in this chapter, Waltsten's study also shows that the greater revenues are accompanied by a reduction in the growth of the telecommunications network of up to 40%.22

22 A similar conclusion is reached by Estache et al. (2002, p. 157) who, on evaluating the progress of telecommunications in Latin America, maintain that "high prices continue to be perceived as

In Mexico's case, the opening up to competition shows distinct characteristics in each of the segments. In 1987, when the mobile market took off, the country was divided into nine regions, with two licences granted in each one. TELMEX, through its subsidiary Telcel, received a licence in each of the regions, the only proviso being that the company could not be the sole provider in any of those regions. In TELMEX's concession licence, the SCT reserved the right to grant another, or other concessions, in favour of third parties in order to develop, in equal circumstances, fixed telephony and national and international LD services, within the same geographical area or in a different one, identical or similar to those that were the subject matter of the concession. However, during the six years following the date in which the concession licence was signed, the SCT pledged to grant other concessions for national and international LD basic telephony public service networks, only if the company had not fulfilled the conditions of expansion and efficiency set forth in the concession licence, which contradicts the above. Chapter V of the said licence defined the general conditions for interconnection and the company's interconnection obligations with other LD public networks when the sector was opened to competition as of January 1, 1997 (Mariscal and Rivera, 2007).

Between 1988 and 1990, Telcel – TELMEX's mobile affiliate – expanded its cellular network in the 800 MHz (Band B) radio frequency, covering the cities of Tijuana, Cuernavaca, Toluca, Guadalajara, Monterrey and the Mexico City metropolitan area. In 1990, the company began to offer cellular telephony in the nine regions into which mobile telephony service was divided in Mexico. In 1998, the 1,900MHz (Band D) radio frequency for PCS in the nine regions was awarded to Telcel. Subsequently, in 2004, a wide-ranging auction was carried out that would make it possible to meet the country's enormous demand for mobile telephony. Towards the end of the last decade, the segment's main operators, together with Telcel, were Iusacell, operating since the end of 1980s, Pegaso which entered in 1999 and Unefon which began operations at the beginning of the year 2000. The entry of Movistar, of Telefónica España, in 2002, contributed dynamism to the market. Nevertheless, up to 2004 the share of companies other than Telcel reached only 22.8% (Mariscal and Rivera, 2007).

By 1994 applications to enter local telephony had already been presented. Iusacell, a joint venture between Alejo Peralta and Bell Atlantic, proposed providing the service by means of fixed wireless technology, but it was not until October 1997 that the rules for the development of competition in local telephony were published. The objectives of the said standards were to promote competition,

an issue in the region. To a large extent this is a result of the limited competition in the sector. For most countries, the exclusivity periods granted to get the privatization deals done, resulted in lasting high connection and usage tariffs. ... Indeed, residential connection rates continue to be high when compared to the US even if they have gone down significantly. These exclusivity periods are now coming to an end as in Argentina or Venezuela and this should lead to a reduction in tariffs driven by the market. However, most countries have not yet defined the rules of the game to ensure competition in a sector where costing the access to bottleneck facilities continues to be at the core of the regulatory debate."

facilitate interconnection and interoperability of public networks, ensure the continuity of the service and avoid the application of discriminatory rates. In 2003, eight companies were authorized to provide local wired service (Ramírez, 2005). The results, however, are not encouraging. In 2004, seven years after the segment was opened up to competition, TELMEX maintained a share equivalent to 95% in local telephony, although the companies that entered the market did increase their share in a sustained manner (Mariscal and Rivera, 2007).

In June 1994, the SCT issued the "Resolution on the Plan for Interconnection with Public Long-Distance Networks" which, among other things, obliged TELMEX to interconnect its networks with all the interested parties. It was expected that by early 1997, 60 interconnection points would be in operation and that they would increase to 200 in the year 2000. With the prospects of the opening up of LD to competition, several companies began to apply for concession licences to operate in this segment.23 In January 1996, the first negotiations among the operators took place, without agreement being reached on interconnection rates, which led them to request the intervention of the SCT. Both TELMEX and the new LD operators questioned the SCT's resolution by means of formal claims. Whereas for TELMEX the charges were insufficient to cover the costs it had incurred to provide the service, for the entrants there were no reasons to justify them. Following a year of analysis, the recently established COFETEL upheld the above-mentioned rate, although according to different international standards it was extraordinarily high.24

In January 1997, seven companies entered the LD market, users being able to pre-select the operators in 60 cities. Among the licencees were Mexican industrial and financial groups associated with US telecommunications operators. Three major operators dominated the market: TELMEX, which established a joint venture with US Sprint to provide international services between Mexico and the USA, Avantel (Banamex and MCI) and Alestra (Alfa and AT&T), which merged with Unicom (Bancomer, GTE and Telefónica). Despite the problems mentioned above with the opening of the market, the companies that entered quickly gained shares in national LD, attaining 30% in 1999. Since 2001, this process has gradually reverted, dropping to 23.2% in 2004.

Seventeen years after opening up to competition in mobile telephony and almost nine years in local and LD telephony, the results from the point of view of operators other than the incumbent were not encouraging. Compared to other Latin American

23 Together with TELMEX, which had a licence since December 1990, Avantel obtained it on October 6, 1995; IUSATEL on February 12, 1996; MARCATEL on February 26, 1996; INVESTCOM on April 4; BESTEL on April 10; MIDITEL on April 26; ALESTRA on April 30 and TELINOR on July 23.

24 A rate was established for cities with Interurban Traffic Centre (CTI; US$0.0257 in 1997 and US$0.0231 in 1998) and another for cities with Routing Capacity Centers but without CTI (US$0.0240 in 1997 and US$0.0219 in 1998). Furthermore, the SCT determined that the companies should pay an additional 58% for traffic termination, which raised the effective rate in 1997 to US$0.0536 and US$0.0513, respectively. By contrast, a World Bank study of the Mexican market suggested an average weighted rate of US$0.019 cents.

countries, Grupo Carso's share of the three segments analysed in Mexico far exceeded the incumbents' share in Argentina, Brazil, Chile and even Peru. It should be pointed out that this situation has occurred despite the fact that it was the companies from the USA, Mexico's great neighbour, which tried to become consolidated in the Mexican market, but without success.

One important specific element has to do with the modalities provided for in the Law for the establishment of interconnection and access charges. As in other countries, the LGT assigns the operators themselves the task of agreeing on those charges. What can be a reasonable, efficient mechanism in a context in which the companies are of a relatively equivalent size is not when companies not only of different sizes but also having strictly contradictory interests come together in the negotiation process. Indeed, as soon as the exclusivity period is over, the entrants face the urgency to begin operations. The incumbent, on the other hand, is interested in hindering the entry of its competitors as much as possible, and, furthermore, every day that the negotiation process is lengthened is another day that it continues to have exclusivity. The more the negotiation period is prolonged, the longer the time during which the entrant has to face its fixed costs and its financial costs without beginning to receive income, which clearly stands as a barrier to the entry of competitors. The delay in the start of operations significantly raises costs for competitors, reducing their competitive possibilities.

Representatives linked to TELMEX and to COFETEL maintain that the entrants' difficulties stemmed from their low levels of investment, from the limited knowledge of US companies and other foreign operators of the characteristics of telecommunications markets in developing countries and from errors on the part of the executive bodies (Mariscal and Rivera, 2007). Nonetheless, the characteristics of the privatization process with a long period of exclusivity that consolidated a single operator and the weakness of the institutional regulatory framework certainly played a central role. It can be said that the institutional framework did not work properly. The various complaints filed by the different competitors both before the CFC and COFETEL took a long time to be resolved, losing all effectiveness. The declaration of dominant enterprise that the LFT provided for as an instrument to avoid anti-competitive conduct by the incumbent company had no effect due to the said way in which the declaration was written and the remedy of amparo which neutralized any administrative measure by the regulatory body and the competition supervisor. Everything points to problems with the legal framework which go far beyond the sphere of regulatory policy and protection of competition (Mariscal and Rivera, 2005a, b).

In the Panamanian model of transition to competition, the opening up was due to be implemented in two phases: the first, corresponding to basic telephony, on January 2, 2003; and mobile telephony in January 2008. The regulatory framework included the following model for opening: the point of departure was acknowledgement that the entrants would face serious difficulties resulting from the consolidation of the incumbent enterprise. In this regard, the new companies were assured interconnection provided by the incumbent in an (apparently) peremptory period and total freedom to choose the services to be provided, as well as the customers. It was believed that by choosing the most profitable markets, entry would be made feasible. The unbundling of networks was also provided for, i.e. the possibility for entrants of leasing segments of the network from the incumbent to provide their services and number portability (González, 2007).

The opening was prepared well in advance. As of early 2002, numerous companies applied for concessions. Up to February 2005, there were 28 concessions for local telephony, 33 for national LD, 59 for international LD, 20 for public telephones and 13 for the hiring of dedicated voice circuits. The basic idea was for the entering operators to have their installations ready to enter into operation on the appointed date. Bellsouth, a company that controlled 45% of subscribers, requested concessions for all the services. Cable Onda, S.A., the dominant cable television operator, did much the same. It was therefore a question of operators with a significant market share and a solid foundation in the new telecommunications technologies.

Despite the existence of a regulatory framework favourable to the entrants, the liberalization of the sector has encountered obstacles that in the best of cases have substantially delayed implementation. There are four fundamental aspects. (i) The interconnection mechanism, in which the regulator only intervened if the parties could not reach agreement, contained incentives for both parties to await the regulator's intervention, which substantially delayed the processes. Even having defined the conditions and the interconnection rate, the problems did not end, for numerous technical difficulties have further delayed the process.25 (ii) With regard to competition in LD, problems arose both in the use of access codes and in automatic routing. The changes demanded by the use of various access codes by users were subject to the speed at which the incumbent implemented them. Regarding the second mechanism, it was assumed that its operation gave rise to change costs, basically derived from the programming the incumbent had to assume. If these costs were to be covered by the user, the change of provider would be discouraged. Alternatively, if these costs were to be assumed by the entrant, its competitiveness would be reduced. (iii) Related to the above, the survey had to be carried out among users so that they choose their LD services provider, even though done as scheduled, mainly because the cost of the calls could not be determined before the negotiations on interconnection charges had concluded. (iv) Finally, number portability had still not been put into operation and judicial proceedings halted efforts regarding unbundling of the network. With regard to the latter aspect, the non-existence of an explicit standard in this respect, within the legal framework, led this policy to be questioned before the courts (González, 2007).

25 The most significant problems that arose were the expansion of routes and new interconnection points. In the case of the expansion of the E1 for POI in the city of Colón and in the provinces, as requested by the entrants, this had still not been achieved until June 2004. Thus, to that date, competition in national LD had not yet begun, despite the fact that, as we saw, there were a significant number of entrants according to the regulatory framework; it had begun a year and a half earlier. For a detailed description of the perverse incentives of the regulatory framework and the "technical competitive" problems encountered, see González (2007).

Panamanian legislation does not encourage the incumbent to allow interconnection, as it would threaten its monopolistic position; nor does the entrant have incentives to reach agreement with the incumbent, since, having analysed the incumbent's negotiating capacity, it is unlikely that the latter would be prepared to grant a lower access charge than the regulatory body would define, and it is therefore in its interest to await the regulator's arbitration. Finally, the regulator has strong incentives to "split the child in half" since according to Article 216, interconnection charges should reflect "at least the long-term incremental costs" (González, 2007). Having added the practical problems,26 the consequence of the above was that in mid-2004, i.e. a year and a half after the opening up, important cities such as Agua Dulce, David, Chorrera, Santiago and Chitré did not have interconnection points.

Despite the above, opening up to competition in LD has meant a substantial drop in prices due to both the impact of the threat of competition and the lessening of difficulties in implementing the required investments. According to estimates the average minute of national LD was to drop from $0.15 in 2002 to $0.07 in 2007. In the case of local telephony the progress has been negligible.

This situation raises a more general debate. Whereas some placed emphasis on the unbundling of networks as an important mechanism for introducing competition in fixed telephony, others maintain that it is an unworkable policy (the owner of the network cannot be compelled to lend their network and even less below its cost) which discourages investment (the entrant, if it can use the incumbent's network, will not want to invest). The Panamanian experience appears to show three basic problems: (i) the difficulties in moving from a system of exclusivity to one of competition are usually underestimated; (ii) the capabilities of network unbundling as a basic instrument for introducing competition in fixed telephony are overestimated; and (iii) the postponement of full liberalization of mobile telephony until 2008 hinders the most effective means of introducing competition throughout the sector – mobile telephony.

As we saw, the third country that granted a period of exclusivity to the former public enterprise when it was privatized was Nicaragua. The privatization process culminated in December, 2001, and therefore the opening up to competition was to begin towards the end of 2004. However, the process was postponed until April 2005, by means of a parliamentary resolution.27 Subsequently, the conflict between the two regulatory bodies, TELCOR and SISEP, had translated into hard bargaining regarding whether the telecommunications should be opening up or not. Whereas

26 The most important problems were those of expansion of routes and new interconnection points. In the case of the expansion of the E1 for the POI in the city of Colón, this had not been achieved in June, which hindered competition in national LD. The ensuing interconnection problems translated into problems of congestion, which were a serious disincentive for consumers to change provider (González, 2007).

27 The National Assembly established that the opening did not have to be on December 18, 2004, three years after the signing of the privatization contract, but had to be three years after ENITEL's concession contract was published in the Gazette, which was in April 2005.

the first entity declared the total opening up of this sector in April 2005, the recently created SISEP disregarded such measure and proposed an extension of ENITEL's private monopoly and a freeze on telephone rates after the start-up of the new regulatory body (Ansorena, 2007). This situation resulted in serious disorder. Towards the end of 2005, ENITEL's monopoly in basic telephony continued, SISEP was engaged in preparing a new project for opening in consultation with the operators and the mobile telephony company had entered the international LD telephony segment, refusing to recognize ENITEL's exclusivity (Ansorena, 2007).

As to mobile telephony, Bellsouth operated in conditions of exclusivity between 1997 and the end of 2002. In 2001, TELCOR decided to open the segment to the participation of other operators by initiating the bidding process for the second mobile telephony licence. Thirteen companies took part in the bidding. The operating licence was finally28 awarded to PCS SERCOM, an affiliate of the Mexican company América Móvil, which began operations in December, 2002, simultaneously with ENITEL, which had the right to a mobile telephony licence in accordance with the privatization contract. Competition in the segment suffered a grave setback when the above-mentioned Mexican company acquired ENITEL in 2004, and thus held two mobile telephony licences. The non-existence of a competition agency that could have established as a condition for the purchase of ENITEL that América Móvil should get rid of one of the mobile licences represents a serious problem for the development of competition in the sector. Bearing in mind that mobile companies are central to the development of overall competition in telecommunications,29 the prospects for competition in Nicaragua are not at all promising. Despite the foregoing, it should be pointed out that the number of mobile subscribers rose from 200,000 in 2002 at the end of the exclusivity period to 686,000 in 2004.

2. Privatization and Opening Up to Competition: The Cases of El Salvador and Guatemala

In contrast with the countries analysed in the preceding section, both El Salvador and Guatemala rejected the possibility of granting a period of exclusivity to the companies that acquired the former public enterprise and opted instead for immediately opening up the industry to competition.

In El Salvador's case, the decision to sell the assets of the fixed company separately from those of the mobile company is also noteworthy. Thus, the presence of at least two telecommunications operators was assured. The option for a

28 In fact, the bidding was won by another operator linked to president Alemán, the owner of which was later accused of corruption. The licence was transferred to the second bidder, PCS SERCOM.

29 On the importance of mobile companies in overall competition in telecommunications, see Mariscal and Rivera (2005a, b).

competitive market in mobile telephony has translated into the fact that despite the country's small size, there are four companies operating: CTE-Telecom (since 1999), Telefónica (since 1998), Telemóvil (since 1992) and Digicel (since 2002). Three companies have national coverage and one of them has centred on the urban area (Digicel).30 After having only one operator in 1998, by 2004 the mobile market was distributed fairly equitably among the four operators.31

As Argumedo (2007) points out, there is another situation in fixed telephony, where CTE maintains its dominant position. In 1998, the company had 386,600 lines; by 2002, the total number of lines had risen to 709,400, of which CTE had 90%; and by the end of 2004, the lines had continued to grow, although at a slower rate, reaching 887,800. CTE's share continued to be high and was equivalent to 88.7%. According to SIGET, there are another nine companies in fixed telephony, but the concentration is greater than in mobile telephony. CTE continues to be practically the only company that offers fixed telephony services to the residential segment, and it faces more competition in the segment of service to businesses.

In Guatemala the process of interconnection between the incumbent operator and the new operators began in mid-1999, and conflicts arose from the start. In the mobile segment, when the "caller pays" mechanism was introduced in October 1999, there were differences of opinion and legal disputes over this system. There were three cases of arbitration for interconnection charges in mobile networks. In international LD there were also tensions due to the fact that, among other things, the earnings from incoming international traffic included subsidies to local telephony. There have been many conflicts over interconnection charges for incoming international traffic, two of which went to arbitration, but those regarding international termination charges continue to be in conflict. The need to gradually eliminate this situation in the context of the interconnection contracts signed has also given rise to other disputes between operators, caused by: (i) blockage of calls to operators who have filed complaints before the SIT; (ii) lack of renewal of interconnection contracts with the incumbent, which has led to three complaints before the SIT; (iii) discriminatory rates for calls to competitors, giving rise to complaints in ordinary proceedings; and (iv) double charges to users of international LD, among others (Urízar, 2007).

The LGT does not establish any restrictions on the entry or exit of competitors, but most operators,32 some international analysts33 and SIT officials, believe that in practice the barriers to entry come from discriminatory treatment and the difficulty

30 According to Argumedo (2007) the competition is intense. The companies offer varied plans to attract different segments of the contract market (individual, family, friends, corporate) and prepaid cards for different amounts. They also offer a wide variety of complementary services such as roaming, text messages, sending of photos and sending of emails. The companies have been very aggressive in expanding the network of points of sale and charges for products and services.

31 Even though Telemóvil keeps a greater participation (28%), CTE, Digicel and Telefónica, follow closely, with 28%, 22% and 21%, respectively (Argumedo, 2007).

32 Interview with Mr. Baldir Garrido, President of the Union of Telecommunications Operators, March, 2005 (Urízar, 2007).

33 Ibid (p. 11).

that the incumbent imposes on other operators to access and interconnect to public networks. This difficulty is generally reflected in: (i) delays in interconnection negotiations; (ii) rates for termination of international calls traffic being higher than those of the market; (iii) difficulty in making payments in cascade between the different service providers, since the latter have agreed that to make telecommunications transfers between companies that do not have direct interconnection, the operator who carries out the function of traffic only will transfer the traffic if there is an interconnection contract between them. (iv) Another reason many operators experience difficulty in interconnection and access to essential resources is that the list set forth in Article 27 does not include key elements for the development of effective competition in the sector, such as the unbundling of infrastructure, joint location of equipment, rights of way and sole invoice service, among others. Thus, for example, operators have come up against difficulties in obtaining rights for the placement of posts or ducts, which are governed by the regulations of the different municipalities and are therefore subject to the discretionary authority of the mayor. (v) Another aspect that has hampered interconnection is that the incumbent operator offers only two interconnection points in the municipality of Guatemala, making entry difficult, since the entering operator must pay for transport of telecommunications to those points, even if it possesses the infrastructure to transport the communications by its own means (Urízar, 2007).

The de facto opening up to competition in Honduras promoted by the TpT has not responded to expectations, despite the progress made. In fact, when the project was launched, it was expected that towards the end of 2005, 200,000 lines would have been installed. By September only one third of these had been installed. Among the main reasons for this situation were: (i) initial interconnection problems that obliged sub-operators to wait approximately seven months after the programme was launched before the first of them went out on the market; (ii) lack of definition and selection of the technology to be used by sub-operators, and many of those that obtained registration as sub-operators acted more as a result of the momentum generated by the project or of false expectations of the telecommunications business than a thorough knowledge of the business (Tábora, 2007). An interesting aspect of the Honduran experience has been the technological variety applied by the different companies.34

IV. Industrial Organization and Competition in Telecommunications in Central America and Mexico

In the preceding section the paths followed by the opening up to competition in the different countries were compared. It was shown how the countries that accompanied privatization with the granting of a period of exclusivity have had more difficulties than the other countries in creating a competitive market. However, the non-existence

34 At the beginning of the programme, most of the interested parties focused on the use of wireless systems; however, due to the limited spectrum availability many rejected this system. It is worth

of institutions for the defence of competition and the inexperience and limitations of regulatory bodies have made the transition to competition difficult even in the countries that did not grant a period of exclusivity. The purpose of this section is to evaluate the forms acquired by the industrial organization of the sector and compare the different models and their results in terms of number of operators per segment and the dynamism of the different markets.

As in the rest of Latin America, the telecommunications market in Central America has become the object of dispute between the two dominant operators in the region, Telefónica España and the Mexican group TELMEX – América Móvil, part of the huge Grupo Carso.

Since the early 1990s, Telefónica España has been developing an aggressive policy of expansion in the telecommunications market in Latin America. This policy, which began in South America, has been progressively expanded to the north of the continent.35

In fact, Telefónica began mobile operations in El Salvador and Guatemala. Despite its interest in participating in the bid for INTEL of Panama, it was unable to do so, since that country's legislation prohibited the participation of companies in which foreign governments had a majority share.36 Subsequently, after Telefónica reached an agreement with Bellsouth in March 2004, to acquire all of its operations in Latin America, Telefónica took control of Bellsouth's operations in Guatemala, Nicaragua and Panama. In parallel, the company began operating in Mexico in 2002 by means of various acquisitions until it became the second operator in that country after Telcel. Thus, the company began to have an increasing presence in the territory of its major competitor at regional level: the TELMEX – América Móvil group.

This latter group, which originated in the privatization of TELMEX in 1990, had concentrated almost exclusively on its operations in Mexico until 1997. The backward state of telecommunications in that country, its enormous size and the threat posed by US companies, particularly in the context of NAFTA, justified the exclusive attention that the company gave to its country of origin. This policy began to change drastically at end of the 1990s.

1. Competition in Fixed Telephony

The problems of competition in this segment are evidenced by the incumbent's high share in these countries. Those which provided an exclusivity period have experienced greater difficulties in moving towards competition.

highlighting the variety of technologies used by sub-operators to undertake their projects: satellite and wireless access such as Teléfonos de Honduras; wireless access such as MULTIFON, UNITEL, CELTEL and Community Telephony, among others; copper networks such as INTELDATA; cable subscription television networks such as SULATEL, AMNET and Cable Color, among others; data transmission and switching networks such as METRORED; Informatics Access Networks (Internet) such as ITTSA, to mention a few cases (Tábora, 2007).

35 For a detailed analysis of this process, see Mariscal and Rivera (2005a, b).

36 As is known, Telefónica from Spain was completely privatized.

Up to 2004, TELMEX of Mexico had a 95% share of fixed telephony. Unfortunately, cable television operators had not developed the so-called Triple Play and hence, there has been practically no competition in broadband.

In Panama's case, entry to the fixed telephony segment has been very difficult. Even though by February 2005, 28 concessions for local telephony had been turned over, competition has concentrated on national and international LD and the business segment of local telephony. The difficulties encountered by the interconnection process have delayed the effective entry of competing companies, and therefore their presence is still minimal.

In Nicaragua, the control package was acquired by the consortium Telia Swedtel Ab – MEGATEL EMCE in 2001, while the state pledged to sell the remaining 49% it owned in a period of three years. In 2004, the privatization process was finally completed with the sale of the shares to the Mexican company América Móvil. That same year, seeking to expand its position in the regional telecommunications market, América Móvil took over the operator MEGATEL of Honduras, thus becoming the only shareholder of the Nicaraguan operator ENITEL. As mentioned above, the exclusivity period was expected to finish towards the end of 2004, despite a number of problems, among them the "competition" between the two telecommunications regulators, which was a sign of the country's serious political conflict. Yet, in 2006, ENITEL continued to be the monopolistic provider in fixed telephony (Ansorena, 2007).

In the case of Honduras, the end of the exclusivity period was scheduled for the end of 2005. The TpT programme, with the introduction of the concept of sub-operator, made it possible to move this process forward. By August 2005, according to CONATEL data, 40 companies had been registered as sub-operators. Of these 40 companies that complied with the requirements established in the process, 17 have begun commercial operations and it is estimated that to date they are operating approximately37 63,000 telephone lines commercially, around one third of which belong to only one sub-operator.38 Taking this information as a reference, to December, 2003, 100% of the fixed telephony market was in the hands of HONDUTEL. By December, 2004, with the entry of sub-operators this percentage dropped to 93.69%, the remaining 6.31% being in the hands of sub-operators. In August 2005, sub-operators had 13.83% of the total telephony market, whereas HONDUTEL maintained dominance with 86.17%. Three aspects should be underscored in this context. First, the great success achieved by the entrants, which contrasts with the cases mentioned before. Despite this progress, 98% of the lines installed were located in the country's three most important cities, deepening the coverage inequity between urban areas and rural areas, thus leaving the government to attend the social problems of access to telecommunications services (Tábora, 2007).

37 Even though the regulations issued by CONATEL for the programme "Telephony for All – Modernity for Honduras" establish compulsoriness in delivery of information on each operator's new subscribers, no evidence was found that this requirement was being complied with, and therefore the data on subscribers are estimated, based on the Tabora's own research.

38 MULTIFON, S.A. de C.V.

This situation confirms the incumbents' repeated claims about the policy of "cream skimming" developed by the entrants. Finally, the presence of a public operator who shares the government's interest in promoting the entry of other operators may have been a contributing factor in reducing conflicts with regard to interconnections. Furthermore, it should be taken into account that at least until 2006, sub-operators had not had to pay charges for the use of HONDUTEL's network. This regime was transitory and concluded on December 25, 2005. As of December 26, 2005, the regime of payment of access charges established in the Regulations for Interconnection was enforced (Tábora, 2007).

What happens in the case of countries which, together with privatizing the enterprise, opened up fixed telephony to competition?

In El Salvador ten companies participated, among which the most important were CTE-Telecom, Telefónica, Emetel, GCA Telecom, El Salvador Telecom, El Salvador Network, Telemóvil and Newcom. However, as we saw in the previous section, CTE's share continued being large (88.7% of the market). In spite of this, the immediate opening up to competition has allowed greater participation by entrants.

An analysis of the different national cases shows that over and above the differences displayed by the privatization processes, the transition to competition in fixed telephony has proved more complicated than expected. Indeed, beyond the intentions what the processes in the region show is that a public monopoly has become a private one.

If we analyse the fixed telephony industry from a regional point of view, the situation is even more complicated. As we can see in Table 3.1, in four of the five countries that have privatized the public telecommunications enterprise, the TELMEX – América Móvil group appears as the owner of the dominant enterprise in fixed telephony. Only in Panama is the dominant company owned by another group. The above-mentioned table also makes it clear that there is no other fixed operator with any capacity to generate competition in this segment. In this sense, as the regional telecommunications market consolidates, it is difficult to expect a competitive market to be generated by the countries' dominant companies. On the contrary, only the consolidation of the dominant position of Grupo TELMEX – América

Table 3.1 Central America and Mexico: fixed operators and market shares, 2004 (percentages) (Own preparation)

 

Grupo TELMEX –

 

 

 

Country

América Móvil

Telefónica España

Third operator

Fourth operator

Costa Rica

 

 

ICE (100)

 

El Salvador

CTE (88.7)

9.7

 

 

Guatemala

TELGUA

 

 

 

Honduras

 

 

HONDUTEL (86)

Several sub-

 

 

 

 

operators

 

 

 

 

(14)

Mexico

TELMEX (95)

 

Avantel

Alestra

Nicaragua

ENITEL (100)

 

 

 

Panama

 

 

Cable & Wireless

 

Móvil would seem possible. We will return to this topic in the following section on analysing the situation of mobile telephony.

In this context, it is likely that if HONDUTEL is privatized, Grupo TELMEX – América Móvil will have very strong incentives to seek to win the bid. Similarly, should telecommunications be opened up in Costa Rica, the possibility of ICE becoming a regional operator, taking advantage of its capabilities and the synergies created by the fact that it is also an electric power enterprise, is not very feasible.

Table 3.2 shows the evolution of the rate of penetration of fixed telephones in the seven countries included in our study. The excellent performance of fixed telephony in Costa Rica, the only country that has flatly rejected privatization of the enterprise, is one of the outstanding phenomena. Compared to Mexico, in 1990 Costa Rica had a penetration 1.55 times that of Mexico; in 2004, the difference widened to 1.84 times. A second interesting aspect is that the growth of coverage in fixed telephony in Costa Rica is very dynamic, between 1998 and 2004 it rose from 19.33 to 31.62.39

Mexico, in contrast with Costa Rica, shows an historical low penetration of fixed telephony (6.48 in 1990) and a very slow expansion of its coverage. It was only in 1998 that it reached 10.36. Interestingly, it was as of that year, one year after the opening up to competition, that the growth of teledensity accelerated. As in Costa Rica's case, although at lower levels, the growth of fixed telephony continues coming close to the South American countries, whose teledensities have a tendency towards stagnation. Nevertheless, compared to the level reached by per capita GDP, Mexico's penetration rate is low (see Fig. 3.1).

Table 3.2 Mexico and Central America: fixed telephone lines per 100 inhabitants, 1990–2004 (International Telecommunication Union (ITU) Online)

 

Costa Rica

El Salvador

Guatemala

Honduras

México

Nicaragua

Panamá

1990

10.05

2.42

2.13

1.72

6.48

1.26

9.27

1991

10.62

2.51

2.21

1.80

6.86

1.27

9.40

1992

10.89

3.22

2.29

1.93

7.54

1.37

9.75

1993

11.61

3.22

2.42

2.10

8.36

1.62

10.30

1994

13.15

4.26

2.51

2.27

9.18

1.99

11.11

1995

14.38

5.03

2.87

2.70

9.39

2.22

11.56

1996

15.47

5.61

3.30

3.10

9.28

2.63

12.16

1997

18.92

6.08

4.08

3.77

9.69

2.75

13.44

1998

19.33

6.41

4.79

3.99

10.36

3.01

15.13

1999

20.41

8.05

5.51

4.42

11.22

3.04

16.43

2000

22.34

9.96

5.94

4.61

12.47

3.12

15.11

2001

22.97

10.15

6.47

4.74

13.72

2.94

12.99

2002

25.05

10.34

7.05

4.80

14.67

3.20

12.20

2003

27.77

11.34

7.05

4.87

15.97

3.74

12.20

2004

31.62

13.42

8.94

5.31

17.22

3.83

11.85

39 As is known, there is a generalized tendency in Latin America, despite the low teledensity, for fixed telephony to stagnate, and even to deteriorate in recent years.

Image

Fig. 3.1 Central America and Mexico: penetration in fixed telephony (International Telecommunication Union, 2005)

The cases of El Salvador and Panama are particularly interesting due to the contrasts shown. Whereas in 1990 teledensity in El Salvador barely reached 2.42, in Panama it stood at 9.27, just behind Costa Rica. In 1997, when Panama's public enterprise was privatized, the penetration rate was 13.44; in 1999, this rate reached a maximum level of 16.43; and the following year it began a continuous decline until in 2004 it reached a teledensity of only 11.85. In stark contrast to this, in El Salvador teledensity grew continuously until it surpassed Panama in 2004, when it reached a figure of 13.42. These different paths cannot be explained by uneven growth in mobile telephony. In fact, as we shall see further on, the number of subscribers per 100 inhabitants was, to 2004, fairly similar in both countries, although El Salvador slightly surpasses Panama. The differences suggest, rather, that the competitive atmosphere favourably affects rates and this has a positive impact on teledensity.40 The performance of the other three countries has been very modest. In 2004, the level of penetration in Guatemala was marginally over 8%. Honduras was barely approaching 6 points. In Nicaragua, the benefits of privatization had simply not become apparent.

Table 3.3 shows a synthesis of what has been discussed in this section and helps to evaluate the development of fixed telephony over the last 15 years. The countries which have liberalized the telecommunications sector, after an exclusiveness period, are characterized by a greater participation of the incumbent than the one it has in those countries which simultaneously liberalized and privatized this sector. Mexico is, in this sense, a paradigmatic case because the main operator had 95% participation

40 The next section shows that whereas rates stood at an average of US$0.06 for 3 min in El Salvador, in Panama they reached 12 cents.

Table 3.3 Central America and Mexico: privatization, liberalization and performance trajectories, 1990–2004 (percentages) (Author, on the basis of official figures)

 

Openness
mode

Coverage
2004

Incumbent
participation

3 min local call
price, 2001

Coverage
growth

Mexico 1990

Exclusivity

17.22

95.0

0.16

2.42

Panama 1996

Exclusivity

11.85

100.0

0.12

-3.28

Nicaragua 2000

Exclusivity

3.83

100.0

0.10

0.91 (three years)

El Salvador 1998

No Exclusivity

13.42

88.7

0.07

7.01

Guatemala 1998

No Exclusivity

8.94

83.6

0.08

4.20

Costa Rica

Public Monopoly

31.62

100.0

0.03

12.29 (1998–2004)

Honduras 2003

Public Monopoly (2003)

5.31

86.0

0.07

1.32 (1998–2004)

even eight years after the market opening up. In contrast, in El Salvador and Guatemala the participation of the main operator reaches around 15%.

If the coverage is considered, Mexico appears with a low one, at least in comparison with Costa Rica. In fact, its higher income per capita does not show in its teledensity. The setback experienced by Panama was already mentioned. As to prices, these are higher for the 3 min call in those countries which have carried out privatization with an exclusiveness period. Table 3.3, illustrates that the average growth rate during the first three years of privatization are substantially higher in those countries that went through a simultaneous privatization and liberalization process.

The Costa Rican experience is unique: the price of local calls is the lowest of the region, has the highest penetration rate and shows a significant coverage increase rate. Even though greater efficiency is not the main reason for this situation,41 Costa Rica certainly gives a different perspective regarding privatization policies and underlines the need for competition regulatory framework that guarantees a competitive post-privatization market.

2. Competition in Mobile Telephony

As in the case of fixed telephony, mobile telecommunications show different forms of development.

In Mexico, the strong presence of the TELMEX – América Móvil group is also registered in mobile telephony. As we can see in Table 3.4, Telcel's share reached more than 77% in 2004. This telecommunications segment, as we saw above, was conceived

41 We will come back to this point in the following section.

Table 3.4 Central America and Mexico: mobile operators and market shares (Author)

Country

América Móvil

Telefónica Móviles

Millicom

Fourth operator

Costa Rica

 

 

 

ICE (100)

El Salvador

CTE Telecom (28.0)

21.0

Telemóvil (29)

Digicel (22)

Guatemala

52.8

22.4

COMCEL (24.4)

 

Honduras

32.0

 

CELTEL

 

Mexico

77.2

15.1

 

Unefon –

 

 

 

 

Iusacell (7.7)

Nicaragua
Panama

51.0

49.0
37.5

 

 

from the start by public policy as a competitive sector. Nevertheless, in contrast to the majority of countries, in which there has been a relatively balanced market distribution, in Mexico the dominant company has tended to strengthen its presence. In 1997, Telcel's share was 63.7%, Iusacell's 23% and other competitors 13%. In the following years, Iusacell began to lose market share. An important factor was not including prepayment among its products. The company was acquired by the consortium made up of Vodafone, one of the world's main mobile companies with operations in more than 25 countries, and Verizon. In 2003, however, these companies abandoned their operation in Mexico after experiencing significant losses. The entry of Telefónica Móviles has revitalized competition in the market. The stagnation in entrants' share continued despite the sharp increase in mobile teledensity in Mexico in recent years.

In Guatemala the influence of the dominant fixed company has also been felt in the industrial structure of the mobile segment. Thus, Grupo TELMEX – América Móvil has attained a share of almost 53% of total subscribers. Nevertheless, the presence of important international companies such as Telefónica and Millicom (see Table 3.4) makes it possible to foresee competitive development towards the future.

El Salvador is undoubtedly the country that shows the promising development of competition in mobile telephony. Indeed, as can be seen in Table 3.4, four operators share the mobile market in a fairly balanced manner.

In Honduras, the original strategy at the time the LMT was approved was to have three operators in the short term: (i) Telefónica Celular, S.A. (CELTEL); (ii) the winner of the capitalization process of HONDUTEL; and (iii) the winner of the bidding for Band B, on 800 MHz (Tábora, 2007). However, the consequences were the difficulties encountered to incorporate a private investor in HONDUTEL and the failure of the bidding for Band B until 2002. On January 6, 1994, the concession for development of Cellular Mobile Telephony Services in Honduras was awarded to the companies Motorola, Inc., Millicom International Cellular, S.A. (MIC) and Proempres, S.A., represented by the company CELTEL. This agreement was signed by both parties on August 7, 1995. The concession granted CELTEL the right to develop mobile telephony services within Honduran territory, using frequency Band "A," as of June 1996. The company officially began its services on September 15, 1996. In 2006, CELTEL offered its services in 13 departments and in the country's ten main cities. Until 2003, the number of subscribers grew slowly until it reached 5.5% coverage.

In 2002, CONATEL successfully carried out an international bidding process for Frequency Band "B." Eighteen companies expressed an interest in taking part. In April 2003, three companies were pre-qualified, Empresa Nacional de Telecomunicaciones (ENTEL) of Chile, Bellsouth Honduras and the local consortium, MEGATEL EMCE, which had been jointly awarded with Telia Swedtel the Nicaraguan enterprise ENITEL. However, the first two declined and consequently the company was awarded to the consortium Telia Swedtel – MEGATEL EMCE for a price of $7.1 million.42 The impact of competition on the growth of the number of subscribers has been significant. Mobile teledensity stood at 10.15% in December 2004, reaching a penetration of 14.82% in August 2005 (Tábora, 2007). The performance of competition has been successful. Having begun operations in November 2003, in August 2005, MEGATEL had a share of 31.8% (Tábora, 2007). This may reflect the fact that there are no relations between the incumbent enterprise in fixed telephony and the company that operated the first concession in mobile telephony.

In contrast to fixed telephony, mobile telephony looks more promising from a competition point of view at the regional level. Indeed, three operators have a presence in more than one market, specifically, Telefónica Móviles, América Móvil and Millicom. In addition, other companies such as Cable & Wireless in Panama and Digicel in El Salvador are in operation. Hence, there is an entrepreneurial base that can significantly stimulate competition in this sector.

The presence of several operators largely explains the more dynamic development of this sector. If we look at Table 3.5, it is evident that as of 1998–1999, teledensity began to grow at very high rates. Crucial features of this process are the introduction of the "caller pays" system and prepayment, which made it possible to broaden the services to groups that were not eligible for post-payment programmes. The two countries with the most dynamic growth are undoubtedly El Salvador and Mexico which rose from a penetration rate of 0.68 and 1.82 in 1997 to 27.61 and 36.64 in 2004, respectively.

Costa Rica, in contrast, despite being the country in 1997 with the broadest coverage (except Mexico), in 2004, only surpassed Nicaragua and Honduras. As we shall see further ahead, this appears to be associated with cross-subsidies from mobile to fixed telephony.

V. Telecommunications Rates in Central America and Mexico

For an analysis of prices of telecommunications services in the region a distinction should be made between the fixed segment, subject to more or less strict regulations, and the segments of national and international LD and mobile telephony, under

42 Representatives immediately pointed out that the process should be annulled as only one bidder took part, see Honduras Revista Internacional (http://hondurasri.com/CLON/detalles/TRIBUNA/SER%C1%20ANULADA%20LICITACI%D3N%20DE%20TELEFON%CDA%20M%D3VIL%20PCS.htm).

Table 3.5 Mexico and Central America: subscribers to cellular telephony per 100 inhabitants 1990–2004 (International Telecommunication Union, 2005)

Country

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

Costa Rica

0.00

0.00

0.10

0.14

0.21

0.56

1.37

1.78

2.83

3.51

5.10

7.57

11.10

18.12

21.73

El Salvador

0.00

0.00

0.00

0.03

0.09

0.24

0.40

0.68

2.27

8.31

11.85

13.40

13.76

17.32

27.71

Guatemala

0.00

0.01

0.02

0.03

0.11

0.30

0.42

0.61

1.03

3.05

7.53

9.81

13.15

13.15

25.02

Honduras

0.00

0.00

0.00

0.00

0.00

0.00

0.04

0.23

0.56

1.24

2.39

3.64

4.86

5.53

10.10

Mexico

0.08

0.18

0.35

0.42

0.62

0.73

1.07

1.82

3.50

7.94

14.24

21.68

25.45

29.47

36.64

Nicaragua

0.00

0.00

0.00

0.01

0.05

0.10

0.12

0.17

0.39

0.90

1.78

2.96

4.47

8.51

13.20

Panama

0.00

0.00

0.00

0.00

0.00

0.00

0.26

0.68

3.10

8.27

14.45

16.40

18.95

26.76

26.98

systems both regulated and subject to competition. In these latter cases, one situation or the other is associated with the existence or absence of exclusive concessions.

For a comprehensive analysis of rate regulation for fixed telephony in the region it is essential to analyse the mechanism for initial setting of rates at the time of privatization, the procedure defined for their establishment as a system and finally, the effective path followed by rates and prices throughout the period.

In the case of mobile telephony, its competitive nature, the existence of two separate markets, the prepayment and post-payment markets, the multiplicity of rate plans and diverse variations within each one and constant innovations in promotions make it difficult to follow-up on price policy. Despite the foregoing, this section gives a general overview of prices and rates, seeking to highlight the main characteristics of their evolution in recent years.

1. Rate Regulation and Evolution in Fixed Telephony and Long Distance

In the case of TELMEX in Mexico, the concession licence determines a form of "authorized rate control," i.e. the SCT authorizes the rates set by the companies in line with the following criteria: (i) the rate structure will seek to favour efficient expansion of the public telephone network and lay the foundations for healthy competition in the provision of services; and (ii) the rates applicable to each service should make it possible to recover at least the incremental long-term cost,43 in such a way that cross-subsidies between services are eliminated.44 This is so that there is the necessary incentive to expand each service and establish fair grounds for equitable competition. The rate structure should lead TELMEX to achieve continuous improvement in productivity to enable it to increase its profitability, which in turn should translate into lower rates for users.45

If the tariff rate of growth of a basket of controlled services is compared with the inflation in 1997, the year of the opening up to competition, the former exceeds the latter considerably. Up to 1998, inflation was higher than fixed phone rates, which meant a fall in real rates. This trend continued strongly up to 2002 and 2003, when the nominal rates finally decreased.

43 Average long-term incremental cost means the sum of all the costs TELMEX has to incur to provide a unit of additional capacity of the corresponding service. Incremental costs should be comparable to those of an efficient company, in such a way that regulated rates are internationally competitive.

44 It is understood that there is a cross-subsidy when a company provides a service with an insufficient rate to cover average long-term incremental costs and simultaneously provides another service with a rate higher than its average long-term incremental costs. The above condition will be applied as long as the difference is substantial.

45 For details of the manner in which the concession licence sets forth the price cap system, see Chapter 4.

According to information from COFETEL, the above has translated into significant reductions in the prices of services. Indeed, in June 2004 in the residential sector, the rate for installation had dropped from 1,393 pesos in 1998 to 1,130 pesos in June 2004, representing a reduction equivalent to 18.9%. Basic rent, for its part, had gone from 182.56 pesos to 156.55 pesos, a reduction of 14.2% in the same period. Finally, measured local service dropped from 1.72 pesos to 1.48 pesos, a reduction of 14%. In the business sector, installation costs dropped from 4,991.85 pesos to 3,500 pesos, a 29% reduction; basic rent from 251 pesos to 191 pesos, a 21.1% reduction; and finally, measured local service from 1.72 pesos to 1.48 pesos, a 14% drop, all in the above-mentioned period (Mariscal and Rivera, 2007). In this context, it is noteworthy that the distinction between residential and business service has been maintained, but perhaps what is most important is that the price reductions of these services are not what could have been expected if the more pronounced drops in the cost of equipment and the reduction of unit costs are considered. Additionally the significant increase in the number of users should have helped to further price cuts.

The above appears to be confirmed in Table 3.6, which shows that Mexican local rates exceed those of all the other countries considered in this study. The same occurs when we compare rates in Mexico with those of the other OECD countries. According to a OECD study, up to August 2004, on comparing the costs of a basket of telecommunications services which includes local calls, national and international LD and calls to mobile phones, it is evident that according to the purchasing power parity in US dollars (USD PPP), the cost of the Mexican basket was 611.74, surpassed only by Czech Republic (707.03) and Poland (699.53). The lowest costs were those of Iceland (259.30), Denmark (259.99), Canada (298.29) and Switzerland (306.19).46 The average cost of the basket in the OECD was 428.62 (OECD, 2007).

In El Salvador, maximum rates for public telephone service are determined and approved by SIGET, without detriment to the provisions of Article 108.47 The overall rate for telephone service is subdivided into: (i) access charge, which is the payment

Table 3.6 Mexico and Central America: average cost of local call (dollars per 3 min) (World Bank)

Country

1997

1998

1999

2000

2001

2002

2003

Costa Rica

0.03

0.03

0.03

0.03

0.03

0.03

0.02

El Salvador

0.05

0.06

0.06

0.06

0.07

n/a

n/a

Guatemala

0.11

0.1

0.09

0.09

0.08

n/a

n/a

Honduras

0.06

0.06

0.06

0.06

0.07

0.06

n/a

Mexico

0.14

0.13

0.14

0.15

0.16

n/a

n/a

Nicaragua

0.04

0.04

0.09

0.11

0.1

0.08

0.08

Panama

0.12

0.12

0.12

0.12

0.12

n/a

n/a

46 It should be pointed out that the small OECD countries have even lower costs for the basket.

47 Article 108 permitted an increase in the rate for fixed telephony based on the operator's investments.

the user makes for being connected to the network; (ii) charge for call traffic (time of use) or services (alarm call, directory enquiries, etc.); and (iii) interconnection charges, plus charges for finishing the call in another network.

Before privatization the rates were set by ANTEL, an institution that between 1980 and 1994 did not adjust the prices charged for traffic in minutes, nor the charges for residential and business access, nor the network installation (FUSADES, 1998), despite the fact that during that period inflation increased by more than 1,000%. The deficits generated were compensated, as was usual at the time, by the income from the high rates charged for international communications. In 1995, the first increase in fixed telephony was registered, increasing the cost per minute of a local call by 200%. The access charge for residential and business telephony rose by 400%, and the installation charge increased by 33.3%. A new rebalancing was applied a year before the privatization, with the rate for a local call experiencing an increase of 200% and the adjustment for the residential charge and installation being lower. International calls made to the USA (the main country of destination) were reduced by 50% in comparison to those of 1994; it should be pointed out that the participation of Americatel was approved in international LD as of the early 1990s (Argumedo, 2007).

The Law established the following mechanism for adjustment of rates: (i) for access charges, a collection base was established which would be adjusted annually in line with inflation; (ii) in the case of interconnection charges, a collection base was determined by type of terminal, which would be adjusted on a quarterly basis according to the inflation rate and devaluation with regard to the USA48; and (iii) in regard to the charge for traffic, a clause was established to allow the rate to be increased as the operator invested more in extending coverage and expanding the network.49 Thus, rate increases were generated despite the fact that they were falling at international level. The continuous increase in rates contributed to the disillusionment of the population with regard to the privatization, since during its planning it had been maintained that competition would make the rates go down (Argumedo, 2007).

If only the inflation rates had been applied, the charge for 1 min access would have reached $0.0185; if, on the other hand, the provisions of the investment contract had been complied with and the maximum increases had been applied, the price would have risen to $0.041/min in 2002 (an increase of 95% over the 1998 price). What actually happened with the rate was that at first the goals for expansion of lines were not met, and therefore the increase authorized was proportional to the effective growth of lines. In June 2002, SIGET and the operator agreed that the increases to which the company had a right were not viable, and they would seek to introduce them in a period of seven years (Argumedo, 2007).

48 In practice the adjustment was only for inflation, since the colon-dollar exchange rate was fixed since 1993 and in 2001 the country implemented dollarization.

49 Based on the existence of 340,000 lines in service in 1997, the investment contract provided for the following rate increases: in 1998 15% on installing 50,000 lines; in 1999 20% on installing 65,000 lines and an annual increase of 20% between 2000 and 2002 as long as 75,000 lines were installed in each of those years.

That is how the rates for fixed telephony have experienced rises, showing the lack of competition and the problems stemming from the investment contract. In the period 1998–2004, there were annual average increases of 4% and 2.5% in basic residential and business charges, respectively. The cost per minute of local calls and national LD, for their part, increased over the past seven years by 5.2% and 4.7%, respectively. The cost of national LD calls reached a figure of $0.04/min, i.e. 40% of the cost of calls to the USA, which could be a sign of market power abuse (Argumedo, 2007).

In the case of Guatemala, the General Telecommunications Law establishes that the rate for consumer, as well as those charged among telecommunications operators, should be freely agreed upon. Government Accord 394–200150 put an end to the actions under Administrative Law Nos. 185–2000 and 186–2000 promoted by the Government against the privatization process. Article 2 establishes that the rates for residential and business services for fixed lines until 2003 should be as follows (using the exchange rate Q. 7.84/dollar): local telephony Q. 0.2036/min ($0.02597/min) and LD telephony Q. 0.3564/min ($0.04546/min).

In 1995, LD rate for calls to the USA, the most important destination for Guatemalan communications, was around $1.50/min, with poor quality and high saturation of calls. In mid-2005, some operators offered rates of $0.10/min.

The rate policy guidelines for Telecommunications Services in Honduras are established in Title Three, Chapter II (Article 31) of the LMT. The latter determines that the rates charged by telecommunications services operators, except broadcasting services, shall be regulated by CONATEL, as long as such services are provided in adequate competition conditions.

Although there was no rate rebalancing in the country,51 the government that took office in 2002 decided, together with CONATEL, to reduce LD rates. The rates effecting place in 2002 for communications to the USA (which represent 91% of all international calls) stood at $1.24/min. That year the rate was reduced to $1.04 and in 2003 it was reduced once again, this time to $0.84.52

50 Diario de Centro América. October 5, 2001–Number 62–Publication of Government Accord 394–2001 of October 1, 2001, Guatemala.

51 Since April 6, 1999, the date on which CONATEL approved HONDUTEL's List of Rates, there has been no rate rebalancing whatsoever. Article 19 of the Regulations on Costs and Rates for Telecommunications Services establishes that there is a period for carrying out the rate rebalancing, which closes on December 31, 2005. The idea of this period was to comply with the contractual conditions that had been established in the capitalization process of HONDUTEL, in order to guarantee the new operator a rate structure that would allow it to recover its investment and, periodically reduce the cross-subsidy between international LD service and the other services. However, since the process failed, these contractual conditions never went into effect, and this provision was not applied (Tábora, 2007).

52 The idea of this plan was for the full rate to the USA to be US$0.42/min by the end of 2004, so that at the time of the opening up the rate should be US$0.28/min, thus making the impact on the market of the opening up less than it would have been at the time. It was not possible to obtain an official explanation for the reason why this plan was not continued, but it may be understood that the failure to apply the plan was influenced by the lack of a clear vision of the future of HONDUTEL, the entry into the market of sub-operators under the TpT programme and above all because of the difficulties of a fiscal nature being experienced by the current government (Tábora, 2007).

The high international LD rates undoubtedly contrast with local rates. As can be seen in Table 3.6, the cost of a 3 min call was $0.06, the lowest in the region after Costa Rica. It should be pointed out that the above data constitute an average, since in Honduras a distinction continued to be made between residential and business rates.53

HONDUTEL has introduced the system of prepayment in fixed telephony. In 2003, it launched Telecard on the market, which may be used in public telephones and in any terminal of its network. Telecard comes in denominations of 20, 60, 125 and 250 lempiras and may be used for local calls, national and international LD and communications to mobile telephones.

As we saw in previous sections, together with the privatization of INTEL, S.A. Panama granted the telecommunications company a period of exclusivity that was to last until January 1, 2003. Unfortunately the bidding process did not include definitions defining the rates each of them would charge. As González (2007) points out, it would have been logical, once the benchmark price for 49% of the shares had been set for the auction, for the process to favour the bidder who offered to charge the lowest prices, with precise quality requirements for the services to be offered.

However, the rates were negotiated after the concession contract had been concluded. Prior to the privatization, the cost of basic local telephony was a fixed charge of $10.50 with no additional variable cost. After privatization a fixed charge of $6.27 was established for the basic plan and a variable charge of $0.03/min. Table 3.6 shows the average costs of a local call. As can be seen, the prices in Panama were the highest in the region, with the sole exception of Mexico.54

Privatization with exclusivity also had as a consequence high prices in LD telephony. Between 1997 and 2002, a national LD call had a cost of $0.15/min. The opening up to competition reduced the price to $0.13/min in 2003, to $0.09/min in 2004 and was expected to fall in 2005 and 2006 to $0.07/min and $0.05/min, respectively (González, 2007). The costs of exclusivity are also reflected in international LD. The average prices of international calls excluding those to the USA reached a price of $1.86/min in 1998, $1.69/min in 1999 and 2000, and $1.57/min in 2001 and 2002. The opening up to competition evidenced how unjustified such high rates are. In fact, in 2003 the rates fell to an average of $0.40, in 2004 to an average of $0.25 and they were expected to fall to $0.12 in 2005. International calls to the USA cost $0.98 in 1998, $0.89 in 1999 and 2000 and

53 For details of the evolution of fixed rates in the period 2000–2004, see Tábora (2007).

54 The figures presented by González (2007) differ from those of the ITU and the World Bank. According to the aforementioned institutions, in 1997, the monthly charge for residential telephony was US$10.00 and in 2001 it was US$3.00. According to the World Bank, before and after the privatization the average cost of a local call remained at US$0.12 for 3 min, i.e. US$0.04/min. The difference may lie in the fact that the ITU figures for the monthly charge in residential telephony do not include business telephony, whereas the World Bank figures show an average of residential and business telephony.

$0.85 in the two years prior to the opening up. In 2003, the first year of competition, the rate dropped to $0.18, to $0.15 in 2004 and it was expected to reach $0.07 in 2005 (González, 2007).

As we saw before, in Costa Rica telecommunications are in the charge of ICE, a public enterprise that also provides electric power generation and distribution services. It is the responsibility of the Regulatory Authority for Public Services (ARESEP) to define the rates for services and supervise their quality. The legislation establishes that the services should be provided in accordance with their cost. Rates for telecommunications services are set under the rate system of the rate of return. In Costa Rica, this system translates into a rate that covers costs and the rate of return on capital (or development rate), understood as the income necessary for ICE to be able to replace or increase the assets it needs to provide its services. In operative terms, ARESEP applies the system of "price caps" to adjust ICE's rates. In order to guarantee that service quality is maintained, an adjustment factor for quality is included that affects only the component of surplus or return for the application of the formula, so that the resources to cover the costs of providing the service are guaranteed.

ARESEP has approved rate increases lower than those requested by ICE, with the purpose of inducing the firm to lower costs and increase productivity. The outcome of this policy during the period 1997–2005 has been a reduction in the basic residential rate equivalent to 27% (Vargas and Hernández, 2007). Moreover, as can be seen in Table 3.6, the average cost of a 3 min local call was $0.03 until 2002, dropping to $0.02 in 2003, thus making it the lowest average rate at that time among the countries considered in this study.

There is wide-ranging debate as to the causes of the Costa Rican peculiarities. A first argument maintains that as a public enterprise, ICE faces lower differential costs than a hypothetical private operator because they do not include the earnings or normal profitability of all economic activities (also known as retribution for the entrepreneurial factor or opportunity cost) because this kind of enterprise cannot generate profits. Also contributing to this situation would be the fact that its nature as a public enterprise exempts it from payment of income tax, other municipal taxes, as well as the right of way on public highways and use of the radio spectrum. As the only operator in the market it also enjoys significant economies of scale. Another feature that contributes to explaining the low rates stems from the synergies the enterprise has by also operating the electric power distribution networks, such as those derived from posts for stringing electrical, telephone and optical fibre cables. Cross-subsidies also play a role. As in the past in the majority of countries, residential telephony was subsidized by the higher rates for fixed telephony service to companies (business) and international telephony. The considerable reduction in rates for international telephony at world level, together with technological progress – which has made possible various forms of contracting this service from the country with companies other than ICE – has obliged the enterprise to seek alternative sources to finance cross-subsidies. Thus, at present mobile telephony has become the source of financial resources to subsidize less-profitable telecommunications services (Vargas and Hernández, 2007).

2. Evolution of Mobile Telephony Rates

An analysis of mobile telephony rates is substantially more difficult than in fixed telephony. What is involved, in general, is a market open to competition and therefore, normally, without the obligation to report rates. It is also a telecommunications service segmented between the post-payment and prepayment markets, often with sharply differentiated rates. Also, since there is considerable competition in this market, each one presents a wide variety of plans with many differences in nature, making it difficult for users to choose among them.

In the prepayment market, competition can take place in relation to prices per minute of airtime, as well as in connection with other variables such as: denomination of the card, duration of such denomination, grace period to recover available balances when its duration expires, and promotions and services offered by the prepayment system. Furthermore, the strategy followed by the licences has been to differentiate themselves to avoid open competition in the majority of the above variables (González, 2007).

In his analysis on competition in telecommunications in El Salvador, Argumedo (2007) draws four conclusions: (i) calling the USA is cheaper than communications between cellular phones in the country; (ii) a call from a cellular phone to a fixed line is more expensive than from cellular to cellular; (iii) calls between cellular phones of the same company are cheaper than between different companies; and (iv) calls from post-payment cellular phones are lower than those made by prepayment.

In Guatemala, the existence of greater competition among the operators has given consumer prices an interesting dynamic. In October 1999, Telefónica entered the market and immediately introduced the "caller pays" system and a 70% reduction in the price of outgoing calls. The entry of Telefónica also contributed to altering the characteristics of competition by diversifying rate plans, introducing post-payment or prepayment systems and offering individual and business (corporate) plans. By January 2005, 88% of mobile telephony was in the form of prepayment and only 12% in post-payment (Urízar, 2007).

In Honduras, the concession contracts of the two mobile operators CELTEL and MEGATEL established rate ceilings for mobile telephony services. Three charges were identified: the basic monthly rate which was set at the equivalent of $30, the charge at full rate set at $0.25/min and the charge at reduced rate set at $0.18/min. These rates set first for CELTEL were to last ten years (until 2006). Operators had the freedom to determine the duration of the different rates and even the possibility of introducing a super-low rate. Even though nominal rates have remained fairly stable in an analysis of consumption per subscriber, in real terms rates went down 31% in 2003–2005. Additionally, on average, during the latter period users used this service approximately 44% more than the time they talked in 2002 and for the same amount of money (Tábora, 2007).

In the prepayment market, CELTEL offered cards in seven denominations, between 25 ($1.31) and 450 ($23.68) lempiras, which are differentiated by the number of minutes and the duration but have the same cost per minute–$0.2632. In the post-payment market, CELTEL offered five plans that include between 60 and 400 min. The prices per minute varied from $0.25 55 (Plan Móvil 15–60 min for a total price of $15) to $0.19 (Plan Móvil 75–400 min for a total price of $75). MEGATEL, Honduras's other mobile telephony operator, had six post-payment plans and their prices per minute ranged between $0.27 (Plan Alo 12–44 min for a total cost of $12) and $0.19 (Plan Alo–525 min for a total cost of $100).56

In Panama's case the companies compete first, in the card denominations. Whereas Movistar has cards worth US$2, 5, 10, 20 and 40, Cable & Wireless has US$5, 10, 15, 30 and 50.57 The price per minute also varies. In Movistar's case, the company had only two rates. For the US$2 card, the price per minute was $0.49. For the rest of the cards the price was $0.43/min. Cable & Wireless, for its part, has a greater variety of rates per minute: US$0.40, 0.34 and 0.32 for the US$5, 10 and 15 cards, respectively, and US$0.25 for the US$30 and 50 cards. Naturally, in the card denominations in which the two companies coincide, users receive a different amount of minutes.

In the case of the US$5 and 10 cards, the licencees do not compete with regard to the duration, as both of them last between ten and 30 days, respectively; nor do they compete with regard to the number of grace days for recovering the balance: their cards offer 15 days to recover balances. Market promotions have adopted the means to double, triple, etc. the value in minutes of each card denomination. According to González (2007), this is the most visible way in which both licencees compete in Panama.

In Costa Rica, as we saw in the preceding section, mobile telephony has replaced national and international LD telephony as a source to finance for other telecommunications segments, particularly residential fixed telephony. Mobile telephony, despite its lesser coverage, contributes more than half of ICE's total revenues for telecommunications services. These circumstances may of course explain the relatively low coverage of mobile telephony in Costa Rica compared both with the levels attained by other countries and with the coverage of fixed telephony (see Section IV).

55 An interesting detail in the prepayment form is that the cards already include 12% sales tax (ISV), and therefore the real nominal rate is US$0.235, an amount lower than the reference rates for the post-payment plans Móvil 15 and Móvil 20, which do not include ISV. This analysis shows that the prepayment rate (US$0.235/min) is less than the post-payment rate of the Móvil 15 (US$0.28/min) and Móvil 20 (US$0.2635/min) plans (Tábora, 2007).

56 In addition to the traditional plans, MEGATEL offers the "Control ALO" Plan, whereby a credit limit on consumption during the month is established (Basic Set Rate) and once the assigned credit limit is reached, the user has the option of entering an Alo Card to continue using the service. With this mechanism subscribers can ensure that their monthly bill will always be for the same amount, with the proviso that any excess will have to be paid in cash by means of prepayment. One of the main advantages offered by this plan is that it accumulates any unconsumed minutes for the following month, including both the minutes assigned and the minutes entered with an Alo card (Tábora, 2007).

57 The information corresponds to April 18, 2005 (González, 2007).

VI. Conclusions and Policy Recommendations

Privatizations and openness of fixed telecommunications promised a world of many operators, a substantive increase in quality and coverage of telecommunications services, as well as a drop in prices. The coverage and quality of these services have improved, but their expansion has been slower than expected, and even Costa Rica – a country which has persevered in keeping these services under the public ownership – has better coverage and prices than the rest of the nations included in this study.

Two central elements help understanding this situation. First, private monopolies have shown a remarkable strength to resist competition. Second, the usual shortcomings of sectoral regulation are greater in smaller economies, particularly in developing ones where two basic institutions – law enforcement and the markets – are weak.

In this context, to discipline the incumbent is quite difficult and high prices limit the population's possibilities to use this service. This is specially so for low-income sectors. Together with the privatization and openness path followed, the fixed telephony problems are associated with an institutional underdevelopment.

The different paths taken by the privatization and liberalization process of fixed telecommunications were a promise of a world with many operators, a much greater coverage and better quality of the service, as well as lower tariffs, especially as regards the granting or not of exclusivity periods, which seem to translate into significant differences when the intensity of competition and the prices of telecommunications services are analysed. The case of El Salvador shows that the opening up to competition in fixed telephony, without a period of exclusivity, has translated into a much higher participation of companies other than the incumbent. By contrast, the exclusivity period allows the consolidation of a non-competitive model which tends to become more pronounced due to institutional weaknesses that allow anti-competitive behaviour by the incumbent, especially during the period of opening up to competition.

Separate privatization of the fixed company and the mobile company appears to constitute a measure of singular importance in promoting competition. In fact, the application of this policy in El Salvador has resulted, despite the country's small size, in the presence of four mobile operators with well-balanced market shares.

The formal presence or absence of a competition agency is not sufficient to make a difference. In fact, in countries which have a competition agency with a history close to, or greater than, ten years (Mexico and Panama), the predominance of the incumbent company does not differ substantially from what happens in countries without a competition authority. This seems to be associated with the weakness of the competition agency.

The mobile telephony represents a very different situation. Its technological characteristics and an active public competition policy made the participation of many operators possible. The intense competition in this industry had an impact on prices and promoted innovative commercialization techniques, allowing very low-income customers to participate in this market. Prepayment reduced the access cost to this service and the business model followed by this industry focused on massive use rather than monopolistic prices to raise its profits. From the users' point of view, the prepayment mechanism is helpful, because expenditure is always under their control and when they lack resources to make a call, they can always receive them. Many transaction costs are eliminated for the enterprise and its customers. In this way, without the need of public subsidies, mobile telecommunications is making headway towards universal access to voice communication services.

Within the net and services convergence framework there is the possibility that the strong competition in mobile telephony may be transmitted to the rest of the industry. This would provide important benefits for consumers and improve firms' competitiveness. Nevertheless, high entry barriers together with weak competition regulation may reinforce the increasing importance of regional duopolies in telecommunications, such as Telefónica y TELMEX – América Móvil, and offset such positive tendency mentioned above.

Tariff regulation has to make sure that competitive market pressures are simulated for the fixed telephony incumbent, if not naturally existent. Price reductions to near competitive levels can increase/generate substantive coverage. This is particularly important in fixed telecommunications because it provides private access to the wide band. This is particularly true since fixed telecommunications still offer important advantages over mobile telecommunications for these purposes.58 Independently from the arguments that favour the total liberalization of this sector, sectorial agencies need to regulate this sector in the best possible way in the medium run.

Regulation, though, is not enough. Competition promotion is increasingly important. It is particularly relevant to eliminate barriers to entry permanently (deregulation and surveillance of interconnection blocking or overpricing, among other anti-competitive behaviour), to control mergers as well as other measures to ensure price competition. For these measures to succeed, four elements are crucial: a strong competition agency, a strong coordination between the latter and sector regulation, improvement of the judiciary system performance in these matters and the development of a supranational competition legal system.

In Costa Rica's case, the future of telecommunications is linked to the more global decisions the country takes after the presidential elections of February 2006. The fundamental decision is undoubtedly the one relative to ratification of DRCAFTA. If approved, Costa Rica should approve a General Telecommunications Law, create a regulatory agency for the sector and open up Internet, private networks services and wireless telephony to competition.

58 Mobile telephony is a very good option for voice transmission, but in the short run it is not a comparable competitive alternative for Internet for the population at large. This will probably change in the longer run.

The free trade agreement does not include the obligation to proceed to the privatization of ICE. Nevertheless, those in favour of maintaining its public nature fear that with the opening up the future of the enterprise may be threatened definitively by competition from international operators, particularly Telefónica España and the TELMEX – América Móvil group.

It is in this context that large telecommunications firms are readjusting their organization structures. As in the rest of the world, there are clear signals that fixed telephony has ceased to be the nucleus of the operation, as it has lost importance as the main source of income and its traditional monopoly over the last mile has been weakened.

Towards the end of the 1990s, Telefónica took two important decisions: first, to acquire the shares of the retail owners of its affiliates, and second, to structure their operations according to the different telecommunications segments. The initiative made it possible to better organize the main lines of business, but somehow it ran counter to the trends defined by the process of convergence.

Users' growing interest in relying on providers that can supply a comprehensive package of telecommunications services in some way raises doubts about the above-mentioned organizational model. This is translating into the search for more efficient forms of organization59 and into a systematic policy of incursion into fixed telephony. Thus, in the case of Panama, Telefónica Móvil has requested a concession for the provision of fixed telephony, and in Mexico the company has established a strategic alliance with Avantel and has also indicated that they are seeking the acquisition of a fixed operator.

One aspect that is taking on growing importance is the debate regarding the best means of treating the industry. Even though an analysis of the different markets into which telecommunications are divided continues to be useful, the way in which the business is structured, the multi-purpose nature of the networks as a result of convergence and the companies' own operation make it necessary to move on from isolated treatment of each segment towards comprehensive treatment of the industry. The formerly different networks are becoming integrated so as to build a single platform for the provision of fixed and mobile telephony, LD, data transmission and television services. Companies use all their assets to compete in the markets for each of the end services.

Furthermore, the companies that divided up just over five-years ago according to the traditional segments, fixed and mobile telephony and Internet, have begun a process of reintegration and are seeking new forms of internal structuring. Though this tendency seems essential to make the best use of new technologies, regulation and competition institutions must prevent anti-competitive practices.

59 Among them being the introduction of de facto organizational formulas in which the different companies converge, with the establishment of directorates of business areas, which are served by the network and planning boards of directors, etc.

Bibliography

Ansorena, C. (2007)s, Competencia y regulación en las telecomunicaciones: el caso de Nicaragua (LC/MEX/L.785), serie Estudios y Perspectivas No. 73, ECLAC/IDRC, July.

Argumedo, P. (2007), Competencia y regulación en las telecomunicaciones: El caso de El Salvador, serie Estudios y Perspectivas N° 73, ECLAC/IDRC, Mexico, February.

Estache, A., M. Manacorda and T. M. Valletti (2002), Telecommunications reform, access regulation, and Internet adoption in Latin America, Economía, vol. 2, N° 2, Spring 2002, pp. 153–217.

FUSADES (Fundación Salvadoreña para el Desarrollo Económico y Social) (1998), "Las Telecomunicaciones en El Salvador: un sector estratégico para el nuevo siglo", Boletín Económico y Social, No 153, August.

González, R. (2007), Competencia y regulación de las telecomunicaciones: el caso de Panamá, serie Estudios y Perspectivas N° 74, ECLAC/IDRC, Mexico, February.

International Telecommunication Union (2005). http://www.ine.int/ITU-D/icteye/reports.aspx#.

Mariscal, J. and E. Rivera (2005a), New trends in the Latin American telecommunications market: Telefonica & Telmex, Telecommunications Policy, N° 29, pp. 757–777.

Mariscal, J. and E. Rivera (2005b), Organización, industrial y competencia en las telecomunicaciones en América Latina: las estrategias empresariales, serie Desarrollo Productivo N° 169 (LC/L.2423-P), ECLAC, Santiago, Chile.

Mariscal, J. and E. Rivera (2007), Regulación y competencia en las telecomunicaciones mexicanas, serie Estudios y Perspectivas N° 83, ECLAC/IDRC, June, Mexico.

OECD (2007), Communications Outlook 2007. http://213.253.134.43/oecd/pdfs/browseit/9307021E.PDF

Ramírez, F. (2005), Interconexión de redes telefónicas en México: solución de disputas y búsqueda de consensos, Mexico, unpublished.

Rivera, E. (2006), Regulación y competencia de las telecomunicaciones en Centroamérica: un análisis comparativo, in C. Schatan and M. Avalos (Coords.), Condiciones y políticas de competencia. Economías pequeñas de Centroamérica y el Caribe, FCE y ECLAC, Mexico.

Tábora, M. R. (2007), Regulación y competencia en el sector de las telecomunicaciones en Honduras (LC/MEX/L.781), serie Estudios y Perspectivas No. 84, ECLAC/IDRC, México, June.

Urízar, C. (2007), Competencia y regulación de las telecomunicaciones: el caso de Guatemala, serie Estudios y Perspectivas N° 75, ECLAC/IDRC, Mexico. February.

Vargas, L. and G. Hernández (2007), Competencia y regulación en las telecomunicaciones: el caso de Costa Rica, ECLAC/IDRC, Mexico, forthcoming.

Waltsten, S. (2000), "Telecommunications privatization in developing countries: the real effects of exclusivity periods". http/www.isnie.org/ISNIE01/Papers01/waltsten.Pdf.

Chapter 4
Competition and Regulation in the Banking
Systems of Central America and Mexico:
A Comparative Study

Eugenio Rivera and Adolfo Rodríguez

Introduction

Competition in the financial sector, especially in the banking sphere, has been a very polemic topic. The adverse selection and moral risk that characterize this sector explain the high regulation of its economic activities. This feature is geared towards solving the important information problems faced by the economic actors, so as to protect savers' interests and ensure the stability of the financial sector as a whole. This objective, nevertheless, contradicts to some extent the aims of competition policy. On the one hand, excessive competition may create difficulties for the bank supervisor: it is easier to control a few large banks than having to do so with a large number of small ones. On the other hand, an excessive competitive dynamic may force banks to transfer a large part of their surplus to the consumer, limiting the banks' capitalization capacity and may lead some of them to undertake excessive risks to face competition. This explains to a certain extent the existence of some regulatory arrangements – such as the requirement of relatively high minimum capital – that promote bank concentration. But, at the same time, the absence of competition can weaken market discipline and allow for inefficient practices, while the excessive concentration can reduce the supervisor's relative power as he faces banks that are "too big to fail".

However, the recent evolution of financial markets tends to solve this contradiction. The development of banking technology has created, more than any other economic sector, a competitive worldwide space, so that local banks are facing, more than ever, a strong competition. This cannot be avoided by measures that promote local bank concentration; on the contrary, concentration can worsen some of the negative effects of excessive competition on banks' behaviour, particularly in the negligent handling of risk. Competition promotion in local markets, together with measures that improve the local bank competitiveness so as to close the gap with the international one, is the best complement to the prudential norms. These should be revised so as to remove aspects that promote concentration and lack of competition in this sector.

Understanding banks' competitive strategies and their performance requires an analysis of the institutional conditions in which financial institutions operate. Such conditions include: (i) the regulatory framework, which includes prudential regulations, monetary policy and taxation; (ii) the characteristics of the region's financial infrastructure – especially those concerning liquidity management and payment systems; (iii) international connectivity; and (iv) non-bank sources of contestability in the banking market, particularly public debt and stock markets. This is the context that determines banks' competitive strategies in the region.

In fact, in this context, the Central American banks have developed a set of strategies to ensure their continued existence in the setting of the industry's growing internationalization and to respond to the demands imposed by the application of the Basel standards. The following feature among those strategies: (i) a process of modernization and efficient organization of activities; (ii) development of offshore banks in parallel to local ones; (iii) an active policy of mergers and takeovers which has reduced the number of banks and increased concentration indicators of the five largest institutions; (iv) regionalization of activities and of the ownership of financial groups; and (v) growing involvement in financial activities formerly barred to the banks.

With this context in mind, the Section I analyses the background and salient features of the history of the banking system in the different countries of the region, as well as the main characteristics of the institutional framework governing regulation and promotion of competition. Section II consists of a study of the main determining factors for the banks' competitive strategies. Section III analyses the main strategies developed by bank operators. Section IV shows certain indicators regarding the banking system's performance from the point of view of both management efficiency and intensity of competition in the system. The study ends with the main conclusions drawn and some policy recommendations in Section V.

I. Organization of the Banking System and Institutional Framework for Regulating and Promoting Competition in Central America and Mexico

1. Historical Background of the Banking System in the Region

Although the bank's historic trajectories in the different nations are extremely varied, they show at least three similarities: the presence of a strong public bank at some point, the generalized tendency to privatization of such bank (with the exception of Costa Rica) and the presence of political or economic turmoil which have threatened the industry's financial solidity.

Recent trends in the banking sector reveal a modernization process and an improvement in its regulation, even if at a different pace in each country. There has been a considerable progress in the prudential regulation application and different steps have been taken in order to achieve a consolidated supervision of the whole financial sector. The authorities' concern for financial stability has limited their effort to promote more competition in these markets. This can be appreciated in the fact that regulatory bodies have not included competition as one of their most important tasks, as well as in the absence of a competition agency, or else in their judicial limitations to undertake these cases.

Over the past 15 years the banking system in Central America and Mexico has undergone major transformations. A brief look at the development of the banking system in Mexico reveals a traumatic track record. In the context of the international-debt crisis in the early 1980s, the Mexican government announced the nationalization of the banking system. Less than ten years later, in May 1990, the Mexican government, then headed by President Carlos Salinas de Gortari, announced the constitutional amendment that made privatization possible. The government's aim was to maximize public revenues. The privatization process was structured as a series of successive auctions in which the losers could participate in bidding processes for subsequent blocks of shares. The process was open not only to groups with banking experience, but also to groups linked to the stock exchange. No kind of assessed value of assets or minimum prices was made known, so the sale was carried out at an average of three and a half times the book value. The high sale prices were subsequently considered to be one of the reasons behind the collapse of most of the banks in the context of the 1994–1995 "Tequila crisis". Several causes were adduced to explain the above-mentioned crisis, among them the fragility of the Mexican banking system, stemming from the low quality of the banks' credit portfolio and the high level of leverage in the corporate sector. These circumstances followed from the rapid expansion of credit and the overvalued prices of assets and guarantees which took place during the years of economic expansion in Carlos Salinas' term of office and were associated with the weakness of prudential regulation. One particularly serious situation was the growth of related credits without adequate backing. Although less than eight years had elapsed between nationalization and privatization, this time it took four years for the privatized banking system to collapse. The tequila crisis brought in its wake high delinquency due to high interest rates, contraction of the supply of uncommitted funds, reduction in the level of financial intermediation and economic slowdown, which made it necessary for the state to intervene in the capitalization of banking institutions. In 1999, only 35% of the 20 banks in operation came from the privatizations; two of these – Bancrecer and Inverlat – were controlled by Institute for the Protection of Bank Savings (IPAB), another two had been taken over by foreign banks (servicios financieros integrados (SERFIN) by Santander Mexicano and Banco de Comercio (BANCOMER) by Banco Bilbao Vizcaya Argentaria (BBVA) ) and Banco Internacional (BITAL) was being capitalized by European banks. Thus, by 1999 only two of these institutions – BANAMEX and Banorte – had retained 100% Mexican stock ownership (see Avalos and Hernández, this volume).

As in Mexico's case, the commercial banks and savings and loan associations of El Salvador were nationalized in 1980.1 Ten years later, in 1990, the process was

1 Legislative Decree No. 158: "Law on the Nationalization of Credit Institutions and Savings and Loan Associations" dated March 7, 1980.

reversed and the government set in motion a process to privatize the sector. This evolution has taken place together with several legal changes, among which are the Organic Law of the Central Reserve Bank of El Salvador, passed in 1991, which included a prohibition on directly or indirectly financing the state (Article 74) and eliminated the authority of that institution to set the exchange rate and interest rates. These have been subject to market forces ever since. Probably the most important subsequent event in the development of the financial system took place on January 1, 2001, when the Law on Monetary Integration went into effect. The law introduced the use of the US dollar as legal tender and stipulated that all banks and financial operations be denominated in that currency. Naturally, the role of the Central Reserve Bank had to be redefined, particularly as regards monetary policy. In 2006, there were ten private banks and two state banks (Herrera, 2007).2

Nicaragua has not been free of vicissitudes. The nationalization of the banking system after the Sandinista revolution was followed by an initial period during which a state banking system was in operation (1980–1990); the next decade was characterized by a private banking system which gained a leading role in the country but underwent a severe crisis (2000–2002) paving the way for a process of concentration between 2003 and 2005 that had left the country with six banking institutions by 2006. Despite the ups and downs in the history of Nicaragua's banking system, or perhaps because of these, Nicaraguan financial capital has a strong presence in the region.

Undoubtedly the characteristic feature of the banking system in Costa Rica is a strong state sector with origins in the 1949 bank nationalization, which also provided for a monopoly on deposits based on the argument that the public's savings should serve the public interest and not private profit-making. Only one bank was not nationalized, Banco Lyon, which specialized in international operations to support exports. Nevertheless, protected by a series of reforms, the private financial sector started attracting funds even since the end of the 1960s. From 1980 onwards, the private banks started having a wider space. Further leeway was afforded to private banks by authorizing access to Central Bank credit programmes and it was allowed to attract savings from the public at progressively shorter terms.

Panama stands out for being an international financial district. The modern Panamanian banking system has its origins in Cabinet Decree 238 of 1970 which established the National Banking Commission, the first attempt at state regulation of banking operations in the country. The Law formalized the presence of international banks in Panama by authorizing both local and offshore operations, depending on the interests of each institution. Banks that engaged exclusively in offshore transactions, however, could make local interbank placements, thus creating a highly active interbank market that forms the basis of the system to this day. The Bank Law gave every

2 Public banking in El Salvador has not been devoid of problems, an example being Banco de Fomento Agropecuario (BFA), which was the scenario of one of the financial system's biggest frauds, in which BFA lost approximately 24.8 million colones (US$2.84 million). Another negative aspect affecting public banking has been a poor-quality loan portfolio. At the end of 2002, Banco Hipotecario registered a 9.3% default rate and BFA 28.6% (Herrera, 2007).

bank the freedom to establish its own borrowing and lending rates and the granting of credit – with a few minor exceptions – was not restricted by general ceilings or by specific portfolio allocation by sectors. This rate-setting freedom for each bank paved the way for the integration of the banking system with the rest of the world, and the London Interbank Offered Rate (LIBOR) is the base for establishing the cost of credits. The number of banks increased steadily between 1970 and 1983. Notwithstanding, since the latter year, some important banks such as First Chicago, Libra Bank and Bank of America, among other smaller ones, began to shift their operations to other financial centres, as their presence in Panama was founded on the accounting records of sovereign loans to Latin America; but when those loans became delinquent, it was more advantageous to register them in centres where those losses could serve as a tax shield. Even so, many foreign banks continued to carry out operations from Panama and to cater to the local market (Fernández, 2006).

In Guatemala's case, a number of problems hindered approval of the new regulatory framework for the banking system throughout the 1990s. It was not until 2002 that the Monetary Law, the Law on Free Negotiation of Foreign Currency, the Organic Law of the Bank of Guatemala, the Law on Banks and Financial Groups, the Financial Supervision Law and the Law on the Stock and Commodities Market were passed. The delay in implementing the reforms has been associated with the outbreak of the 1998 financial crisis,3 which pointed out deficiencies and poor financial and administrative management within the framework of legislation that proved obsolete in the face of the financial innovations that had taken place (Balsells, 2007).

2. Institutional Regulatory Framework and Promotion of Competition

The modernization of the banking system, its growing internationalization, bank disintermediation processes and the importance of competition in ensuring the industry's efficient performance raise a series of institutional problems of prime academic and public-policy interest. This evolution, together with the intensification of prudential regulation, has demanded the development of greater competence from regulatory agencies. Domestic banks have grown in size and in their international links, while international banks, because of their large size and its global way of operating, pose a challenge for regulators mostly operating on a domestic basis. Hence, international cooperation among regulators takes on vital importance.

3 The 1998 crisis was brought on by risky investments in agro-export companies linked to traditional products such as coffee and cardamom, which suffered marked price drops that year. In addition, important banks invested in securities affected by the "Asian crisis", including Russian government bonds that fell into a non-payment situation. International problems, together with the coffee crisis and the appearance of natural phenomena such as Hurricane Mitch, provoked a liquidity crisis in the banking system which rapidly raised the country-risk situation and led to the bankruptcy of a considerable number of financial firms (Balsells, 2007).

An intensely debated issue is the independence of the institutional regulatory framework from the government. At one end, the appointment of regulatory authorities by the Congress is seen as a guarantee for these authorities' autonomy, while at the other end this autonomy is seen as a potential source of a lack of coordination on economic policy and a kind of "total political irresponsibility". As we shall see, in most of the countries studied Executive Branch authorities intervene in the appointment and dismissal of regulatory authorities, and adequate mechanisms for safeguarding the independence of regulators are non-existent. The problem is compounded when the sector being regulated has a say in the designation of authorities and in approving the principal measures, an example being the Monetary Board in Guatemala.

Another institutional issue raised with growing insistence is competition. As we will see, recent developments have given rise to a process of mergers and takeovers which has produced a concentration of ownership and a regionalization of the banks. This process has taken place in parallel with a certain reduction of intermediation margins and commissions which nevertheless remain at relatively high levels, indicating a low degree of competition in the sector. This justifies discussion on the need to promote the role of competition agencies. Until late 2004, four countries in the region lacked a competition supervisor, and in the other three the effective powers of such agencies in the banking sector were highly limited or non-existent. Bank supervision agencies fail to appreciate the importance of competition and the involvement of competition agencies; in point of fact, bank supervisors still tend to consider banking as a special industry which on account of its importance to the overall performance of the economy should be protected from the "excesses" of competition regulation. The competition laws recently passed in El Salvador and Honduras acknowledge the importance of competition in the financial sector and establish that the competition agency should become involved in that sector just as in any other economic activity,4 for which purpose the means of coordination with financial supervision authorities should be appraised.

In Mexico the chief supervisor is the Ministry of Finance and Public Credit (SHCP), which authorizes the operation of financial groups, banks, insurance companies, pension fund administrators (Afores) and stockbrokerage firms and follows up on these entities through four commissions in charge of regulation and prudential supervision: the National Banking and Securities Commission, the National Insurance and Sureties Commission, the National Commission on the Retirement Savings System and the National Commission for the Protection and Defense of Users of Financial Services. Analysts maintain that the Mexican financial system is dominated by the major financial groups which own banks, stockbrokerage firms, insurance companies and Afores, complicating regulation due to the overlapping of activities engaged in by each entity, the existence of cross-subsidies not properly registered in accounting ledgers and coordination difficulties among the regulatory agencies.

4 In Mexico's case the powers of the Federal Commission on Competition (CFC) apply to the financial sector in the same way as to any other sector. Nonetheless, the CFC's actions have focused mainly on reviewing mergers and takeovers (Avalos and Hernández, this volume).

In contrast to other countries of the region, since 1993, Mexico has had a competition agency, the Federal Commission on Competition, which is in charge of supervising mergers in the industry. Despite the modernization of the regulatory agencies, one problem that has affected regulation is the lack of modernization of the Judiciary.5

The Salvadoran financial system is governed by four institutions: the Superintendence for the Financial System (SSF), in charge of supervising and regulating the activities of commercial banks, insurance companies, bonded warehouses, bureaux de change and other concerns related to financial activities; the Securities Superintendence, in charge of supervising securities intermediaries, central securities depositories and the stock exchange; the Pensions Superintendence, responsible for supervising and regulating Pension Fund Administrators; and the Central Reserve Bank, which is in charge of promoting and maintaining monetary stability and developing an efficient, competitive financial system. The SSF forms part of the Central Reserve Bank of El Salvador and possesses autonomy in administrative and budgetary matters and in exercising the powers conferred on it by the Law. The SSF's management is headed by a Directive Board and the Superintendent, the latter nominated by the council of ministers and approved by the President of the Republic. The Bank Law establishes that the SSF is in charge of authorizing bank mergers and such mergers must be carried out in accordance with the rules established by the Code of Commerce. Throughout the period studied, El Salvador has lacked a competition agency. The Law on Competition was passed in December, 2004, and the agency began operating in January, 2006 (Herrera, 2007).

In Guatemala, the law stipulates that supervision of the entire financial system, including banks, is incumbent upon the Superintendence of Banks, a Central Bank agency that acts under the general directorate of the Monetary Board. The latter is responsible for submitting a list of three candidates from which the President of the Republic appoints the Superintendent, whose resolutions in regard to oversight and inspection may be appealed before the Monetary Board. The Monetary Board grants or denies authorization for the incorporation of banks and foreign bank branches, and for bank mergers. The Monetary Board's powers and composition translate into a low level of autonomy for the supervisory body both from the government and from the private sector itself (Balsells, 2007).6 Guatemala had neither competition law nor agency yet in 2006.

Nicaragua also has a system with a sole regulatory agency. The Superintendence's higher organs are a directive board, a superintendent and a vice-superintendent.

5 According to a study, 60% of mercantile executory processes do not go beyond the first stage of the proceedings, i.e. summons and attachment. Moreover, of the remaining 40% of cases, only half reach a final decision, in other words, of every 100 complaints admitted, only 20 reach a verdict (Avalos and Hernández, this volume).

6 The board is chaired by an official appointed by the President of the Republic for a four-year term who is also chairman of the Bank of Guatemala (BANGUAT), and its members include the Ministers of Public Finance, Economy and Agriculture, a member elected by Congress, a member elected by trade, industry and agriculture business associations, a member elected by the National Banking Association and a member elected by the Higher Council of the University of San Carlos of Guatemala.

It authorizes, regulates and supervises the operations of the following entities: banks, finance companies, insurance companies, the stock exchange, seats on the stock exchange, general bonded warehouses, leasing companies and Tourism Investment Capital Funds (FONCITURs). Its sphere of duties makes no reference to considerations of concentration, dominant position or abuse of market power and their implications for competition; the criteria for approval of mergers or takeovers contain only aspects that denote concern over the system's solvency. In accordance with the current provisions of the Bank Law, by virtue of bank secrecy in Nicaragua that sector remains outside the jurisdiction of any future Competition Agency (Ansorena, 2007).

In Honduras, the Law of the National Banking and Insurance Commission (CNBS) was passed, becoming a de-concentrated agency of the Presidency of the Republic, assigned to the Central Bank of Honduras (BCH), but with absolute technical, administrative and budgetary independence. Its mission is to supervise and regulate but the law does not assign any competition supervision function to it (Tábora, 2007).

II. Determining Factors in the Banks' Competitive Strategy

In the struggle to secure market share and an adequate rate of return, the banks try out different strategies that are not only determined by their objectives and goals, but also to a large extent constitute reactive behaviour in the face of the external conditions in which competition operates. The determining factors of these external conditions could be classified into four groups, mainly of institutional origin and relatively independent of the banks: (i) the regulatory framework; (ii) the infrastructure of the financial sector; (iii) the connectivity of the financial system; and (iv) the internal sources of contestability. This section discusses these factors, their effects on banks' competitive strategies and their results.

1. The Regulatory Framework

The banks face a regulatory framework that operates in different fields and has consequences on their performance and on the way competition operates in each country and in the region as a whole. One could hardly explain the course of the banking system's development in the different countries without taking that framework and its vicissitudes into account.

1.1. Discriminatory Barriers to Entry

The regulatory provisions with the most obvious effects on competition are those concerning entry barriers. There are barriers that hinder the entry of new banks due to existing banks' "natural" advantages related, for example, to economies of scale or relationship lending. Other barriers stemming from the regulatory framework indistinctly affect all banks seeking to establish themselves in a particular market, such as minimum capital requirements, for example. Finally, there are barriers that do not equally affect all the banks seeking to establish themselves in the market, but that discriminate on two grounds: origin of capital (foreign or domestic) and ownership of capital (public or private). These are known as "discriminatory entry barriers".

At present there are no absolute barriers to the entry of particular banks, although they have existed in the past. The most extreme case is Costa Rica which, as we saw in the preceding section, nationalized all the existing banks and implemented a state monopoly on deposits from the public in 1949; this was followed by a lengthy period during which for all intents and purposes only state-owned banks existed. This monopoly began to be loosened in the mid-1980s when private banks were authorized to take deposits from the public for terms longer than six months and was eliminated in 1995.7 The extensive network of branches developed by the public banks and the existence of an unlimited state guarantee on their deposits has made Costa Rica the only country in the region in which public banking clearly predominates.

The opposite extreme is exemplified by Panama, which imposes no restrictions on the banks to establish in its territory. The 1970 Bank Law authorized operations, from offices established in Panama, carried out, completed, and having effects in other countries, as well as the establishment of bank agencies in Panama. This set of provisions, complemented by a deregulation process that included the liberalization of interest rates and later the elimination of reserve requirements, gave rise to the development of the "financial district" which became an important offshore destination for the region's banks and for some of the world's largest banks.

Between those two extremes there are various degrees of discriminatory entry barriers in the region. With regard to ownership of capital, the barriers that discriminate private banks as against public ones concern the obligation of private institutions to manage their funds through public banks, income tax exemptions for public banks and the existence of a state guarantee on deposits in public banks, which in countries such as Costa Rica and Guatemala play an important role for the public. By contrast, the law in some countries stipulates a number of specific uses for public bank profits and the legislation imposes a series of restrictions on them regarding employment and administrative hiring.

As to the origin of capital, no country in the region has restrictions on the establishment of foreign banks, although in some of them there are certain limitations on the establishment of offices or branches: only banks subject to local regulation and supervision can operate in the national territory. This restriction eliminates the

7 In exchange for eliminating the state monopoly on deposits, Costa Rican legislators provided that private banks taking sight deposits (current accounts) should place 17% of their acceptances in public banks so that the latter funnelled them into development credit.

possibility of banks established abroad to openly offer their services in the country, although they do so surreptitiously, as will be seen on analysing the degree of connectivity in the region and the role of offshore banking.

1.2. Prudential Regulation

Other than El Salvador and Panama, all the countries of the region have undergone bank crises (Mexico, Nicaragua) or periods of serious bank stress that have included bankruptcies and interventions (Costa Rica, Guatemala, Honduras). This has paved the way for considerable efforts to update prudential regulations and supervisory capacity. The countries of the region have gradually been adopting the prudential norms issued by the Basel Committee. These norms seek, among other aims: (i) to define minimum conditions for the establishment of a bank as regards its capital and the honour of its shareholders; (ii) to adjust each bank's capital base to its risks as a guarantee for absorbing unforeseen potential losses; (iii) to establish mechanisms to assess asset quality, risk and coverage by creating reserve funds; (iv) to set limits on portfolio concentration and restrictions on extending credit to persons linked to the bank; (v) to foster procedures for assessment and comprehensive management of other sources of risk, such as country, infrastructure and market risks; and (vi) to regulate the use of internal controls and client assessment systems to prevent money laundering.

Among the Basel standards that most influence competition in the banking sector are those on capital requirements, which have gradually been increasing since the 1990s. On the one hand, the minimum capital required to establish a bank was around US$8 million on average in 2006. This amount is far below the minimum capital required in OECD countries and even in other Latin American countries (US$25 million in Chile). And on the other, the compulsory capital adequacy ratio (CAR) is generally higher than that recommended by Basel, which is explained by the greater opaqueness in information and the greater risk associated with these markets. The compulsory ratio fluctuated in 2004 between 7.5% in Guatemala and 13% in Panama (see Table 4.3). This ease, however, can be partly explained by the liberality with which the credit portfolio tends to be classified, the quality of which would require provisions that would lower the effective CAR. In any event the gradual increase in capital requirements has played an active role in furthering mergers and takeovers since the beginning of the present decade.

Other Basel standards affecting competition are those concerning credit, particularly limits on portfolio concentration and credit to related persons. The main problems facing the banks in the region are associated with credit risk, since information on portfolio quality – debtors' capacity to pay and loan security – is frequently lacking, until the economic situation deteriorates and a high proportion of the portfolio becomes delinquent. The excessive concentration of the portfolio in a few sectors of the economy, as well as with loans to related persons, is the origin of most banking crises and interventions in the region.

Table 4.1 Central America and Mexico: capital requirements and reserve coefficients, 2005 (%) (Rodlauer and Schipke, 2005)

 

Requirements

 

For Capital

Reserve coefficient

 

 

In local currency

In foreign currency

Costa Rica

10.0

10.0

10.0

El Salvador

11.5

20–25

n/aa

Guatemala

10.0

14.6

1.6

Honduras

10.0

12.0

12.0

Mexico

8.0

Nicaragua

10.0

16.3

16.3

Panama

8.0

n/aa

a El Salvador and Panama are totally "dollarized".

1.3. Monetary and Exchange Policy

Monetary policy in most of the economies of the region has been marked by central bank losses and monetary policy. Aside from Panama, officially dollarized since 1904, all the countries had fixed dollar exchange rates which were abandoned during the course of the 1980s as a result of inflationary processes, armed conflicts or financial turbulence: Nicaragua abandoned the fixed exchange rate in 1979, Costa Rica in 1981, El Salvador in 1983, Guatemala in 1984, Honduras in 1990 and Mexico in 1994. Abandonment of the fixed exchange rate often took place after an accumulation of imbalances which provoked a free fall, after which various exchange systems were adopted: following a relatively short period, Costa Rica, Honduras, Guatemala and El Salvador adopted a crawling peg, generally adjusted in line with expected inflation; later, El Salvador adopted dollarization as of 2001, and Guatemala an independently floating system since 2003. Mexico, for its part, has allowed its currency to float freely, although there have been episodes when the Central Bank has intervened forcefully in the foreign exchange market according to pre-established rules. In all cases, leeway in monetary policy is tightly restricted by exchange goals.

Moreover, the foreign exchange crises experienced by the Central American countries during the 1980s obliged the Central Banks to intervene in order to avoid the collapse of the commercial banks. This caused excess liquidity which had to be siphoned off through the placement of securities issued by the Central Bank itself, the service of which entails quasi-fiscal losses of several percentage points of gross domestic product (GDP) and creates chronic excess liquidity. The need to continually siphon off this excess liquidity and keep to foreign exchange goals has compromised the autonomy of monetary policy. This is aggravated by deficiencies in the payments infrastructure and in public-debt management, which has hindered the organization of interbank markets that enable the monetary authority to intervene through indirect instruments. Hence, the main monetary-policy instrument is cash reserves: aside from Panama, which lacks bank reserves, reserve coefficients fluctuate between 10% (Costa Rica) and 20–25% (El Salvador) (see Table 4.1). This way of implementing monetary policy has a high cost for banks, which transfer them to their clients through high intermediation margins (as cash reserves are not remunerated). This places local banks at a considerable disadvantage compared with their international competitors and motivates them to shift their operations elsewhere.

In Mexico the elimination of reserve requirements on new deposits by Banco de México coincided with a considerable net inflow of foreign capital just at the time of the re-privatization of the banks in 1991. This led to an increase in liquidity in the banking system and hence by a marked expansion of credit, also made possible by the weakness of the banks' supervision and risk management during the decade when they were nationalized. This expansion contributed to turning the 1994 exchange crisis into a serious solvency crisis.

Another consequence of the exchange and monetary environment is the growing informal dollarization experienced by these economies. The Central American Monetary Council (Consejo Monetarios Centroamericano, (CMCA) ) has estimated that in 2001, around US$1.401 billion was in circulation in the region, representing 44% of the monetary emission; in 2004 that dollar availability had risen to US$2.297 billion, 72% of monetary emission (CMCA, 2004).8 Between 1996 and 2003 dollar deposits in commercial banks – excluding Mexico and Panama – increased from 20.2% to 31.1% of overall deposits, and dollar credits from 15.7% to 28% of overall credit. This trend heightens the banking system's liquidity risk, insofar as there are less mechanisms for protection from the mismatching of maturities between dollar deposits and credits, as well as the credit risk, since a large proportion of debtors do not have dollar incomes and in the event of a devaluation many of them would become insolvent, which would compromise the health of the creditor banks' assets. This outcome of monetary and exchange policy, as well as the preceding one, has implications for competition in the region, particularly if not all the countries have prudential rules that make the price system reflect the risks inherent to informal dollarization.

2. Financial Infrastructure

The shortcomings suffered by practically all the countries of the region, except for Mexico, in the development of the financial infrastructure, have direct consequences on competition between local banks and their foreign competitors, as well as between large and smaller banks.9 In addition, they help to explain some of the

8 These estimates give an idea of the intensity of the dollarization process, although the amounts exclude Mexico and Panama and include the Dominican Republic.

9 In 1994, Banco de México undertook a comprehensive reform of payment systems which reduced the character risk and attained a high degree of security and operative reliability in settlement of payments.

competitive strategies that will be analysed in the next section. Two components of the infrastructure are especially relevant from the point of view of bank competition: interbank markets and interregional payment transfer systems.

2.1. Interbank Markets

Commercial banks manage their liquidity mainly in interbank markets by trading and borrowing public-debt securities. The existence of efficient interbank markets from the point of view of price-setting, but in particular of the timeliness with which operations are agreed and settled, is essential to them for optimal liquidity management. None of the countries of the region, with the exception of Mexico, have such markets. Two circumstances account for this.

First of all, central government-debt management has not provided, among its objectives, for the development of the public-debt market: non-standardized governments bonds are issued, there is no constant presence on the market, placements are made through mechanisms lacking in transparency – such as retail windows or unilateral negotiations with public or private institutions – and the organization of secondary and bond-execution markets has been left in private hands. The result is a practically non-negotiable stock of public debt which cannot be used, as in more developed countries, to manage banks' liquidity and risk and for the execution of monetary policy.10

Second, local payment systems stand as an additional hurdle to the organization and operation of interbank markets. Aside from Mexico, Guatemala and Costa Rica, no country in the region has payment systems that enable transfers between banks to be made in real time, which is a basic infrastructure for the development of an interbank market. Furthermore, in Guatemala and Costa Rica no measures have been taken to allow the use of real-time gross settlement (RTGS) infrastructure for an interbanking market to work. Therefore, the payment system lacks the mechanisms required to handle liquidity and is not used by banks for high value payments.11 This means that, except for Mexico, in order to transfer money, banks have to resort to two types of channels. On the one hand, clearing houses limit payment timeliness (because cheques are generally cleared in t + 1) and increase the systemic risk, as clearing houses lack risk management mechanisms for large payments; the funds a bank expects to receive are generally tied up in other operations, so that one bank's default can cause a chain of defaults. On the other hand, the channel used by banks to make their payments is through the Central Bank. Here a commercial bank can request the Central Bank, by telephone or fax, to debit its reserve account in order to credit another bank's account for the amount in question.

10 See World Bank and IMF (2000) and Litan et al. (2003), on the importance of an adequate management of the public debt the limitations faced by emerging markets.

11 See IMF (2005 and 2006), on the shortcoming of the regional payments systems, especially for its judicial framework.

This channel is too inefficient to allow the development of an active interbank market.

Beyond the consequences that the absence of public-debt interbank markets has for the state, since the lack of market liquidity increases the cost of public indebtedness and hinders the use of modern monetary-policy instruments, it distorts competition and limits local and small-bank competitiveness. In fact, the lack of mechanisms to handle liquidity forces commercial banks to maintain excessive liquidity at great financial cost in order to meet the vicissitudes of their daily needs and contract lines of credit with foreign banks to address their diverse contingencies. This increases intermediation costs and creates a dependence on international markets with regard to liquidity management and introduces competition distortions in the different banking systems. On the one hand, it gives foreign banks an advantage over local banks, since liquidity management is less costly for the former than for the latter. On the other, the large banks have an advantage over the smaller ones, since many of these cannot obtain adequate lines of credit from foreign banks or at any rate can only do so at a much higher cost than their competitors.

2.2. Intraregional Transfers

The other important infrastructure component for bank competition is the one governing intraregional transfers. Despite the region's high degree of integration in its productive structure, migratory flows and financial system, there are no efficient mechanisms for making payments between one country and another. Intraregional payments have to be carried out through a costly chain of correspondent banks that increase transaction costs and raise portfolio management costs for both local clients and the banks themselves.

How does the system work at present? A bank customer in Costa Rica wishes to transfer a sum of money to a customer of another bank in El Salvador – a typical case involving imports of Salvadoran goods, but which also applies to a remittance from a Salvadoran immigrant. The Costa Rican bank holds an account in at least one correspondent bank in Miami or New York, as does the Salvadoran bank. To make a transfer from the Costa Rican bank to the Salvadoran bank, the Costa Rican bank requests its correspondent bank in New York or Miami, through a swift message, to debit its account and transfer the sum in question to the correspondent bank of the Salvadoran bank, which, once notified, will credit the amount to the receiving customer's account. Commissions have to be paid which in the chain as a whole can amount to more than 5% for the client. Obviously the accounts in the correspondent banks have to have funds, and those resources are generally managed by the correspondent bank. This transfer mechanism makes all regional transfers more costly and imposes time restrictions on the operations: although local banks close at 6 p.m., interregional transfers must be made before 1 p.m. if they are to be executed on the same day, otherwise the funds will not be credited to the receiving account until the following day.

The regionalization of banking groups has made it possible to overcome such limitations, for they have developed compensation systems between their various banks which enable a customer of one of the group's banks in Costa Rica to make transfers to a customer of another of the group's banks in El Salvador, with more flexible hours and relatively lower rates – or in any case, rates that are not comparable to the chain of correspondent banks but are left entirely to the regional banking group. This regional transfer mechanism through compensation between the group's banks places regional groups in a strong position to attract clients engaged in intraregional business, which is part of the motivation behind the regionalization process that has been taking place in recent years. There are less efficient mechanisms than those available with present technology. If the countries were interested in greater regional integration and competitiveness of their banking systems, they would have to update such technologies.12

3. Degree of Connectivity

The evolution of communications and banks' technological development, which have paved the way for e-banking, has radically transformed the shape of financial markets and the different groups' strategies. Five years ago, in its report on consolidation in the financial sector, the Group of Ten stated that: "The continued evolution of electronic finance could expand greatly, or even eliminate, existing geographical limits and lower entry barriers, thereby altering the potential effects of consolidation" (Group of Ten, 2001, p.10). Traditional concentration indicators no longer reflect the degree of competition to which local banks are subjected; after having been protected by laws and different barriers to entry, these banks are rather suddenly facing an intensive competitive dynamics.

Transaction costs limiting local banks' exposure to external competition continue to exist, particularly in certain market segments, but a growing segment of local banks' activities is subject to strong competition from foreign banks and in general from foreign intermediaries.

On the one hand, local investors have the choice of opening accounts abroad and having their capital managed there through stockbrokerage firms in Miami or New York. But the majority of investors who invest in foreign markets do so through the local banks with stock market branches abroad, with which they can share management fees. This loyalty to local banks is explained by transaction costs and the individualized attention provided by local brokers. In fact, to operate directly through a foreign stockbrokerage firm, there are relatively high minimum sums required for opening accounts, which small-scale investors have difficulties to meet. Furthermore, these agencies usually give individualized attention to large investors only and, therefore, the medium-size investors and even some of the

12 The IMF (2006, p. 3) has repeatedly pointed out that the "efficiency gains that would be reached through the adoption of regionally integrated payment frameworks and stock market settlement".

large ones prefer to rely on the local banks. Hence, the competition from foreign intermediaries is restricted by these barriers and is an obstacle for the direct relationship with local investors.

On the other hand, sufficiently large local companies have the option of issuing American Depositary Receipt (ADR)s on Euro markets, as has been done by some of the region's breweries and cement industries, among others. But that particular segment includes a very small number of companies, although the amount of their financing needs is very high. Those companies, as well as a segment of smaller and medium-sized companies, receive offers of credit from foreign banks, which visit them regularly through their local offices when this is permitted, or through salesmen who make periodic visits, just as foreign brokers visit investors. The cost of dollar credits depends partly on the country's risk, partly on the company's soundness and partly on how far its links to the foreign bank date back, but in general it tends to be lower than the cost of credits offered by local banks.

Recent research shows the perception on the degree of openness and development of the financial systems of various Latin American countries, among them those of Central America, Mexico and Panama. Table 4.2 shows this perception for a set of indicators on the degree of exposure to competition and international markets.

As we can see, the predominant view is that it is very easy for investors to invest in both foreign and local markets, but much more difficult for companies to obtain financing in foreign and local markets. This confirms that in order to secure the financing they need, local companies continue to be relatively dependent on

Table 4.2 Selected countries: perception on openness (Arnoldo Camacho, Finance Alternatives in Latin America. Recent Developments and Future Outlook, Sumaq Summit, 4 de mayo de 2004)

 

Investment
Access in
foreign marketsa

Access of
foreigners to
local marketb

Access to
financings in
foreign marketsc

Access to
financing in
local marketsd

Entry to local
banking
industrye

Brazil

4.7

6.3

4.8

5.5

5.3

Colombia

5.3

3.1

3.1

3.6

4.4

Costa Rica

6.5

6.6

2.6

3.8

4.7

El Salvador

6.3

6.0

2.8

3.7

3.8

Guatemala

5.0

5.7

2.3

2.3

4.4

Honduras

6.0

6.2

2.0

2.8

3.8

Mexico

6.1

6.4

4.9

4.8

4.5

Nicaragua

6.1

5.9

2.5

3.2

4.3

Panama

6.0

5.9

3.2

4.1

5.2

a Local citizens who wish to invest on stocks and bonus and open banking accounts in other country (1 = forbidden to do it, 7 = free to do it).

b Foreigner investors (1 = forbidden to invest on stocks and bonus in the country, 7 = free to do it).

c Local enterprises may get into debt in bonus foreign markets (1 = practically impossible, 7 possible enough for a good company).

d Capture money emitting actions in the local market (1 = practically impossible, 7 = possible enough for a good company).

e The entry of new banks in the local banking industry (1 = difficult enough and rarely allowed, 7 = easy and subject to reasonable regulations).

domestic banks, a perception reinforced by the view that it is still relatively difficult for new banks to enter the market.

4. Internal Sources of Contestability

Apart from the competitive pressure arising from the globalization of the financial system and exerted by foreign banks and stockbrokerage firms, banks face various local sources of contestability. The most important in the region, which operate especially in Mexico and Costa Rica, are public debt and the stock market, the latter not so much in the financing of enterprises as in deposit-taking from the public through mutual funds, which in turn invest mainly in public debt.

4.1. Public Debt

Public debt is a source of competition for banks in attracting savings from the public. The returns on public debt are generally higher than those offered by the banks' term deposits, and the public in all the countries of the region have been able to purchase public-debt securities at the Finance Ministry, Central Bank and commercial bank windows. This has been an investment alternative for medium and large investors who invest part of their portfolio in local public debt purchased on the spot market (auctions and windows) or on the over-the-counter market through banks and stockbrokerage firms. This involves a market segment for which the banks are unable to compete, for in order to capture it they should expect to see their funding costs increase appreciably and in any case they already have the most captive market for sight deposits (current accounts) and term deposits from small investors who have no access to public debt. Indeed, public debt has not been an investment option for the latter due to the lack of a distribution network for such bonds, which is explained by the cost for the state of issuing physical certificates. The struggle to attract small investors and sight deposits has been opened by the development of jointly administered instruments, particularly mutual funds.

4.2. The Stock Market

The stock market offers two sources of contestability for the banks: through attraction of savings and through company financing. However, they have been unimportant in most of the countries of the region except Mexico and Costa Rica.

With regard to company financing, in no country of the region, not even Mexico, has the local stock market seriously competed with bank credit. In Mexico the capitalization of the stock market has been falling since 1994, and corporate-debt placements represent a low percentage of GDP. In Costa Rica the relative importance of the stock market and corporate acceptance is even lower. In both countries the stock market basically consists of public debt and bank securities. In Panama stock market activity is virtually non-existent and the high rate of stock market capitalization is accounted for purely tax reasons: the stock issued by companies is a mechanism for receiving bank credit and benefiting from more favourable tax treatment.13

As regards deposit-taking, the stock market has played a certain role in the development of the banking system in Mexico and Costa Rica, although that role has been uneven. As we have mentioned, in 1949, Costa Rica established a state monopoly on deposits from the public which for many years prevented financial intermediation by private concerns. This situation, coupled with the existence of a high level of public sector internal debt, favoured the emergence of a stock market which enabled the state to meet its financing needs and allowed private financial capital to compete with the public banks in the colón market (the colón is Costa Rica's currency). Costa Rica established a stock market long before the rest of the Central American countries and promoted two instruments which were to play a key role in the development of the banking system. First, joint investment portfolios administered by stock exchange seats, the design of which was copied from Mexico. Investment in public debt was done through these portfolios and it yielded higher returns than those offered by the public banks. This was a bank instrument, for it formed part of the stock market seat's balance and guaranteed previously agreed returns and ready cash. Second, it permitted the borrowing of public-debt securities at relatively short terms, thus constituting a means for investors to invest at very short terms with relatively high returns. These two instruments and foreign exchange activities as well as the support from the Agency of International Development (AID) led to the development, during the 1970s and 1980s, of private financial capital. Credit institutions and subsequently banks were established, which took deposits from the public by issuing negotiable investment certificates on the stock market, an instrument developed to replace term deposits, which were prohibited to them until the 1990s. Even at present, the banks used both joint portfolios and the borrowing of securities to manage their liquidity in the absence of an interbank market.

Thus, in Costa Rica the stock market constituted a means for the development of private financial capital and private banks in the face of the public banks' monopoly on deposits. This made it possible for the market's infrastructure, regulation and supervision to make more headway than in the rest of Central America and, at a certain point, to become a source of competition for the banking system, especially with the development of mutual funds. This instrument was initially developed by non-bank stock market seats, and its rapid development, particularly after the legislation passed in 1998, gradually obliged all the financial groups to create mutual funds and propose them to their customers. This instrument replaced the former joint portfolios – which were prohibited – and has been a source of competition in the

13 Tax treatment for companies listed on the stock exchange and their dividends is highly beneficial. Hence a considerable portion of bank credit has taken the form of purchase of share issues which the banks themselves structure and which never pass through the secondary market.

Costa Rican financial system, and its growth over the past five years, as such, reached almost 40% of the banks' liabilities. Some of them, though, in the absence of an interbank market use these as an instrument to handle their own liquidity.

In Mexico the importance of the stock market in the financial system increased following the 1982 nationalization of the banks. As a result of the latter, a growing percentage of bank deposits was earmarked for financing the fiscal deficit, which displaced the private sector from bank credit, and restrictions were imposed on lending and borrowing rates and other quantitative and qualitative limits on credit. Companies' need for funds encouraged the creation of a parallel banking system. In this regard, stockbrokerage firms were the main institutions competing with the nationalized banks, both in deposit-taking and in channelling savings. Their development at that time has allowed them to position themselves with a series of jointly administered instruments which are invested mainly in public debt and to a lesser extent in bank instruments. During the second half of the 1990s and in more recent years, the legal framework for mutual funds has been improved and that instrument has been gaining ground over the banks in deposit-taking.

In the rest of Central America the development of the stock market has been incipient, due largely to the fact that the financial system is dominated by the banks. In El Salvador there is a type of joint portfolios, similar to Mexican and Costa Rican portfolios, which enables stockbrokerage firms to attract sight deposits which they invest in public debt, and in Honduras the stockbrokerage firms issued securities backed by public debt, in both cases earning from the spread. In Panama and Guatemala mutual funds exist, in the latter country with deficient regulation, but in neither of the two countries have they succeeded in attracting significant amounts.

III. Competitive Strategies of the Banks in the Region

A series of common actions by the banks could be taken to mean strategies aimed at strengthening their competitive situation. The first strategy was the reorganization of, and modernization of, the banks to be able to meet the Basel standards and the increasing exposure to competition from foreign banks and stockbrokerage firms. The second strategy has been an active policy of mergers and takeovers as a means to expand operations, attract more clients, enter new fields of operation and eliminate competitors. The outcome of this strategy has been a significant reduction in the number of operators and an increase in the five main operators' share of total assets in each country. The third strategy has been regionalization which has taken place together with the mergers and acquisitions. With few exceptions, the main operators in each country are setting up operations in the other countries, thus accompanying the regionalization of Central American industry and becoming regional operators. The fourth strategy has been the establishment of offshore operations which makes it possible to operate at lower costs for clients and bank stock holders. The fifth strategy has been the combination of specialization and diversification of the products and services offered by the banks.

1. Efficient Organization and Modernization of the Banks

There has been a generalized improvement in the banks' operative efficiency. However, the intensity of this improvement varies substantially. El Salvador during the 1990s is a case in point: as we can see in Table 4.3, in 1997, the administrative costs/total assets ratio stood at an average of 3.42%, far below the other countries; since then and until 2005 that indicator continued to improve until it reached 2.81%, which was surpassed only by Panama. This country had an administrative costs/total assets ratio of 1.10%, a long way from the other countries being studied, in spite of that indicator's deterioration since 2000, when it reached 0.78%. Efficiency levels in Panama are associated with the existence of the financial district, where large part of the operations are barely registered and do not involve costs as high as local operations.

Nicaragua's experience contrasts sharply with that of El Salvador. These two countries' banks are the most active as regional operators, and they would therefore be expected to share similar efficiency levels. This, however, is not the case: Nicaragua features among the countries with the most inefficient banking systems, far below El Salvador, and without registering any significant improvement during the period from 1997 to 2005.

Mexico's case is paradoxical. The majority of the country's largest banks were taken over several years ago by some of the main bank operators at international level. It was to be expected that the high efficiency levels that characterize those operators in their countries of origin and in the other countries where they have operations would be transmitted to their activities in Mexico. This, however, has not been the case. As we see in Table 4.3, operative efficiency measured by the administrative costs/total assets ratio reaches 4.7% in Mexico, just below Costa Rica and Honduras. This situation can be explained by lax regulations in this matter and the existence of a strong trade union movement which has been successful in defending its achievements. The high fluctuation shown by the indicator in this case is also noteworthy.

Table 4.3 Central America and Mexico: operative efficiency of the banks, 1997–2005 (administrative costs/total assets average) (Central American Monetary Council at www.secmca.org/Estadisticas_Indicadores_Ban.htm and Mexico's National Banking and Securities Commission, at www.cnbv.gob.mx)

Country

1997

1998

1999

2000

2001

2002

2003

2004

2005

Costa Rica

6.08

6.06

6.04

5.85

6.01

5.87

5.76

5.42

4.96

El Salvador

3.42

3.42

3.46

3.53

3.04

3.06

2.98

2.83

2.81

Guatemala

5.84

5.92

5.82

5.71

5.26

5.32

4.67

4.66

4.28

Honduras

6.34

6.22

6.26

6.43

6.23

5.99

5.62

5.30

5.13

Nicaragua

5.03

5.67

5.05

4.20

4.18

4.27

4.26

4.74

4.64

Panama

n/a

n/a

0.80

0.78

0.82

0.98

1.06

1.00

1.10

Mexico

n/a

5.56

6.00

6.35

3.79

5.37

5.01

5.08

4.70

Note: Figures are for December each year from 1997 to 2004. For 2005, the figures are for June, except for Panama, which are for April.

In Costa Rica, the low operative efficiency evidenced by Table 4.3 is associated with the public nature of its main banks, which are subject to regulations inherent to the public administration with regard to procurement and personnel management, not to mention that they often have to produce results reflecting social rather than profit-making criteria, such as the extent of the branch network. Finally, the case of Honduras, last in the ranking, reflects the country's general backward state and the lower level of development of its banking system. In this context it is interesting to look at Table 4.4, which contrasts various efficiency indicators for Honduran-owned banks with foreign banks operating in the country. The table clearly shows that these have a better performance than the domestic banks in all the indicators.

The table above shows that the average CAR is approximately 4% points higher in the foreign banks than in the domestic banks; the bank with the highest CAR (38.65) is a foreign-owned bank, whereas the one with the lowest CAR (10.36) is a domestic bank. The foreign banks, on their part, generate greater average financial returns (FR) and in turn register a lower cost of liabilities (CL), and consequently their financial intermediation mark-up (9.69%) is 1.5 times higher than in the domestic banks (6.12%). This is related to two facts: higher administrative costs, but higher profitability of foreign banks. In fact, the relationship between administrative costs and average productive assets of foreign banks is 13% higher than the domestic banks, probably because of staff expenses associated with the hiring of international personnel. The returns on productive assets (RPA) of foreign banks is double that of the domestic banks and the returns on assets (RA) is approximately 14% higher (Tábora, 2007).

2. Mergers and Takeovers

As a result of the implementation of the Basel standards and the increased exposure to international competition stemming from the globalization of the financial system, the banking industry has experienced a major process of concentration.

Table 4.4 Honduras: efficiency differences between foreign and domestic banks, September 2004 (%) (Prepared by Tábora, 2007)

Variable

CARa

FRb

CLc

IMd

Administrative costs
/productive assets

RPAe

RAf

Domestic Banks

14.85

10.63

4.51

6.12

6.58

1.18

9.47

Foreign Banks

18.48

13.90

4.21

9.69

7.45

2.43

10.78

a Capital adequacy ratio.

b Financial returns.

c Cost of liabilities.

d Intermediation margin.

e Returns on productive assets.

f Returns on assets.

In addition, the deficiencies in regulation and supervision suffered by various countries throughout the 1990s created the right conditions for a series of crises which stepped up the concentration process. As can be seen in Table 4.6, the five main banks' share of total assets is very high. Nevertheless, the consequence of the relatively similar sizes of the main operators was that the Herfindahl-Hirschman Index (HHI), which measures all the individual firms' share of the industry,14 stood at around 2,000 in the majority of cases.

In Mexico, as of 1990 the privatization of the banks led to a period of rapid growth in the activity and in the number of banks. In 1991, there were 12 banks, rising to 22 in 1992, 27 in 1994 and 33 in 1995. The 1995 crisis reversed the process: by 1996 the number of banks was 31, despite government rescue efforts, and continued to fall over the following years to only 20 in 1999. During that period a mass entry of foreign capital began, and the mergers and takeovers coincided with the country's main banks being taken over by foreign institutions. Thus, the Spanish bank, BBVA, which started operating in 1994, absorbed three existing banks between 1995 and 1999: Oriente, Probursa and Cremi; Santander, also a Spanish bank, began operations in 1995 and in 1997 absorbed Banco Mexicano; whereas Citibank, which was already operating in 1991, took over Confia. Meanwhile, in 1999, another three banks, all of them among the country's largest financial concerns, were capitalized by foreign banks: BANCOMER, capitalized by the Bank of Montreal; SERFIN, capitalized by the Hong Kong & Shanghai Bank; and BITAL, capitalized by Banco Central Hispano and Banco Central Portugués. Likewise, both BANCOMER and BITAL absorbed other entities: the former, PROMEX and Unión, and the later, Atlántico, Interestatal and Sureste.15 During the period 2000–2003, the total number of banks fell from 20 to 18, which involved large-scale bank mergers and changes in control from Mexican owners to foreign ownership: BBVA, recently merged in Spain, absorbed BANCOMER, and Citibank acquired Banco Nacional de México (BANAMEX), Mexico's largest bank16; HSBC took over BITAL; Scotia acquired Inverlat and Santander absorbed SERFIN. These developments left all the major Mexican banks in the hands of foreign capital. (see Avalos and Hernández, this volume).

Central America also underwent a process of bank consolidation over the same period, although the different cases vary among themselves. As we can see in Table 4.5, taking all the banks into account, including affiliates of foreign entities, the most

14 Defined as the sum of the squares of the relative shares of all the firms in the entire industry. The result falls between 0 and 10,000. Zero indicating perfect competition and 10,000, absolute monopoly. The Department of Justice considers a market with HHI lower than 1,000 as competitive; with a value between 1,000 and 1,800 as moderately concentrated and 1,800 or greater as highly concentrated.

15 Throughout the period analysed (1990–2003), foreign banks' affiliates were not taken into account.

16 The operation involved resources in excess of US$12 billion (see Avalos and Hernández, this volume).

Table 4.5 Central America and Mexico: evolution of the number of banks (Central American Monetary Council at www.secmca.org/Estadisticas_Indicadores_Ban.htm and Mexico's National Banking and Securities Commission, at www.cnbv.gob.mx/default.asp?com_id = 0)

Country/year

1998

1999

2000

2001

2002

2003

2004

2005

Total

231

214

207

196

193

184

176

174

Costa Rica

23

23

22

21

21

19

19

19

El Salvador

17

15

14

12

12

12

12

11

Guatemala

34

34

32

31

31

26

25

25

Honduras

23

22

21

21

19

16

16

16

Nicaragua

12

12

8

6

6

6

7

6

Panama

85

78

75

74

71

73

67

68

Mexico

37

30

35

31

33

32

30

29

Note: From 1998 to 2004, the figures refer to banks in existence in December each year; for 2005, to those in existence in June.

intense process of reduction took place in Nicaragua, where the number of banks dropped from 12 to 6. As in Mexico's case, the concentration process in Nicaragua took place after a period in which the banking system experienced major growth: as of 1990, following the end of the Sandinista regime, ten private banks were created while at the same time the importance of the state banks declined. The banking sector became consolidated with the entry of Nicaraguan-owned private banks which in 1999 concentrated over 90% of the system's deposits, with a relatively well-balanced distribution of assets and liabilities. However, in the year 2000, the Nicaraguan financial system was shaken by a series of bankruptcies that threatened the country's financial stability. The crisis broke suddenly with the takeover by the Superintendence of the country's largest bank, Banco Intercontinental (Interbank), which at the time held 14% of the system's total assets.17 The Interbank intervention had serious consequences as regards account holders' confidence in the stability of the banking system. Between June and December 2000, an outflow equivalent to 8.2% of the public's deposits was registered, reducing the total during the same period from US$1.539 billion to US$1.414 billion. Thus, other banks in the system which were already facing difficulties were affected by this withdrawal of deposits; the loss of credibility combined with portfolio irregularities led to the collapse of another four banks18 (Ansorena, 2007).

In view of the size of the institutions affected, the authorities decided to guarantee the totality of deposits and the majority of liabilities. The assets and liabilities were transferred to other local banks by auction, whereas the Central Bank of

17 What appeared to be a well-managed portfolio turned out to be a situation in which 80% of loans, mostly unrecoverable, had been made to individuals or companies linked to an agro-industrial economic group (Ansorena, 2007).

18 The four institutions taken over by the Superintendence during that period represented in June, 2000, approximately 40% of the system's assets and 39% of its deposits. The liquidity and solvency problems arose as a result of the rapid growth of risk assets, their concentration and the proliferation of related loans, in addition to external factors.

Nicaragua (BCN) covered the difference between liabilities and assets by issuing medium-term internal-debt securities (CENIS) equivalent to almost 16% of GDP. This decision led to increased concentration in the sector: as a result of the auctions, Banco de la Producción (BANPRO) absorbed three banks, Pribanco, Banco Intercontinental and Banco Nicaragüense (BANIC), while Banco de Finanzas (BDF) absorbed Banco del Café and Banco de Credito Centroamericano (BANCENTRO) took over Banco Mercantil. As can be seen in Table 4.6, this process increased the five largest banks' share of total assets to 95.71% in 2001, reaching 99.57% in 2005. This situation was reflected in the evolution of the HHI from 0.09 in 1999 (when there was greater distribution among the nine largest banks) to 1,900 in 2004, above the critical level of 1,800 (Ansorena, 2007).

After Nicaragua, El Salvador was the country that suffered the greatest reduction as a percentage of total national banks during the period 1998–2005. It is also the country, after Nicaragua, showing the greatest asset concentration: in 2005, the five largest banks held 91% of assets (see Table 4.6). This is expressed by the pronounced growth of the HHI for total assets (see Table 4.7). In this case the mergers

Table 4.6 Central America and Mexico: five largest banks' share of overall assets, 1997–2005 (%)a (Central American Monetary Council at www.secmca.org/Estadisticas_Indicadores_Ban.htm and Mexico's National Banking and Securities Commission, at www.cnbv.gob.mx/default.asp?com_id = 0)

Country/year

1997

1998

1999

2000

2001

2002

2003

2004

2005

Costa Rica

79.65

77.57

76.80

75.24

72.54

73.00

71.54

76.00

76.83

Honduras

51.21

51.97

51.17

59.66

60.98

64.03

66.83

66.12

66.08

Guatemala

38.12

38.84

39.81

43.45

53.86

55.29

59.70

60.62

62.15

El Salvador

68.74

69.71

75.04

85.03

89.46

89.32

88.40

86.93

91.00

Nicaragua

63.32

56.54

60.82

80.00

95.71

95.63

95.96

95.93

99.57

Panama

n/a

36.80

37.59

39.33

42.83

41.92

41.51

43.84

45.77

Mexico

71.79

67.26

66.39

69.65

73.62

72.76

75.81

74.67

80.02

a Equivalent to the five main banks' share of the total banking sector in each country.

Table 4.7 Central America: Hirschman-Herfindahl index total assets, 1996–2005 (Based on official sources from Ávalos and Hernández in this volume, Balsells, 2007, Herrera, 2007, Tábora, 2007 and Yong, 2007)

Year

Costa Rica

El Salvador

Guatemalaa

Honduras

Mexico

1996

2,095

1,136

503

722

n/a

1997

1,743

1,168

490

735

n/a

1998

1,716

1,158

495

743

1,639

1999

1,735

1,294

489

752

1,547

2000

1,680

1,790

559

889

1,485

2001

1,533

1,921

711

901

1,621

2002

1,515

1,876

778

971

1,537

2003

1,598

1,865

855

1,077

1,528

2004

1,842

1,585

861

1,057

1,458

2005

1,875b

n/a

n/a

n/a

n/a

a Equivalent to the HHI of total credits.

b Figures correspond to March.

and takeovers process did not stem from a financial crisis, but from strategies aimed at strengthening the banks' positioning in the domestic market and attaining scales of activity to enable them to operate at the Central American level. The first merger took place in 1997 and involved an international bank from outside the region: in October that year The Bank of Nova Scotia (Scotiabank) acquired 53% of the shares of Ahorros Metropolitanos, making it the first international bank to hold a majority share of a Salvadoran bank. Nova Scotia increased its shareholding in Ahorromet to 98.3% in December 2000, when it changed its name to Scotiabank El Salvador. In August 1998, two small banks, PROMERICA and Banco Corporativo (Bancorp),19 merged. In July 1999, BANCOMER took over Banco Atlacatl. In May 2000, the merger began between Banco Agrícola Comercial, the country's largest bank (with a 20.27% share of total credits, 24.55% of the deposits market and 23.04% of the system's overall assets) and Banco Desarrollo, which in turn was in fifth place in the system. In July 2000, the merger between Banco Salvadoreño and Banco de Construcción y Ahorro (BANCASA) took place.

In Guatemala's case, the number of banks dropped from 34 in 1998 to 25 in 2005. The reduction in the number of banks was associated with the 1998 crisis20 and with corporate strategies aimed at strengthening the institutions' positioning in the domestic market. In 1999, a period of mergers began which continues to this day. That year Banco del Café absorbed Multibanco, thus rising to second place by size of assets; in 2000, Banco Reformador took over Banco de la Construcción, passing from eighth to fourth place in the same ranking; in September, authorization was granted for the merger between Banco del Agro and Banco Agrícola Mercantil, forming the new Banco Agromercantil de Guatemala, second place in assets that year; in March 2001, Banco Granai & Townson and Banco Continental merged, thus creating Banco G & T Continental, which became the country's principal bank; in November, 2002, Banco CHN took over Banco del Ejército, as well as another small bank, Banco Del Nor-Oriente (BANORO), the following year; finally, in April, 2004, Lloyds TSB Bank, Guatemala Branch, made over the totality of its credit assets in favour of Banco Cuscatlán de Guatemala, as a result of which Banco Cuscatlán rose from eleventh to eighth place in the market upon increasing its assets by 50% (Balsells, 2007).

The mergers process in Guatemala began in a context of low concentration levels. In fact, in 1998 the five largest banks' share of overall assets was 38.84%. Despite the major increase in the five largest banks' share of total assets in 2005 (62.15%), Guatemala continues to register the lowest concentration levels, the only exception being Panama. Analysed on the basis of the HHI with respect to total credits, the

19 Banco Promerica established its operating strategy through alliances with important chains of supermarkets and shoe stores, Pricesmart and Payless Shoe Source in various Central American countries. This bank's equity is 99.99% in the hands of Salvadoran shareholders according to data in the Boletín Estadístico, December 31, 2004 (Herrera, 2007).

20 In 2001, three private entities were intervened due to causes related to the 1998 crisis. Those banks were Banco Empresarial (February), and Banco Metropolitano and Banco Promotor, both in March that year (Balsells, 2007)

index has never risen above 881, which denotes low levels of concentration. The other characteristic aspect is that apart from the last case involving a foreign bank, Cuscatlán, of Salvadoran origin, the leading role in the mergers and takeovers process has been played by Guatemalan concerns.

Beginning in 1998, Honduras has also experienced a process of consolidation of its banking system, which went from 23 banks to 16 (see Table 4.5). During the period 1998–2002, a number of events had an influence on that reduction: the forced liquidation of Bancorp in 1999, and Banco de Crédito y Servicios (Bancreser) in June 2001. In July 2000, authorization was given for the merger between Bancahsa and Banco El Ahorro Hondureño to form Banco BGA, the system's largest bank until December 2001. In 2001, the operation of Banco Promérica, a member of the Network of Grupo Corporativo PRO, was authorized. In May 2002, CNBS intervened Banco Capital21 and Banco Sogerín. The latter was taken over, together with Banco de las Fuerzas Armadas, by Banco del País in July 2003 (Tábora, 2007).

In Costa Rica the banking system was made up of 16 institutions in 2006, four of which were state-owned. Of these four, three were commercial banks and one was a second-floor bank. The three commercial public banks, Banco Nacional, Banco de Costa Rica and Banco Crédito Agrícola de Cartago, concentrate 65% of the entire system's assets. The private banks, for their part, have undergone a series of mergers and takeovers. Prominent among the mergers was the acquisition in September 1995, of 80% of the shares of Banco Mercantil by The Bank of Nova Scotia, which became Scotiabank Costa Rica; in December that year, Lafise took over Banco del Exterior and became known as BANCENTRO; in December 1996, Banco Cuscatlán took over Banco BFA, which acquired Cofisa in May 1998; in December 1997, Banex took over Banco Continental, and in 2000, Banco del Istmo, of Panamanian origin, made a takeover bid for 75% of the capital stock of Corporación Banex, whose shares were listed on the stock exchange, and then merged with Corporación Metropolitana (which owned Banco Metropolitano); in May 2000, Corporación BCT merged with BANCOMER. In September 2001, Banco de San José (BAC San José, of Nicaraguan capital but established in Costa Rica since a long time ago) acquired Banco Finadesa and UP Bank (an offshore operation in the Bahamas); and in 2002, Banco Istmo acquired Bancrecen.

Two aspects of Costa Rica's experience stand out. First of all, the strong involvement by Central American banks in the mergers and takeovers process is noteworthy. Second, despite this process of takeovers, the five largest operators' share of overall assets remains fairly stable, which reflects the important weight of the public banks (see Table 4.6). As we can see in Table 4.7, Costa Rica is the country with the highest levels of concentration according to the HHI.

For Panama, two features, probably associated with the existence of the international financial district, are worth noting: the large number of banks operating in the country (Table 4.5) and the relatively low share of total assets of the country's five main banks (Table 4.6). The principal stakeholder in Panama's merger and takeover processes is Primer Banco del Istmo: in September 2001, it took over

21 Banco Capital went into forced liquidation due to insolvency in late 2002.

Primer Banco de Ahorros; in September 2002, Banco de Latinoamérica; and in October 2003, Banco Mercantil del Istmo. As a result, Primer Banco del Istmo's total assets increased from US$1.97 billion in December 2000 to US$4.186 billion in September 2004, hence becoming the country's main bank. In July 2002, Banco Continental acquired Banco Internacional de Panamá and Banco Cuscatlán Panamá, of Salvadoran origin. Later, in April 2004, Banco Continental took over Banco Panamericano and Banco General acquired Bank of Boston.

3. Regionalization and Internationalization

The banks' globalization process has been heterogeneous. While Mexican and Panamanian banks are highly internationalized – in Mexico because the local banks were massively acquired by foreign banks, and in Panama because of the creation of a special financial district – foreign banks in the rest of Central America still had a limited presence at the end of 2005.22 The internationalization process has taken place in three different ways in this area: the regionalization of local banks, the services provided to the Central American migrant population in the USA and the creation of offshore banks (which will be discussed in the following section).

In fact, the concentration process that has taken place in the different Central American countries has been interwoven with a dynamic process of regionalization of the banks. Three objectives appear as determining factors in this process. The first objective has to do with providing a competitive service to the productive enterprises that are increasingly operating on a Central American scale, among them companies engaged in telecommunications, beer, energy, air transport, sugar cane and foods in general. The private sector is making rapid headway in the economic integration process and the likely ratification of Dominican Republic-Central American Free Trade Agreement (DR-CAFTA) will give renewed impetus to this process. The second objective of bank regionalization is attaining a larger scale to facilitate modernization processes and the development of infrastructure for the circulation of capital and means of payment throughout the region. The third objective is associated with the expectation that major international banks will arrive in the region in the short term, and that to start operations they will probably be prepared to pay considerable charge to access operating infrastructure and a significant client portfolio.23 The other component of internationalization is serving a

22 This situation started changing radically at the end of 2006, when two regional banks were bought by extraregional banks.

23 This process has developed even though, as stated by the corporation IDC, "Unfortunately the legislation does not yet permit an appropriate regulatory framework for the functioning of a true concept of regional banking, since the regulations and supervisory bodies continue to act separately in each country and banking legislation is still national in scope. In practice, however, the region's principal financial groups compete with one another, with or without physical presence in each country, to cater to the main corporate clients in the Central American region" (Quoted by Balsells, 2007, p. 21).

large contingent of emigrants to the USA. This process has a positive impact on competitiveness of regional banks, but at the same time it poses serious challenges to bank regulators.24

In the case of El Salvador, Banco Cuscatlán is an example of the first strategy. To April 2005, it appeared as El Salvador's second largest bank according to paid-in capital (US$90 million), and as shown in Table 4.8, the bank operates in five of the six Central American countries. Its chief operation is naturally in El Salvador, but as can be seen in Table 4.10, when comparing 2003 and 2004, its profits grew substantially more rapidly in its operations in Guatemala and Costa Rica. Excluding Panamanian banks, Banco Cuscatlán appears as the largest regional group, with assets totaling US$3.101 million, as shown in Table 4.9. Its aggressive expansion policy also includes Panama, where as a result of the merger with Banco Panamericano its assets grew from US$35 million in April 2004, to US$553 million in June the same year.25 Another two Salvadorian banks that have operations in other Central American countries are Agrícola Comercial (in Honduras and Panama) and BANCOMER (in Guatemala). At the same time, El Salvador is also a place where regional banks, with another Central American origin operate: BAC, Grupo Financiero Uno and PROMERICA have a relatively strong position in that country.26

An alternative internationalization strategy is that of Banco Salvadoreño, the country's fourth largest bank, which has placed emphasis on serving Salvadorans abroad by offering services related to attraction and management of remittances. A very important item if we take into account is that the flow of remittances amounted to US$2.547 million in December 2004 and has maintained an annual growth rate of around 10%. By 2006 Banco Salvadoreño was present in Houston, Los Angeles, San Francisco, Las Vegas and the entire West Coast of the USA (Herrera, 2007).

Nicaragua's banks have played a particularly important role in the regionalization process, even though it is one of the region's poorest countries. In fact, the expansion of the Nicaraguan banks to other nations of the region, which started in 1980, with the flow of capital from that country to Miami and other countries of the region, enabled them to strengthen, notwithstanding the weakness of Nicaraguan economy and the abrupt changes in economic policy over the 1980s. BAC for example, as shown in Table 4.10, derived many more profits from abroad than from Nicaragua. A similar situation occurs in the case of Grupo Financiero Uno, with the peculiarity that profits from its operations were of a similar size in each country. The importance of this group's investments in Nicaragua had dropped from 15.95% in 2001 to 8.88% in 2004 (Ansorena, 2007).

24 Among these are the difficulties that arise from regulatory arbitrage, the effects of intra-group transactions and the higher risk of this phenomenon transmission. A detailed analysis on these issues can be found in IMF (2006).

25 Banco Cuscatlán also purchased the local credit portfolio and private banking operations of Lloyds Bank of London (Fernández, 2006).

26 By April 2005, the capital paid for its operations reached 11.5 million, 11 million and 13 million of dollars, respectively (Herrera, 2007).

Table 4.8 Central America: regionalization of the banks (Ansorena, 2007; Balsells, 2007; Fernández, 2006; Herrera, 2007; Tábora, 2007; Yong, 2007)

 

Name of the subsidiary bank in each country

Name and origin of the group

Costa Rica

El Salvador

Guatemala

Honduras

Nicaragua

Panama

Grupo del Istmo (Panama)

Banco Bañes

 

 

Banco BGA

Banistmo

Primer Banco del
Istmo.

Banco Agrícola Comercial.
(El Salvador)

 

Banco Agrícola
Comercial

 

Banco de la
Producción

Banco Caley Dagnall.

Banco Agrícola
Comercial

Corporación Cuscatlán Cuscatlán
(El Salvador)

Cuscatlán

Cuscatlán

Cuscatlán

Cuscatlán

 

Cuscatlán de
Panamá

BAC Internacional (Nicaragua)

BAC San José

Banco de América
Central, ex
Credomatic

Banco de
América
Central

Credomatic
and Ficensa

Banco de América
Central (15.3%)

BAC
Internacional

Banco BICSA Offshore of
public banks (Costa Rica)

BICSA

 

BICSA

 

 

BICSA

Grupo Financiero Uno
(Nicaragua)

Banco Uno

Banco Uno

Banco Uno

Banco de la
Exportación

Banco Uno (formerly
Banco de la
Exportación)

Banco Uno

Red Promerica (Nicaragua)

Banco Promerica

Banco Promerica

 

Banco
Promerica

Banco de la Producción
(22.9%)

 

Lafise/BANCENTRO
(Nicaragua)

BANCENTRO

 

 

 

BANCENTRO

 

Corporación BCT

Banco BCT

 

 

 

 

BCT Bank

Banco Catia

Banco Cathay

 

 

 

 

 

Citigroup

Citibank

Citibank

Citibank

 

 

Citibank

Bank of Nova Scotia

Scotiabank

Scotiabank

 

 

 

Scotiabank

Lloyds TSB

 

 

Banco Lloyds
TSB

Banco Lloyds
TSB

 

Banco Lloyds
TSB

Table 4.9 Assets – regional groups, December 2004 (millions of dollar) (Actualidad Económica, April – May, 2005 [online] (www.actualidad.co.cr/315–316/20.especial_bancos1.html#patrimonio) )

Group

Assets

Cuscatlán

3,101

BAC

1,884

Banistmo

1,421

De Comercio

1,303

De Occidente

1,267

Uno

923

Scotiabank

694

PROMERICA

438

Citibank

420

Table 4.10 Regional groups: ranking by profit level, 2003–2004 (Herrera, 2007)

 

Profits (millions of dollars)

REGIONAL GROUP

December 2004

December 2003

Variation (percentage)

CUSCATLÁN

30,506

23,989

27.17

Banco Cuscatlán El Salvador

24,531

19,522

25.66

Banco Cuscatlán Guatemala

1,732

1,046

65.58

Banco Cuscatlán Costa Rica

4,109

2,895

41.93

Banco Cuscatlán Honduras

134

526

-74.52

BAC

48,670

38,009

28.05

BAC Guatemala

3,902

3,308

17.96

Banco Credomatic El Salvador

7,878

8,652

-8.95

BAC Honduras

2,913

2,644

10.17

BAC Nicaragua

20,230

11,893

70.10

Banco de San José Costa Rica

13,747

11,512

19.41

Grupo Financiero UNO

15,917

15,284

4.14

Banco Uno Guatemala

2,805

2,254

24.45

Banco Uno de El Salvador

3,827

5,483

-30.20

Banco Uno Honduras

1,123

1,050

6.95

Banco Uno Nicaragua

6,955

5,417

28.39

Banco Uno Costa Rica

1,207

1,080

11.76

PROMERICA

18,082

14,293

26.51

BANPRO Nicaragua

13,473

11,574

16.41

Banco Promerica de El Salvador

1,604

1,146

39.97

Banco Promerica de Costa Rica

1,982

776

155.41

Banco Promerica de Honduras

1,023

797

28.36

CITIBANK

6,070

7,001

-13.30

Citibank N. A. Guatemala Branch

2,868

5,715

-49.82

Citibank El Salvador

-173

-1,250

-86.16

Banco de Honduras (Citibank)

1,789

777

130.24

Citibank Costa Rica

1,586

1,759

-9.84

Four Nicaraguan banks were participating in the regionalization process in 2004, as can be seen in Table 4.8. Both BAC and Grupo Financiero Uno carried out operations in all the Central American countries, including Panama. PROMERICA, for its part, had operations in three countries, Costa Rica, El Salvador and Honduras, while LAFISE/BANCENTRO had opened operations in Costa Rica. As can be seen in Table 4.9, BAC, Grupo Financiero Uno and PROMERICA were in second, sixth and eighth places among the main regional groups. In contrast to El Salvador, Nicaragua is not a field of operation for regional financial groups from the other Central American countries. In December 2004, the Panamanian Banco del Istmo began operations in Nicaragua and the Panamanian financial group, ASSA, acquired 20% of the shares of the Nicaraguan BDF (Ansorena, 2007). Finally, it is worthwhile noting that Banco Agrícola Comercial took over Banco Caley Dagnall in El Salvador (Herrera, 2007).

In contrast to Nicaragua, the Guatemalan banks have had little regional activity. Furthermore, regional banks' operations in Guatemala are considerably smaller. The share of the mortgage portfolio of the four banks of non-Guatemalan origin operating in that country (BAC, Citibank, Cuscatlán and Grupo Financiero Uno) amounted to barely 3.8% in 2004 (Balsells, 2007).

Honduras is the Central American country in which regional banks had the largest market share. As shown in Table 4.11, in 2004, there were seven foreign banks with non-Honduran capital whose assets amounted to 32% of total assets, a loan portfolio of 30%, deposit-taking of 28%, capital and reserves of 33% and their profits 29%. In general, it may be said that over the first half of the present decade the entry of regional banks was one of the main features of the banking system's consolidation process in Honduras. Eight of the most important groups in the Central American region have penetrated the Honduran market. Panama's Primer Banco del Istmo group took over Banco BGA. BAC Internacional was linked to BAC Honduras (formerly Credomatic) and to Banco Ficensa. Until 2002 Banco Agrícola de El Salvador was associated with Banco de la Producción, which was acquired by Banco Ficohsa in 2003. Grupo Pacific-Banco Uno took over Banco de la Exportación (Banexpo), later called Banco Uno. In 2001, Banco Promérica began operations, linked to Red PRO. In 2004, Lloyds Bank, always considered a foreign bank, was acquired by Grupo Corporación Unión de Bancos Cuscatlán Internacional. In 2005, the arrival of the Lafise group meant that 50% of Honduran banks had foreign ownership interest. Banco de Honduras Citibank has always been considered a foreign-owned bank (Tábora, 2007).

The Costa Rican banks are among the largest in the region, though this does not imply an important position outside the country. In terms of net worth, in 2004,

Table 4.11 Honduras: participation of foreign banks, 2001–2004 (Tábora, 2007)

Item

2001

2002

2003

2004

Foreign-owned banks

7

8

7

7

Foreign banks/totals banks (%)

33.33

42.11

43.75

43.75

Total assets (%)

26.22

30.22

31.18

32.86

Loan portfolio (%)

23.83

25.87

27.99

30.05

Deposits (%)

25.68

28.88

27.46

27.94

Capital and reserves (%)

27.36

29.47

30.20

33.14

Profits (%)

31.21

25.78

28.11

28.91

Costa Rica's three public banks featured among Central America's five main banks (excluding Panama). From the point of view of assets, Banco Nacional de Costa Rica, with assets amounting to US$3.132 billion, stood as the region's largest bank (excluding Panama) and Banco de Costa Rica was in fourth place with assets totaling US$2.128 billion. However, these banks' regional presence – through their subsidiary Banco BICSA in Guatemala and Panama – was minimal. By contrast, Costa Rica's second largest private bank, Banex, was owned by the Panamanian group Banistmo. Questions arise as to why Costa Rica's main banking institutions did not appear to be starting up activities in the other countries of the region, and this will certainly have an impact on its banking sector and on the productive branches of the economy. The country's more advanced economic development may not translate into an equivalent position for its banks in the region.

4. Offshore Operations

One of the strategies followed by banking groups to contend with the local regulations and at the same time compete with both local and foreign banks has been the establishment of banks in markets subject to laxer regulations, especially in tax and monetary matters. Arbitrage in those fields has turned some financial markets into what are known as "offshore financial centers", which means that most of the financial institutions handled by them are from clients resident in other jurisdictions.

Carrying out offshore operations has various advantages, though some of them are an obstacle for the financial system to work efficiently. Offshore deposits remain free of taxes at the source, and depositors are not obliged to include the interests earned in their income tax declaration, since in most of the countries of the region income received outside the national territory is not subject to taxation. Furthermore, banks do not have to keep part of those deposits in the central bank as cash reserves, which enable them to pay higher interest on their deposits or charge less interest on loans. Finally, depositors believe that the risk of their assets becoming known is lesser in the custody of an offshore bank.

On the other hand, some of the regulations governing credit operations can be evaded if carried out by an offshore bank, especially as regards limits on the amount of the credit extended to one same interest group or to companies belonging to a particular economic sector, as well as limits on persons related to the interest group that owns the bank. Also, when a non-domiciled bank is established in a centre with deficient regulation, it is common practice for some of them to clean up their credit portfolio and evade capital adequacy requirements by transferring their deficient portfolio to the offshore bank. Finally, offshore centres can be used by banking groups to minimize their tax payments by simply transferring a large proportion of the group's income to the offshore bank through accounting operations and leaving most of their expenses in onshore banks.

Among the activities carried out by offshore banks, credit operations are the ones that most concern the authorities, especially when the banks are in poorly supervised centres. Offshore banks' depositors are generally the same as those of the onshore bank, so if the quality of the credit portfolio compromises the solvency of the offshore bank and endangers its depositors' funds, this would produce a loss of confidence in the onshore bank and could place the stability of the financial system as a whole at risk. More recently there has been concern that offshore operations could be linked to money laundering and the financing of terrorism. Therefore a growing number of multilateral institutions, countries and jurisdictions are doing their utmost to raise the quality of supervision and regulation of offshore financial centres.

In the second quarter of 2003, total credits from offshore financial centres to non-domiciled private and public agents amounted to US$1.8 trillion, one third of which were extended in the USA. Since the end of 2000, US banks have been the largest users of offshore banking services (BIS, 2005).

Deposit-taking by offshore banks within the national territory is prohibited in all the countries of the region except Guatemala: no bank domiciled abroad can take savings from the public unless it has an authorized branch in the country. There are, however, no restrictions on nationals investing in banks domiciled abroad, or on local banks carrying out commercial transactions with such banks; this enables local banks to take savings from the public on account of their offshore banks, and this is presented to the authorities as fund transfer services provided to their customers.

According to a study carried out by the Central American Monetary Council, in 2004, the deposits taken by Central America's offshore banks (excluding Mexico and Panama) amounted to just over 50% of the deposits taken by banks domiciled in the region, whereas the credits extended by offshore banks that same year represented 62% of those granted by domiciled banks (see Table 4.12). These figures reflect the importance of offshore operations for the region's banks.

These proportions tend to be confirmed by the information supplied by the countries' regulatory bodies. In Guatemala, for example, all the country's largest financial groups have offshore banks which handle assets equivalent to 35% of those of the onshore banks and to 20% of the entire group's assets; over 54% of offshore assets are concentrated in three of the country's most important banks: Industrial, G & T Continental and Bancafé. In Costa Rica, where the situation is more pronounced,

Table 4.12 Central America: credits and deposits of domiciled banks that report to BIS, 2002–2003 (million dollars) (CMCA. Implications of the operations with offshore banks, February, 2004)

 

        Credits

        Deposits

Country

2002

2003a

2002

2003a

Total

9,967

7,631

11,035

7,317

Costa Rica

4,122

3,029

3,702

2,169

El Salvador

2,030

1,193

1,023

417

Guatemala

2,179

2,070

3,122

3,254

Honduras

850

655

2,614

1,054

Nicaragua

786

684

574

423

a Data up to June, 2003.

to December 2004 offshore banks held assets, liabilities, credits and deposits representing 55%, 56%, 60% and 77%, respectively, of the assets, liabilities, credits and deposits of all the private banks (i.e. excluding the public banks, which have no offshore operations) and 71%, 72%, 77% and 101% of the assets, liabilities, credits and deposits of the private banks whose group possesses an offshore bank.

5. Specialization and Diversification

One of the banks' competitive strategies is specialization or diversification. This strategy can be applied in three main areas: products, services or market niches dictated by clients' income level or economic activity. Decisions on specialization or diversification in one of these areas imply decisions in the other areas: priority attention on one type of clientele often leads to emphasis on the subset of products and services that best meet the needs of that clientele. Although the size of the Central American market does not allow highly specialized banks to emerge, and even less second-floor banks, certain patterns of specialization or diversification can be detected as competitive strategies implemented by different banking groups. The general trend, however, has been for financial groups to become involved in all types of business, even though the legislation has led them to do so through independent concerns belonging to a holding company.

First, banks can specialize by their clientele's income level. Some banks specialize in small-customer deposits, which means they have an extensive network of branches and automated teller machine (ATM)s to facilitate deposit-taking from the public at large, and also credit programmes specializing in small-scale farm operators or micro and small-sized enterprises. By contrast, banks specializing in corporate clients, such as BCT of Costa Rica – one of the few remaining locally owned private banks – have practically no branches but instead rely on highly specialized account executives in charge of providing personalized service and meeting all the client's needs. Both these banks and those specializing in lower-income segments do not offer the whole range of financial services, but rather those that respond to their clientele's needs. For example, banks specializing in small savers in general do not have offshore banks, investment banking or a significant international operations division – except where remittances are concerned – whereas banks specializing in corporate banking do not manage pension funds domiciled in their countries (although they do trade in international funds).

Foreign banks operating in the region, such as Citigroup or NovaScotia, typically specialize in the corporate segment, and prominent among their clients are transnational corporations. Such is the case with Costa Rica and Panama. On the other hand, small savers and customers requiring credit have been catered to by medium and small banks – which find it harder to compete with larger banks and foreign banks in the corporate sector – or by large public banks. Among the public banks that deal with small savers and people requiring credit are the Costa Rican ones, which concentrate around 65% of the country's deposits and together possess the largest portion of the branch network. Among the small- and medium-sized banks serving this segment are, for example, the Guatemalan banks Banco de Desarrollo Rural, Banco Crédito Hipotecario and Banco de Antigua (the latter Chilean-owned, with operations also in Costa Rica under the name Financiera Miravalles). These three Guatemalan banks are relatively small and have specialized in catering to small-scale farm operators and small-credit customers who are practically neglected by the larger banks. In addition to direct credit for farmers, the first of these has established alliances with a network of operators which it serves with second-floor banking services.

Recently the segment of small savers and credit customers has begun to be considered as customers by banks which had so far specialized in higher-income sectors. To that end they are relying on technical and financial support from international agencies such as Inter-American Development Bank (IDB) which are interested in promoting the supply of banking services to the rural sector, especially in countries where it has a low level of modernization.

Second, banks can specialize in serving specific economic groups. Such is the case with banks that have emerged to solve the lack of financial services for a particular economic group, especially in agriculture or industry. A typical example of this type of banks is the above-mentioned Banco de Desarrollo Rural of Guatemala. However, changes in the productive structure of the countries of the region have gradually eroded the base of these banks, which little by little have been compelled to open up to other sectors and are losing the specificity of their clientele. Many non-bank financial entities, particularly savings and loan cooperatives, are specializing in clients from a specific type of economic activity.

Third, banks can specialize by type of services. As we have mentioned, this kind of specialization is determined by the needs of the segment covered by the bank. For example, remittance transfer services are provided by a few Guatemalan, Salvadoran and Nicaraguan banks in response to the large-scale emigration experienced by those countries. BAC has gone beyond the simple transfer of remittances and has taken advantage of its position as one of the first regionalized banks to develop a payments network through which it provides cash clearance and settlement services to banks and companies in the region. This bank made greater progress than others in the integration of regional treasurer's offices and has developed money market desks with a regional strategy to manage the bank's and its clients' exchange and trading-portfolio risks.

Some banks also specialize in investment banking activities which are generally carried out through one of the financial group's companies. This activity is especially important among the Panamanian banks, which for tax purposes lend to local companies through the purchase of bonds issued by the latter and structured by the banks themselves. But there are also banks in the region that invest in venture-capital enterprises through their financial group's companies, like Banco Interfin of Costa Rica, for instance.

Another activity in which there has been a certain degree of specialization as a result of tax legislation and enforcement of guarantees is financial leasing, which has a significant volume in practically all the countries of the region. A number of factors may underlie the upswing in leasing, either combined or separate, depending on the country's legislation or institutional framework. First, the routine judicial procedures leading to the enforcement of a guarantee can be very trying and involve considerable notary costs, all of which can be obviated by a leasing contract. In case the debtor does not fulfil the contract, the financed good can be recovered quickly and without high costs. Second, leasing costs may have advantageous tax treatment in view of the depreciation of assets and therefore leasing may be useful as a tax shield. Third, leasing contracts tend not to be recorded in the accounts as credits for purposes of banks' capital adequacy, which enables those that have little leeway to channel their credit resources in this manner. It should be mentioned that leasing contracts are generally handled by one of the financial group's companies, not directly by the bank.

Finally, there is the whole chapter on the stock market. Here it is worth recalling that stock exchanges have had disproportionate weight in the organization of the Central American stock market due to the legal provision in the countries of the region – except Mexico – that all transactions involving securities, whether shares or debt must be carried out on the stock exchange and through its members (stock-brokerage seats or firms).27 This has not been the case with the neighbouring country, Mexico. The conditions just described have hindered the development of interbank markets and has compelled the banks to establish stockbrokerage firms in order to deal at least with their own bank's liquidity needs. These stockbrokerage firms provide their clients with securities intermediation services and in some cases hold contracts with foreign stockbrokerage firms for the management of those clients' accounts, for which they obtain lower service fees.

In addition to stockbrokerage firms, pension funds and more recently mutual funds have been organized in the region. Costa Rica pioneered the organization of these products. Initially, pension funds were managed by private banks under the concept of trust fund or trust commission. The legislation in all the countries has gradually obliged financial groups to establish companies for the group's different activities and efforts have been made to clearly differentiate strictly banking products from medium-term investment portfolios (mutual funds) and long-term portfolios (pension funds). In general, foreign banks have not entered the business of mutual funds, because restrictions such as "Chinese walls" and risk diversification imposed by head offices on their branches are practically impossible to comply with on the scale of the Central American market; nor have the banks specializing in service to corporate clients, whose sales force is small and highly specialized, entered the pension fund market.

The development of stock market activities in Costa Rica is interesting from the viewpoint of competition. Originally developed by non-bank financial groups, mutual funds have gained growing importance and all the financial groups have gradually been obliged to establish fund management companies and offer them to bank customers before they transferred part of their resources to mutual funds managed by

27 This obligation is very restrictive and has tended to be abandoned at an international level in recent years.

the competition. Most financial groups allowed fund management companies to compete with the bank for deposit-taking, so that much of the competitive pressure on banks came from mutual fund management companies, including those belonging to the same financial group. The development of mutual funds in Costa Rica paved the way for improvements in banking services and for increases in interest rates on deposits. It is no coincidence that in the other countries of Central America, where financial development has been determined more by the banks, the development of the stock market, and particularly mutual funds, has been much more limited.

IV. Performance of the Banking System and Development of Competition in the Financial Sector

The past 15 years have been a period of substantive changes in the region's regulatory frameworks. These changes have paved the way for the consolidation of private financial systems, with the exception of Costa Rica, where a private system has also become consolidated, but in coexistence with a powerful public sector. Furthermore, progress has been made in modernizing the institutional regulatory framework and applying the Basel standards. The banks have also undergone a period of intense activity and transformations. Many banks have appeared and disappeared. The sector has been subject to an active process of mergers and takeovers. The banks have begun to operate on a regional scale en masse and have extended their operations to all types of financial activities. From the institutional point of view, a notable absentee has been the competition agency, and the consequences of this institutional vacuum have been aggravated by the scarce or non-existent concern of the regulatory institutions for competition in the industry.

Bank performance will be analysed in three spheres. The first perspective, from the point of view of prudential indicators, is: solvency and quality of the portfolio. A second perspective focuses on its role in the canalization of savings to the private sector. A third one focuses on the point of view of the sector's profitability.

From the prudential perspective, banks have had to increase their net worth in order to respond to the new capital requirements imposed by the Basel standards, which has contributed to the expansion of their operations and activities. However, as can be seen in Table 4.13, the results are varied. Whereas in some countries the system's average solvency has been substantially strengthened, in others it has remained stagnant and even deteriorated. In fact, in several countries the net worth/total assets ratio has increased substantially, such as in Nicaragua (which has risen from 4.85% in 1997 to 8.25% in 2005) and in Mexico (8.03–12.27% in that same period). El Salvador and Panama also stand out in that regard.

Guatemala and Honduras contrast with the above. The former has still not recovered in 2005 from the 2001–2002 crisis. This explains why the net worth/total assets ratio dropped from 9.98% in 1999 to 7.54% in 2005. In Honduras, the ratio has remained practically stagnant since 1998, although the levels were not as precarious as Guatemala in 2005.

Table 4.13 Central America and Mexico: bank capitalization, 1997–2005, net worth/total assets (%) (Central American Monetary Council at www.secmca.org/Estadisticas_Indicadores_Ban.htm and Mexico's National Banking and Securities Commission at www.cnbv.gob.mx/default.asp?com_id = 0)

Country

1997

1998

1999

2000

2001

2002

2003

2004

2005

Costa Rica

8.71

8.79

10.07

9.94

9.02

9.12

9.49

9.71

9.82

Honduras

8.39

8.53

8.70

8.79

9.00

8.08

7.65

8.40

8.52

Guatemala

7.81

8.11

9.98

9.08

6.70

5.27

7.92

7.58

7.54

El Salvador

7.76

7.50

8.34

8.86

8.47

9.01

10.17

10.72

11.23

Nicaragua

4.85

5.03

6.55

7.46

6.13

7.06

7.48

8.81

8.25

Panama

n/a

n/a

8.66

9.34

9.43

10.71

12.42

13.43

13.27

Mexico

8.03

8.31

7.97

9.56

9.42

11.14

11.83

11.20

12.27

Table 4.14 Central America and Mexico: payment arrears ratio in the banking system, 1997–2005, overdue portfolio/total portfolio (%) (Central American Monetary Council at www.secmca.org/Estadisticas_Indicadores_Ban.htm and National Banking and Securities Commission, Mexico, at www.cnbv.gob.mx/default.asp?com_id = 0)

Country

1997

1998

1999

2000

2001

2002

2003

2004

2005

Costa Rica

5.24

3.20

2.71

3.49

2.44

3.20

1.71

2.04

1.61

El Salvador

2.96

4.10

4.55

4.34

3.47

2.83

2.26

2.09

2.01

Guatemala

6.27

6.30

7.69

7.14

8.82

13.80

4.77

4.46

4.45

Honduras

5.36

10.69

10.94

12.19

12.84

12.01

9.66

6.90

6.33

Nicaragua

2.01

2.43

2.61

3.98

3.37

3.49

2.77

2.26

2.48

Panama

n/a

n/a

1.73

1.55

2.63

3.48

2.63

1.71

2.02

Mexico

11.34

11.3

8.91

5.81

5.14

4.56

3.15

2.51

2.27

Note: Figures correspond to December each year from 1997 to 2004. For 2005, the figures correspond to June, except for Panama which corresponds to August.

Table 4.14 shows that the solvency of banking activities was subject to serious threats, to a great extent as a result of instability suffered during that period. This weakness symptom of the banking system explains why regulators have given priority to the sector's stability over competition policy. We should emphasize the grave situation of the Mexican banking system, with high rates of payment arrears which were not overcome definitely before 2004. This situation has to do with the problems caused by the tequila crisis, but it also responds to the mistrust which affected the system's operation.

Throughout the period, the two cases that have shown the poorest performance are Honduras and Guatemala. For the latter, 2002 was the worst year, in the midst of the financial crisis that affected the country. During the rest of the period the figures tend to feature among the region's poorest, but without reaching Honduras's extremes, which are very serious. During the period between 1997 and 2005, the default rate was in double digits. This situation is associated with the ravages of Hurricane Mitch and the persistence of its catastrophic effects.28 Since 2004, however,

28 It should also be pointed out that the destructive effects of Hurricane Mitch followed a period of vigorous growth of credit (Tábora, 2007).

there has been a substantial improvement, even though the figures for that year and for 2005 were still very high and Honduras is in the last place in the region as regards past due to indebtedness. Though with significant fluctuations, the other four countries show similar improvements in their performance, with clear evidence of more solid solvency.

Table 4.15 confirms the precarious situation of Guatemala and Honduras and the degree of solvency of Costa Rica, El Salvador, Nicaragua, Panama and Mexico: whereas in the former two countries the overdue portfolio surpasses funding, in the latter five funding levels are acceptable.

From the point of view of credit access for the private sector, the situation is heterogeneous, though, in general the evolution has not been very positive. Figure 4.1 shows the evolution of private sector credit as a proportion of GDP. First of all,

Table 4.15 Central America and Mexico: credit portfolio quality: funding/overdue portfolio, 1997–1998 (Central American Monetary Council at www.secmca.org/Estadisticas_Indicadores_Ban.htm, Mexico's National Banking and Securities Commission, at www.cnbv.gob.mx/default.asp?com_id = 0, and Statistical Reports of Commercial Banks, several years, seen at www.cnbv.gob.mx/default.asp?com_id = 0)

Country

1997

1998

1999

2000

2001

2002

2003

2004

2005

Costa Rica

110.48

136.84

131.62

101.96

114.76

103.63

147.75

126.88

167.34

El Salvador

82.90

86.73

93.61

93.80

110.34

127.56

148.12

141.98

135.66

Guatemala

29.25

34.54

47.56

53.28

68.45

60.73

68.45

70.30

70.74

Honduras

36.94

19.25

23.01

26.42

28.71

36.91

36.49

60.33

61.80

Nicaragua

149.66

121.72

137.47

83.60

175.01

161.47

170.00

187.90

187.02

Panama

 

 

102.12

128.14

97.24

129.70

136.29

149.85

132.26

Mexico

 

 

104.3

115.3

123.8

138.1

167.1

201.78

198.4

Note: Figures correspond to December each year from 1997 to 2004. For 2005, the figures correspond to June, except for Panama which corresponds to March.

Image

Fig. 4.1 Central America and Mexico: bank credit to private sector as a percentage of GDP, 1970–2004 (Authors' preparation based on official sources)

Image

Image

Image

Image

Image

Image

we should note the case of Mexico, where this indicator was barely above 15% in 2004. A number of factors explain this precarious situation, the 1995 crisis being one of them. Since that year, the banking system began a process of recovery in which foreign capital played a key role. Nevertheless, this has largely been attained to the detriment of credit, for the banks concentrate their investments in securities, particularly those aimed at financing the salvage operation implemented by the state following the generalized collapse of the privatized banks. Whereas securities represent more than 50% of the banks' investments in Mexico, trade and housing credit has remained stagnant since 2001 following a decline with respect to 2000 and only consumer credit has registered vigorous growth in recent times (see Avalos and Hernández, this volume). The low level of private sector credit as a proportion of GDP is of even greater concern if we consider that in relative terms it falls below countries as poor as Honduras and Nicaragua (see Fig. 4.1).

Guatemala places second among the countries of the region in which private sector credit is lowest. As shown in Fig. 4.2, private sector credit does not reach 20% of GDP. This figure comes close to the maximum levels reached in the early 1980s. The political problems and violence affecting the country during that period explain the sharp drop in private sector credit until 1992, when it began to recover. The vigorous growth registered between that year and the year 2000 runs parallel to a major increase in the number of branches and the expansion of the banks to areas which previously had been practically devoid of banking services: during the period 1990–2004, the number of bank agencies rose from 259 to 1,315, an increase of more than 400% and a percentage that reaches almost 790% if we consider only the opening of agencies in the Guatemalan high plateau and Verapaces. However,

Image

Fig. 4.2 Domestic credit to private sector in 187 countries as percentage of GDP, 2004 (Authors preparation based on official data)

Note: The horizontal axis orders 187 countries and some country groups ("world" and "countries with low incomes") according to their domestic credit/GDP ratio. These ratios vary from near 0–300%.

most of the growth in the number of agencies was registered towards the end of the last decade and the growth rate has fallen notably over the past five years (Balsells, 2007). Private sector credit, for its part, has tended towards stagnation. This reflects the problems experienced by the sector at the beginning of the decade. As in Mexico's case, a high proportion of deposits is invested in securities: only around 60% of savings is allocated to credit.

In Nicaragua, as in Mexico, the emergence of private banking was accompanied by dynamic growth in private sector credit. Nonetheless, the low quality of the credits, associated with deficiencies in regulation, led to a marked decline in private sector credit from which the country has not recovered. In fact, Fig. 4.2 shows that in 2004 credit fell off steeply to around 25% of GDP (see Fig. 4.2).

El Salvador is an interesting case. Between 1990 and 2001, private sector credit rose dynamically from slightly over 15% of GDP to more than 45%, reflecting the economy's fast growth and the effects of liberalization. However, as of 2001, private sector credit declined significantly as a result of dollarization and the deterioration of the economy's competitiveness.

Panama contrasts with all the other countries of the region. As shown in Fig. 4.1, private sector credit went up from around 50% of GDP in 1990 to 100% in 2002. These figures have no equal in the region. A drop was recorded between 2002 and 2004 which can be explained for accounting reasons.29

Table 4.16 shows that in all countries, with the exception of Costa Rica and El Salvador, the proportion of deposits from the public that are not used to provide credit has been increasing. This indicates that the evolution of the banking system in those years has reduced its capacity to play the role it should have in a market economy, i.e. that of directing savings to productive activities.

Table 4.16 Central America and Mexico: total deposits as percentage of total credit, 1997–2005 (Central American Monetary Council at www.secmca.org/Estadisticas_Indicadores_Ban.htm and Mexico's National Banking and Securities Commission, at www.cnbv.gob.mx/default.asp?com_id = 0)

Country

1997

1998

1999

2000

2001

2002

2003

2004

2005

Costa Rica

196.89

163.40

159.21

146.95

129.50

130.26

128.19

145.29

148.25

El Salvador

111.14

108.67

112.33

115.83

126.16

113.19

107.81

104.71

100.50

Guatemala

153.74

136.29

123.56

147.31

153.44

161.79

157.92

155.38

154.91

Honduras

119.87

111.17

117.96

120.59

120.96

133.00

127.91

131.99

134.71

Nicaragua

166.12

147.06

130.25

163.45

224.11

223.26

195.17

180.23

177.30

Panama

n/a

n/a

123.16

125.41

120.51

134.77

134.37

123.06

120.72

Mexico

90.48

92.73

102.64

102.42

121.90

115.05

128.67

131.63

123.20

Note: Figures correspond to December each year from 1997 to 2004. For 2005, the figures correspond to June, except for Panama which corresponds to April.

29 As a result of tax treatment on interests received, traditional loans documented with promissory notes were replaced by issues of corporate bonds. These bonds have the advantage for the bank that the interests received are not taxable, but payment of interests by the issuer are income tax deductible. The same banks that extended the credit took charge of structuring and purchasing the totality of the issues (Fernández, 2006).

From the viewpoint of profitability of the banking system, the results have been much better. Tables 4.17 and 4.18 provide information in this regard. In general terms, there has been a significant growth trend in recent years in the rate of return on assets (ROA). Costa Rica is notable for the relative stability of this indicator and its growth trend throughout the period (except 1998 and 2002). In 1999 and 2000, it registered the best performance in the region and over the next two years it was surpassed very slightly only by Nicaragua.

Interestingly, Nicaragua has registered high rates of ROA since 2001 in comparison with the other countries. Mexico's ROA attained the same level only in 2005. Good performance, although affected by the crisis in the sector in the year 2000, has remained at levels that place it second. Similarly, on comparing the rate of return on equity (ROE), the Nicaraguan banking system shows a particularly high performance. Table 4.18 shows that from 2001 to the first few months of 2005, ROE reached an average level of over 30%, more than 10% higher than the next country. The rise in ROE, from 20.66% in 2000 to 30.12% in 2001, coincides with the reduction in the number of banks from 12 in 1999 to 8 in 2000 and 6 as of 2001 (see Table 4.5). The high level of returns also reflects the concentration of private credit in the first five banks, which rose from 60% in 1999 to 80% in 2000 and to

Table 4.17 Central America and Mexico: bank returns, 1997–2005, ROA (Central American Monetary Council at www.secmca.org/Estadisticas_Indicadores_Ban.htm and Mexico's National Banking and Securities Commission, at www.cnbv.gob.mx/default.asp?com_id = 0)

Country/year

1997

1998

1999

2000

2001

2002

2003

2004

2005

Costa Rica

1.58

1.21

1.61

1.80

1.89

1.74

1.99

1.93

2.32

El Salvador

1.65

0.49

0.51

0.64

1.14

1.19

1.16

1.12

1.45

Guatemala

1.39

1.11

1.49

1.21

0.47

0.57

1.37

1.65

1.73

Honduras

3.135

2.235

1.546

1.014

0.993

0.921

1.328

1.454

1.722

Nicaragua

0.94

0.88

1.57

1.46

1.91

1.77

2.03

2.73

2.55

Panama

n/a

n/a

1.35

1.29

1.04

0.63

2.01

2.37

1.96

Mexico

0

2.05

0.89

0.66

0.81

1.1

1.65

1.47

2.55

Note: Figures correspond to December each year from 1997 to 2004. For 2005, the figures correspond to June, except for Panama which corresponds to May.

Table 4.18 Central America and Mexico: bank returns, 1997–2005, ROE (Central American Monetary Council at www.secmca.org/Estadisticas_Indicadores_Ban.htm and Mexico's National Banking and Securities Commission, at www.cnbv.gob.mx/default.asp?com_id = 0)

Country

1997

1998

1999

2000

2001

2002

2003

2004

2005

Costa Rica

17.78

13.46

18.04

17.93

19.70

18.93

20.72

20.20

23.83

El Salvador

21.66

6.16

6.24

7.43

13.15

13.66

11.80

10.88

13.23

Guatemala

17.63

14.26

16.66

12.67

6.10

10.75

15.56

20.24

21.78

Honduras

36.47

26.95

17.90

11.54

11.02

10.90

16.70

18.91

20.13

Nicaragua

18.18

17.84

26.48

20.66

30.12

27.65

28.97

34.56

30.22

Panama

n/a

n/a

15.62

14.03

10.91

6.39

16.74

18.15

14.01

Mexico

n/a

25.56

10.93

6.76

8.58

10.4

14.2

12.98

21.38

Note: Figures correspond to December each year from 1997 to 2004. For 2005, the figures correspond to June, except for Panama which corresponds to May.

95.71% in 2001 (Table 4.6). These levels may be associated with the sharp rise in the HHI, from 900 in 1999 to 1,900 in 2004.

V. Conclusions

Several conclusions can be derived from the analysis done before on the bank development in the region.

The banking sector in the region is very heterogeneous and has undergone important transformations as a result of political and economic convulsion. Banks in some of the countries have reflected the political upheaval of the 1970s and 1980s. Such is the case with El Salvador, Nicaragua and Mexico, where the banks went from being predominantly private to a public banking system which was then privatized once again. The economic crises with serious banking components that affected Mexico and Nicaragua have also had a strong influence on the fluctuations between public and private predominance in the sector. As in other sectors, Costa Rica has shown an almost immutable predominance of the public banking system for nearly 60 years. Panama, with its financial district, has the advantage of possessing the most advanced banking system in the region. Guatemala and Honduras are stable cases which nevertheless have an industry that lags behind the other countries.

At the same time, contestability faced by the sector has increased, particularly from abroad. The development of telecommunications and the increasing capital movement have meant great competition from foreign banks and stockbrokerage firms: from the liability side, many large- and medium-size investors handle their portfolios abroad, and on the assets side, many corporations receive loans from foreign banks. The economies of scale in this activity have allowed local banks to keep control on small and medium-size customers, but there are indications that this barrier will eventually disappear. In those market niches, the bank may face contestability from domestic sources, such as credit cards issued by non-banking operators, the public debt or investment funds. The Costa Rican case is especially relevant in this sense.

In a parallel way, because of this greater foreign exposure, the banking system has tended to concentrate. All countries have experienced this tendency: the number of banks has fallen systematically and various concentration indicators have risen. Nevertheless, this phenomenon does not necessarily seem to have reduced competition. On the one hand, the HHI reflects this greater concentration, but the fact that relative symmetry prevails between leading banks, supports the idea that there is workable competition; on the other hand, the exposure to foreign contestability has forced banks to keep a relatively competitive stand; finally, concentration has occurred within a framework of a regionalization process, with positive consequences for competition in the region.

After having traditionally concentrated in domestic activities, local banks are increasingly participating in regional markets. A growing number of the countries' large banks are developing activities in other countries and the regional nature of these operations becomes, in time, a competitive advantage. Its importance will also grow as the needed regional institutional framework for these private sector activities is completed. Furthermore, the regionalization of this industry seems to intensify competition: for example, while its impact in Guatemala is negligible due to the low penetration of the regional bank, in Honduras there is a very positive effect from regional large bank activities, since the performance indicators of regional banks are substantially better than that of the domestic banks.

The contribution of extraregional banks does not seem to have been as helpful for competition as regional banks have. In fact, and contrary to what has traditionally been thought, the entrance of foreign banks has not meant an improvement in the banking sector performance neither in cost of services nor in other costs, such as commissions, or intermediation margins nor in service efficiency (coverage and proportion of credit going to the private sector, among others). In contrast, what has really increased has been profitability of banks. In Mexico – which, together with Panama, is the country with the greatest presence of foreign banks in the region – bank performance indicators are quite deficient in several areas, especially those regarding credit for the private sector.

Public policies have great challenges ahead. There is no guarantee that by taking control of local banks, foreign banks will improve competition in the banking system or in the services it provides to customers. For the moment, the domestic banks have demonstrated great dynamism and adaptation capacity, but competing with foreign banks is not easy because of some public policies' anachronism. Particularly unfavourable for the financial sector is the poor public infrastructure for payments and public-debt management. The fact that in all countries studied, except for Mexico, the emissions policies do not meet with some basic international standards30 prevents the development of competitive mechanisms for liquidity and risk management. It also makes it difficult to rely on market mechanisms for an efficient monetary policy to work without a rise in financial intermediation costs. While the state is in the process of meeting these challenges and takes on its responsibility for developing an adequate financial infrastructure and public-debt markets, local banks will still be in better shape both to assign savings more efficiently and to handle risks and liquidity, hence transferring the fruits of a growing proportion of competition and technology development to its customers.

Bibliography

Ansorena, C. (2007), Competencia y regulación en las telecomunicaciones: el caso de Nicaragua, (LC/MEX/L.785), serie Estudios y Perspectivas No. 73, ECLAC/IDRC, July.

Balsells, E. (2007), Competencia y regulación en el sector de la banca en Guatemala, ECLAC/IDRC (LC/MEX/L.728/Rev.1), Mexico, January.

30 Public debt emission is done through a variety of instruments in the region which individually do not reach the necessary volume for its secondary market to count with a minimum liquidity level. Besides this, there are problems with information and the way in which distribution, negotiation and settlement of financial instruments are designed, which means further problems for the secondary markets to work well.

Bureau of Industry and Security (BIS) (2005) [online] (www.bis.doc.gov), September.

Camacho A. (2004), Finance Alternatives in Latin America, Recent Developments and Future Outlook, Sumaq Summit 04, May.

CMCA (Consejo Monetario Centroamericano) (2004), Indicadores de dolarización de las economías de Centroamérica y la República Dominicana, September, unpublished.

Fernández, M. (2006), Análisis de la competencia en un sistema bancario con integración financiera internacional: el caso de Panamá, Project ECLAC/IDRC, unpublished.

Group of Ten (2001), "Report on consolidation in the financial sector", January (http://www.bis.org/publ/gten05.pdf)

Herrera, M. (2007), Competencia y regulación en la banca: el caso de El Salvador, serie Estudios y Perspectivas N° 68, ECLAC/IDRC, Mexico, January.

International Monetary Fund (IMF) (1988), Direction of Trade Statistics, Yearbook 1981–1987, Washington, DC.

International Monetary Fund (IMF) (1994), Direction of Trade Statistics, Yearbook 1987–1993, Washington, DC.

International Monetary Fund (IMF) (2000), Direction of Trade Statistics, Yearbook 1993–1999, Washington, DC.

International Monetary Fund (IMF) (2003), Direction of Trade Statistics, Yearbook, Washington, DC.

International Monetary Fund (IMF) (2005), Central America Financial Sector Regional Project (FSRP), Payments Systems Assessment Module, Monetary and Financial Systems Department, Washington, DC., May.

International Monetary Fund(IMF) (2006), Central America: Structural Foundations for Regional Financial Integration, produced by a staff team under the leadership of Patricia Brenner, Washington DC.

Litan, R., M. Pomerleano and V. Sundararajan (2003), The Future of Domestic Capital Markets in Developing Countries, Brookings Emerging Markets Annual Series, World Bank and International Monetary Fund (IMF), Washington DC.

Rodlauer, M. and A. Schipke (2004), Central America: Global Integration and Regional Cooperation, International Monetary Fund (IMF), Washington DC.

Tábora, M. R. (2007), Competencia y regulación en la banca: el caso de Honduras, (LC/MEX/L.818), serie Estudios y Perspectivas No. 91, ECLAC/IDRC, Mexico, November.

Yong, M. (2007)," Competencia y regulación en la banca:el caso de Costa Rica", ECLAC/IDRC, Mexico, unpublished.

World Bank and International Monetary Fund (IMF) (2001), Developing Government Bond Markets: A handbook, Special Meetings Edition, Washington, DC.

This page intentionally left blank

Chapter 5
Advantages and Limitations of Costa Rica's
Experience in Competition Policy:
A Benchmark for the Rest of the Countries
of the Central American Region

Pamela Sittenfeld

Introduction

The Law on Promotion of Competition and Effective Consumer Protection (N° 7472) was enacted on December 19, 1994 (LPCDEC). Its chief objective is to promote and protect the process of unrestricted competition, as well as consumer rights and interests, by preventing and prohibiting monopolies, monopolistic practices and other restrictions on efficient market operation and eliminating unnecessary regulations governing economic activities.

To implement antimonopoly standards the Law created the Commission to Promote Competition (CPC), which began operating in January 1996. The purpose of this study is precisely to appraise the work of these ten years of CPC, with the aim of assessing the experiences, challenges and certain lessons that could serve other countries of the region. In this regard, the context in which Costa Rican regulation came about is similar to that of other countries. The latter are characterized by being small economies, with scarce resources and with a scant culture of competition. Nevertheless, within this reality significant progress has been made which could serve as a reference point for these nations.

Since Costa Rica's enactment of LPCDEC and creation of CPC, there has been dynamic interaction among the different agents involved in the market, and the manner in which they do so has varied over time. For example, at least two types of reaction on the part of companies have been perceived: some have become aware that monopolistic practices are illegal and do not practice them, and others continue to do so but much more surreptitiously. Thus, CPC's work has become more difficult, since at first it could base its findings on sound and easily accessible evidence, but over time its investigations have become increasingly costly and complex. The accumulation of experiences in resolving cases is real capital in improving the manner and the responsiveness with which anti-competitive behaviour can be addressed, and the purpose of this study is precisely to compile the legacy of CPC's decade of experience.

In order to meet the proposed goal, a detailed study will be made of the regulation of prohibited practices and concentrations and the most important cases CPC has dealt with. This will be in order to analyse the most significant lessons in application of the legal and institutional framework and compare them with best international practices. The conclusions of this paper will make a general appraisal to provide certain recommendations for the countries of the Central American region that have recently approved competition legislation or are about to do so.

I. Analysis of Monopolistic Practices and Concentrations

Investigating anti-competitive practices and concentrations is any competition agency's principal task. According to the registers of the Technical Support Unit (hereinafter UTA), CPC carried out 55 investigations in 1995. Of these, eight were related to absolute practices, 13 to relative practices and one case related to a concentration. However, rather than considering the number of cases investigated by the Costa Rican authority, the aim of this section is to consider CPC's experience in investigations and their analysis and resolution. To that end, each part of this section will briefly describe the applicable legislation on the subject, the most representative cases, and best international practices.

Absolute Monopolistic Practices

1. Regulatory Provisions

Absolute or horizontal monopolistic practices are regulated in Article 11 of LPCDEC. Specifically, this Article prohibits acts, contracts, agreements, arrangements or combinations between competing agents among themselves, the purpose or effect of which is to: fix or manipulate prices or exchanges of information for the same end; restrict or limit the supply of goods or services; divide markets by means of clientele, suppliers, time spans or areas; and coordinating offers in bidding processes.

Such practices are analysed according to the per se rule. Precisely because of the harmful effect of these conducts on the process of unrestricted competition, Costa Rican legislators attributed them with the concept of ipso facto null and void.1

In accordance with Articles 27 and 67 of LPCDEC, the main tool available to CPC in carrying out investigations of these practices is the possibility of requesting information, whether public or private, from any economic agent The agency cannot carry out in situ inspections, attachment of documents or seizures, which undoubtedly hampers investigation of absolute or horizontal practices, since obtaining evidence tends to be the most complex factor in this type of cases.

According to Article 28 of LPCDEC, with regard to sanctions CPC can resort to two types of measures. First of all, it can order the discontinuance, correction or

1 In this regard, see López, 1995.

elimination of the practice. Second, it can impose fines of up to 680 minimum wages. In particularly serious cases, it can charge fines of up to 10% of annual sales or 10% of the infringer's assets.

Finally, sanctions can be imposed on both individuals and corporate persons. In order to clearly determine those responsible for the execution of a particular practice, Article 70 of LPCDEC enables CPC to dispense with any juridical forms adopted by economic agents that do not correspond to the facts investigated.

2. Significant Resolutions

This section will study CPC's jurisprudence in order to analyse its evolution. The cases selected seek to illustrate as completely as possible the criteria for analysis used to carry out the investigations of these conducts, as well as the different obstacles encountered by the authority. Special reference will be made to the type of practices analysed, their investigation and the evidence found, the use of the per se rule and the sanctions imposed. Similarly, regulatory limitations will be appraised, as well as the intervention of other public administration agencies.

2.1. The Case of the Bean Market (CPC-19–99)
Case Description

The investigation of the bean market clearly exemplifies the type of cases that CPC dealt with during its early years. This case began with an article on the bean market published in the newspaper Al Día on March 12, 1998. On June 16, 1998, CPC approved the institution of administrative proceedings against the members of the National Chamber of Bean Processors and Similar, for carrying out three possible absolute practices: horizontal price fixing for the 900 g bag of black beans, exchange of information to manipulate or agree on the purchase price for beans in bulk and finally, for dividing up the territory for the purchase of beans in bulk.

Investigation and Analysis of the Practices

Price cartel

To investigate price fixing, the Commission requested the minutes of the sessions of the National Chamber of Bean Processors and Similar, in which the new price for the 900 g bag of beans was clearly agreed upon, as well as the names of the participants in the said agreement. Furthermore, in order to determine whether the cartel had actually been effective in the market, it requested the invoices of the different companies investigated. It succeeded in proving that the prices set by the agents denounced for the 900 g bag of black beans tended to be standardized.

The pleadings of the parties were aimed at certain particular aspects of the participation or not of the agent in the agreement. Specifically, they argued that the agreement was never put into practice and that the latter was not binding either for those who took part in the meetings, or for those who did not attend. However, since it was a question of a per se practice, none of the arguments were taken into account in determining the illegality of the conduct. Moreover, the evidence not only clearly demonstrated the existence of the price agreement, but also proved that it had been applied in the market. The parties' arguments were used only to aggravate or mitigate the sanction.

The agency sanctioned the companies investigated with fines ranging from 22 to 37 minimum wages. It did not succeed in sanctioning the National Chamber of Bean Processors and Similar, since the description of the prohibited conduct established in the regulation only permits sanctioning the competing agents.2

Exchange of information

The second practice investigated was an exchange of information on prices for the purchase of beans in bulk. The main evidence of this practice can also be classified as direct. Specifically, the Commission succeeded in collecting a copy of the minutes of the sessions of the Chamber of Bean Processors and Similar which contained the statements or opinions of the different economic agents discussing the problems that were arising with the purchase price from farmers, as well as the need to seek mechanisms to face the significant increase in prices.

The file contains no evidence that an agreement was actually arrived at or that any proposal to pay a particular price was negotiated. Nevertheless, some of the aims of the exchange of information did reveal a latent aspect of harming the process of free competition.

The parties investigated attempted to justify their conduct. However, since it is a per se practice the alleged justification served only to aggravate or mitigate the conducts, not to free the companies from responsibility.

Thus, the Commission sanctioned the competing companies in the market for purchase of beans in bulk for exchanging information to manipulate prices on the market.3 Finally, the agency declared that the illegal act was ipso facto null and void and in general ordered both the offending individuals and corporate persons to abstain from carrying out similar acts or any other provided for in Article 11 of the Law.

2 One of the companies was exonerated because it was not a competitor in the relevant market and another two because it was not possible to demonstrate whether they knew about the agreement reached in the sessions of the Chamber of Bean Processors and Similar.

3 In this agreement also one of the companies was exonerated because it was not a competitor in the relevant market and another two because it was not possible to prove whether they knew of the agreement reached in the Chamber of Bean Processors and Similar.

Territorial division

Finally, CPC investigated an alleged division of territories. However, according to the resolution, although there was evidence to justify opening proceedings for this practice, during the investigation it was determined that there was never any intent to carry out a territorial division in terms of Section C of Article 11. The companies investigated were therefore exonerated.

Comments on the Investigation

As has been stated, this investigation exemplifies the type of cases CPC dealt with during its early years.4 That is to say, the investigations did not pose much difficulty in terms of obtaining evidence. This generally becomes the most complex element where cartels are concerned. Furthermore, the evidence found in some cases was not only of a direct nature, but also public. Obviously this fact simplified investigation and analysis in such cases.

In addition, it should be pointed out that the fines imposed were relatively low. That is, they were symbolic, probably with very little or no dissuasive or exemplifying effects.

Nonetheless, as agents increase their knowledge of competition legislation, they will probably learn to act less conspicuously in reaching prohibited agreements. Thus, not only will investigations be more complex, but some agreements are bound to remain undetected. As a matter of fact, the airlines case begins to display this new reality.

2.2. The Case of the Airlines (CPC-37–03)
Case Description

One of the cases marking a difference in CPC's jurisprudence is the investigation of an alleged agreement among the airlines to fix travel agents' commissions. This is because it is one of the few proceedings in which the evidence analysed was of an indirect nature, which obliged the agency to go more deeply into an analysis of the circumstances surrounding the case and the existing evidence.

The investigation began with an article in a national daily on January 6, 2000. The article, entitled "Dispute Between Airlines and Agencies," announced the decision by various airlines of reducing the commissions they pay travel agencies for the sale of airline tickets. According to the information published, the reductions would become effective as of January 1, 2000, and the commission per ticket sold would drop from 12% to 6%.

4 In this regard, see CPC-9–95 and CPC-32–00.

Investigation and Analysis of the Practice

Based on these publications, CPC ordered administrative proceedings to be instituted against the airlines that allegedly took part in the price cartel. Specifically, it opened the case against the following airlines: American Airlines Inc., Continental Airlines Inc., United Airlines Inc., Líneas Aéreas Costarricences, S.A., Taca International Airlines, S.A., Grupo Taca, S.A., Compañía Panameña de Aviación, S.A., Aviateca, S.A., Delta Airlines Inc. and Compañía Mexicana de Aviación, S.A.

In this context, the competition authority basically had to prove two things: (i) that the participating agents were competitors among themselves; and (ii) that the agents had taken part in one of the acts prohibited by Article 11 of LPCDEC.

With regard to the first point, CPC defined the relevant market as services provided by travel agencies to the airlines, specifically services related to the marketing of tickets for air transport of persons. In other words, the conduct investigated referred specifically to the commercial relationship between airlines and travel agencies and not to the provision of air transport service.

This point was crucial during the case, since the parties alleged that CPC was incompetent to take cognizance of the acts stemming from concessions to provide air transport service. This is because public services are excluded from the application of LPCDEC. However, in this case the events investigated did not refer to this service, but to a related market such as the commercial relationship between the airlines and travel agencies.

In regard to evidence to determine the existence of the practice, during the investigation CPC failed to find any document directly proving an agreement among the airlines to unify the rates at 6%. The facts the competition authority analysed are set out below.

First of all, it should be pointed out that before the proceeding was initiated, the airlines under investigation paid different commissions to travel agencies (defined as percentages of the value of the ticket).5 It was after communications were sent to travel agencies by the majority of the airlines in December 1999 that the rates were almost simultaneously unified at 6%,6 to be applied in the same geographical region (Central America). Likewise, the date on which the reduction went into effect was very similar among the airlines.7

On the basis of the above, the Commission demonstrated that similar behaviour existed on the part of the economic agents involved. However, it had to explain that the only explanation for such behaviour was none other than the existence of a cartel among the economic agents.

5 The percentages of commissions to travel agencies were between 6% and 12%.

6 The majority of the airlines communicated their decision between December 27 and 30, 1999. Only two airlines communicated it between January 5 and 7, 2000.

7 The majority of the airlines had as the date of entry into effect of the reduction January 1, 2000. Two of them set it for January 7 and 16 and one of them for February 1, 2000.

Throughout the proceeding the airlines pointed out that this similar behaviour was the result of a unilateral decision by each one of them to follow the airlines of Grupo Taca, S.A. In general, the airlines maintained that they found out about that decision through the reservations information systems.8

As a result of these declarations, CPC requested information from the airlines investigated and from various travel agencies on the reservations information systems they use. It was shown that the commission percentage paid to the travel agencies was not included in these systems and that changes in these percentages were communicated by some kind of formal means.

Thus, the analysis of all the above-mentioned evidential elements led the competition authority to conclude that the only reasonable explanation for the uniform fixing of the commission was that this information was already known previously. That is, that there was an exchange of information between the airlines that make up Grupo Taca, S.A. and the other airlines which culminated in an agreement between the latter to standardize those percentages and thus avoid competing for the demand for the service provided by travel agencies.9

Measures and Sanctions Applied

Based on the background information mentioned above, CPC decided to sanction the airlines for forming a cartel to lower the commissions they paid travel agencies. Despite the foregoing, the Commission decided to exonerate two of the airlines: Delta Airlines Inc. and Compañía Mexicana de Aviación, S.A. since these airlines communicated the reduction in their commissions on January 5 and 7, 2000, respectively. By those dates the reduction of the other airlines had already been made public.10

The Commission imposed fines of between 141 and 280 minimum wages. It also ordered all the persons participating in the absolute monopolistic practices to suspend them and abstain in the future from carrying out any act in violation of both Articles 11 and 12 of Law N° 7472.

Appeals for Reconsideration

The fact that the airline resolution basically used circumstantial or indirect evidence to prove the existence of the prohibited conducts gave rise to a major discussion

8 Except for one of the airlines which stated it had known about the information through third parties.

9 In its analysis, CPC used the standard set by the Spanish agency to appraise evidence of an indirect nature. In this regard, see rulings by the Spanish Constitutional Court 174/1985, 175/1985, 169/1986, and 150/1987.

10 It should also be pointed out that the Commission exonerated Grupo Taca S.A., as this company was not a competing economic agent.

within CPC, as well as between the parties in the proceedings. All the arguments put forward in the appeals were rejected on the basis of the information contained in the file and in the arguments set forth in the final resolution.

During the filing of the appeals, however, new factors unexpectedly came to light which made the agency reconsider the position upheld in the final resolution. Specifically, that the conduct adopted by the airlines of Grupo Taca, S.A. was known and therefore foreseeable several months in advance of the events that gave rise to the proceedings, which could have allowed the other airlines to take a unilateral decision to follow the market's leading company at the time when the latter took the anticipated decision.11

Thus, CPC considered that since a further possible rational explanation of the facts existed, and therefore doubts as to whether the conducts reflected a prohibited practice, it was in order to interpret such doubts in favour of the companies sanctioned and to reconsider the challenged resolution. However, the competition agency made it clear that it was not a question of whether or not use was made of the circumstantial evidence to justify the reconsideration, but rather the existence of new factors that admitted another reasonable explanation for the actions investigated.

For all the above reasons, it excluded from all responsibility the companies Líneas Aéreas Costarricences, S.A., Taca Internacional Airlines, S.A., Aviateca, S.A., United Airlines Inc., Continental Airlines Inc., American Airlines Inc. and Compañía Panameña de Aviación, S.A.12

Comments on the Investigation

The airlines investigation undoubtedly set a new course in analysis of evidence in absolute practices. Although in the end, the resolution was reconsidered, CPC made it quite clear that it is possible to use circumstantial or indirect evidence in this type of proceedings, but with certain limitations. The circumstantial evidence should be sufficient and unequivocal. That is, there should be a sole reasonable explanation of the evidential elements. Obviously if the Commission had a wider range of investigation instruments, it would very probably have to depend less on this type of evidence.

This case is also significant because it somehow shows the need for agreements between competition agencies, whether formal or informal, in order to

11 In this regard, it was the representative of the association of travel agents who expressed in the appeal filed that the issues discussed at a meeting in Montreal and the agreements to lower the commissions to travel agencies were disseminated among all International Air Transport Association (IATA) airlines as part of the general obligations of this organization with respect to its members.

12 Two of the members of CPC differed from the majority opinion. In general terms, they pointed out that of the evidence contained in the file, the only reasonable explanation that could be upheld, was an agreement among the airlines to fix travel agencies' commissions. That is, they upheld the opinion of the final resolution.

proscribe anti-competitive practices. In Panama, for example, a similar investigation was conducted on the case of airlines and travel agencies. Had an agreement between the two agencies existed, perhaps the result of the investigation would have been different.

However, the competition authority has also faced other types of challenges and limitations. These are related to exceptions to the competition legislation provided for in other regulations. In this regard, one of the most controversial cases the Commission dealt with was the investigation of the shipping companies. The most important aspects of this investigation are set out below.

2.3. The Case of the Shipping Companies (CPC-39–01)
Case Description

The National Coffee Exporters Chamber filed a complaint against various shipping companies for alleged rate agreements between them, which led CPC to carry out a preliminary investigation, as a result of which it found evidence of a possible anticompetitive practice. Nevertheless, during the course of the preliminary investigation, it was determined that Costa Rica formed part of the Code of Conduct of the United Nations Maritime Conferences, which has the status of an international agreement ratified by the Legislative Assembly in 1975, in Law N° 6074.

Articles 2 and 13 of this agreement establish the possibility of economic agents who are members of the Conference sharing out the market among themselves, as well as setting rates. This means that the economic agents involved in the shipping sector are exempted by the law from the application of competition regulations, since as the Code of Conduct of Maritime Conferences is an international agreement; it possesses a higher status than ordinary law, as provided for in Article 7 of the Political Constitution. Furthermore, the norm is special to the maritime sector, so that even if it were a question of legislation with the same status, the former would prevail over competition regulations.13

Analysis of the Conduct and Complaint File

On rejecting the complaint, CPC criticized the situation and pointed out that effective application of competition norms would depend on consistency with the current legal system. Also, it stated that it was neither advisable nor fair to apply competition regulations to certain sectors only and that this situation was contrary to the Principles of Equality and Freedom of Enterprise, embodied at constitutional

13 This criterion has been upheld by the Office of the Attorney General of the Republic in Opinion N° C-008–2001, in which CPC consulted the force of Articles 4 and 5 of the Law of Cooperative Associations N° 6756 of May 7, 1982, in relation to the provisions of LPCDEC.

level (Articles 33 and 46 of the Political Constitution, respectively). The above was valid particularly when there was no type of justification for excluding a specific sector from the application of competition regulations, as occurred with the maritime sector. It also pointed out that the international regulations could even run counter to the Political Constitution, particularly to Article 46, which prohibits any monopolizing practice or tendency.

Therefore, CPC stated in its resolution that it was necessary to make amendments to the law so that the different norms that comprised it should be consistent among themselves. In the end, the agency had to file away the case, since as the Code of Conduct of Maritime Conferences had higher legal status than the competition legislation; it was unable to open the proceeding.14

Comments on the Investigation

This case highlights the typical limitations that new competition agencies face upon becoming established in countries with a scarce culture of competition, or with legal systems having a protectionist slant. Hence, the activity of competition law is vital in the effort to change long-standing patterns.

In this context, we can cite the interventions of other agencies of the public administration in competition matters. The case of the Chorotega pallets is a good example of the type of obstacles CPC has faced in applying competition regulations.

2.4. The Case of the Pallet Market (CPC-39–96)
Case Description

In April 1993, the Ministry of Economy, Industry and Trade (hereinafter MEIC) issued Decree N° 22111 by means of which it reduced tariff protection for pallets, pallet boxes and other load box platforms from 15% to 1%. At the time, the measure was taken to address supply problems, problems with the quality of domestic pallets, and to protect the nation's forests.

On November 27, 1995 several pallet production companies signed a letter addressed to MEIC. The letter requested the Minister to increase the pallet tariff

14 The Coffee Exporters Chamber instituted amparo proceedings against the final resolution of the case of the shipping companies CPC-39–01 and against the resolution of the appeal for reconsideration confirmed by this resolution. It also filed an unconstitutionality action against Chapter I and Articles 2, 13 and 16 of the Convention on the Code of Conduct of Maritime Conferences. The Constitutional Court considered that the powers conferred by the said agreement on the shipping companies were not contrary to the Principles of Free Competition contained in Article 46 of the Political Constitution and enlarged upon by LPCDEC. It therefore rejected the unconstitutionality action and the amparo proceedings instituted by the Coffee Exporters Chamber.

and if this could be done, the companies pledged to maintain the maximum price to plant of pallets at US$9.50 per unit.

On January 26, 1996, by means of Executive Decree 24783-MEIC, the Cost, Insurance and Freight (CIF) tariff for pallets, pallet boxes and other load box platforms was increased by 10%. This decree stated the following: "The national wooden pallet industry has the capacity to meet domestic demand for pallets in the country's banana sector and other productive and commercial branches at competitive prices and has pledged to sell at the same prices as import prices, or lower, as a result of which it will use 100% of its installed capacity, since it is currently employing an underutilized production scale."

The company Tarimas Chorotega, S.A., which was importing wooden pallets, filed a complaint with CPC (in April, 1996) against eight pallet manufacturing companies for the alleged forming of a cartel to fix the price of their products. It also requested the competition agency to issue its opinion on the actions of the administrative authorities that approved Executive Decree N° 24783, which according to the complainant supported the concerted agreement. Following this company's action, in June 1996, the previous decree was annulled by means of Executive Decree N° 25275.

Investigation and Analysis of the Practice

The Commission instituted ordinary administrative proceedings against the offending companies in June 1996. Once the stages of the proceedings had been completed, CPC appraised the existing evidence and pleadings. First of all, it considered that the letter of November 27, 1995, signed by the companies denounced, constituted clear evidence of a horizontal agreement between competitors to fix the maximum price of their products. The companies claimed that the said letter was signed at the request of the Minister of Economy, Industry and Trade but this fact could not be proved in the proceedings. In any event, by being a per se practice, the circumstances surrounding the agreement were not relevant to determining its illegality.

During the investigation, CPC also assessed the effects of the agreement in order to determine the sanction to be imposed. It requested the manufacturers to furnish the prices actually charged and succeeded in confirming that the majority of the manufacturers did not raise the agreed price of US$9.50. Nevertheless, it pointed out that even with the tariff increase and the prices charged by the producers, the importing companies were able to compete for the sector's tax incentives.

The competition authority also referred to Executive Decree 24783-MEIC. The Commission did not concern itself with assessing the legality of this document, nor whether MEIC had acted within its powers when it raised the tariff. However, it pointed out that the grounds for the said decree were reprehensible and indicated to MEIC that it should not use its procedures to carry out practices prohibited by the law. In particular, it recommended to the Minister of Economy, Industry and Trade in the future not to endorse such agreements by means of an Executive Decree.

Measures and Sanctions Applied

The Commission ordered the elimination of the agreement for seven of the offending companies. It exempted one of the manufacturing companies from all responsibility because it was not a competitor in the market. However, it did not apply any fine, since it considered that the effects on the market were minor.

Comments on the Investigation

It is worth highlighting that this was one of the agency's first cases and its value in terms of jurisprudence lies in having clearly marked the competition authority's independence in technical matters and criteria from MEIC, which appoints its commissioners and allocates its budget. This is encouraging, for it shows that such independence is possible for the competition agencies of small economies or developing countries.

Nevertheless, the Commission showed a certain weakness by not having applied a fine, as the forming of a cartel was so evident. Furthermore, the study on the effect on competition took a poor approach, since it concentrated on analysing particular economic agents' possibilities of competing rather than on the process of competition as such.

3. Conclusion and Recommendation

Having analysed the competition legislation with regard to cartels, as well as the most representative cases in this type of practice, we will now proceed to compare the most significant conclusions from the previous sections in the light of best international practices. To that end, we will use as references the recommendations of the International Competition Network (ICN) and the Organization for Economic Cooperation and Development (OECD), as well as competition agencies with acknowledged experience.

3.1. Regulatory Restrictions

An essential requirement for promoting and defending competition is the existence of a consistent legal framework that permits cross-sectional application. The degree of consistency of the law directly affects the degree of effectiveness in the application of competition legislation.15 As a matter of fact, one of the difficulties CPC has

15 In this regard, see Curiel, 2000.

encountered in the application of competition legislation is that part of the law pursues contradictory objectives. The clearest example is Article 9 of LPCDEC, which excludes public services from the norms established in the chapter on competition. In the same way, we can cite many other sectoral regulations that exclude certain markets from the application of competition norms.16

This limitation has been reflected in several investigations of absolute practices of which the Commission has taken cognizance. Thus, one of the competition authority's most significant challenges is to make sure that the legislation is applied to all the agents participating in the market, in whatever capacity. To that end it must intensify its efforts to promote competition in the public sector and pay particular attention to the opening up of regulated sectors.

3.2. Intervention by Other Public Institutions

Public institutions can play a key role in investigations of horizontal practices, but they can also favour this type of conduct. The agency should therefore approach civil servants by holding seminars, publishing newsletters or simply by making informal contacts with them.

In this context, certain officials are particularly strategic to the agency in investigating cartels, with emphasis on those in charge of procurement or supplies departments in public institutions. These civil servants are the ones who are in the best position to detect collusive bidding among competitors. If we analyse the cases dealt with by CPC in recent years, most of the investigations have been related to price cartels or supply restrictions. This is not to say that price cartels alone exist in the Costa Rican market, but rather that these are easier to detect than others where investigation is more difficult and costly.

In fact, OECD points out that establishing contacts with the procurement departments of public institutions is one of the best practices. This organization's experts aver that such officials have an excellent knowledge of the sector and can also observe certain behaviour patterns in bidding processes that are key indicators in detecting collusive bids.17 It would therefore be very useful for CPC to establish a programme to systematize best practices in this regard, in order to have a wider range of instruments and information to detect collusive bids.

16 An example of the foregoing is the case of the Law on the Sugarcane League N° 7818 of September 22, 1998 and the Law on Cooperative Associations and creation of Infocoop, N° 4179 of August, 1968.

17 As examples of successful training programmes OECD points out the cases of the USA and Canada. Among the activities carried out, mention is made of the publication of a list of behaviours that could be suspicious to public officials of procurement departments, or the establishment of procedures that make the forming of cartels difficult. In Korea, in addition to training officials, the agency monitors certain bidding processes directly (OECD, 1998).

3.3. The Role of the Judiciary

The Judiciary is the public entity that gives CPC's work greater legitimacy, since its resolutions are reviewed by this body. Therefore, one of the best practices recommended by the agencies of more developed countries is to train judges on the subject. In the absence of qualified judges within the country, it is recommended that judges from other countries with more experience in the field be brought in to train their counterparts.

The task of training should be ongoing, since in countries with limited culture of competition judges tend to favour other values over competition law. Major efforts have been made in Costa Rica to train the judges of the Judiciary, and these efforts have directly or indirectly produced very encouraging results. In this regard, the resolution on the real-estate brokers' cartel is a good example.18 Nevertheless, not all the experiences have been positive. Here we can cite the case of the "Code of Conduct of Maritime Conferences," which the Constitutional Court considered did not run counter to Article 46 of the Political Constitution.19

To overcome this problem, the agency should intensify its efforts in the field of competition law in the Judiciary. This entity is part of the regulatory role in matters of competition. Both the legitimacy and the effectiveness of the agency's resolutions depend directly on the Law Courts.

3.4. Business Chambers and Associations

CPC's jurisprudence has shown that business chambers and associations have somehow participated in the execution of anti-competitive practices, whether as promoters of cartels or even as those directly responsible for these conducts. In this regard, we can cite the case of the bean market, the case of the ice manufacturers20 and the case of the airlines. However, the agency has been unable to establish measures against these economic agents, since the legislation lays down that they cannot be sanctioned. Only agents competing among themselves characterize the description of the standard on cartels.

In other legislation,21 this type of organization can be sanctioned for participation in cartels and other anti-competitive practices. It would therefore be important to amend the legislation so that business chambers and associations can also be sanctioned. It is essential for CPC to discourage this type of agreement in the places where competitors typically meet.

18 In this connection, see Court for Matters under Administrative Law. Second Section. Ruling No. 275–2005.

19 See note 15.

20 In this connection, see CPC-9–95.

21 For example, the European Union's Treaty of Rome permits sanctions against business associations or chambers for anti-competitive conducts.

3.5. The Per Se Rule or the Rule of Reason

One of the most important points of discussion regarding cartels is about the standard of analysis that should be used to determine the legality or illegality of these conducts. Specifically, the discussion centres on whether the per se rule or the rule of reason should be used. In the former case, the agency is not required to prove the anti-competitive effects of the conduct, since these are assumed. This rule provides far more legal certainty to economic agents and is less burdensome in terms of the resources needed to investigate a monopolistic practice. In the latter case, the anticompetitive effects of the conduct must be demonstrated, which may sometimes allow an analysis of possible efficiencies. This rule has the advantage of being much more flexible than the per se rule, since the agency must analyse the circumstances surrounding each particular case.

Although each rule has its advantages and disadvantages, for a small agency with limited resources and a nascent culture of competition it is much more advisable to use the per se rule during its early years. Also, it should be pointed out that most experts agree that cartels never or rarely have pro-competitive effects.22 For example, if we analyse the jurisprudence of CPC, many of the cases involving this issue were sanctioned simply as the result of proof of the existence of an agreement. Specifically, in the case of overland container transport,23 it was impossible for the Commission to analyse the effects on the market due to the diversity of the services provided by the economic agents. If it had done so, it would have cost a large amount of resources and possibly the final decision would have been very similar.

3.6. Investigation Powers

The fundamental problem for every competition agency with regard to cartels is furnishing proof of their existence. Owing to the clandestine nature of this type of practice, there is a serious risk of evidence being destroyed or altered. Hence, most competition agencies consider that in addition to the power of requesting public or confidential information, it is essential to be able to carry out seizures and attachments of documents. Some agencies, moreover, have pointed out that this type of tool is their first option when carrying out an investigation for horizontal practices (ICN, 2006).

However, if we analyse the Costa Rican experience, we can confirm that a considerable number of the cases have been detected through newspaper advertisements or the minutes of business associations or chambers. This information alone has been sufficient to launch investigations. Undoubtedly this does not exemplify

22 In this regard, see ICN, 2005a.

23 In this regard, see CPC-32–00

the typical experience of competition agencies in investigating cartels. If we analyse the jurisprudence, and in particular the pleadings of the parties, it may be seen that in several cases the agents investigated were simply unaware of the existence of competition legislation. However, as the culture of competition develops, this type of case will tend gradually to disappear. Moreover, in recent cases investigated by CPC the parties' arguments are much more sophisticated and the agency is facing increasing difficulties in detecting all the cases it could investigate because it does not have access to all the necessary tools.

This situation merits an amendment to the legislation so that the Commission broadens its powers of investigation. It should be pointed out, however, that the type of instruments needed require large amounts of financial and human resources, which could be a limitation for a small agency. Such an agency would therefore have to be very prudent in the use of the said resources and restrict them to especially serious or difficult cases.

3.7. Ten Per Cent of Sales or Assets

One of the most effective instruments for dissuading companies from committing horizontal practices is fines. These must obviously be sufficiently high to discourage economic agents from carrying out this type of practices (i.e. if by forming a cartel an economic agent is likely to obtain a higher profit, it prefers to run the risk of being fined). If we analyse CPC's experience, we find that there has been a significant increase in the amounts of the fines imposed.24 Nonetheless, according to UTA files, up to mid-2006 the agency had not applied any fine of 10% of the annual sales or assets of the wrongdoer in relation to cartels.

However, one possible interpretation of this Article is that the Commission can only apply a fine of 10% of annual sales or the assets of the infringer, which means that it would be difficult for it to apply such a fine in practice, since in the majority of cases it could break the company. This would not only exceed the purpose of competition legislation, but could even be unconstitutional.

It would thus be advisable to clarify this subparagraph of Article 28, or else amend the legislation so as to be able to apply fines ranging from 1% to 10% of the sales or assets of the wrongdoer. Otherwise, in practice it will only be able to apply the parameter of 680 minimum wages, which in cases of small companies could mean a high fine, but for larger companies would probably not succeed in fulfilling the objective of discouraging this type of practice.

24 The investigation on the palm-fruit cartel has been one of the cases in which the highest fines have been imposed. In this case, three absolute monopolistic practices carried out by competing companies in this market were sanctioned. The fines fluctuated between 28 and 569 minimum wages, which was directly related to the size of the companies investigated and their ability to pay. In addition, the individuals involved in the agreements were sanctioned with fines ranging from 1 to 21 minimum wages (CPC-32–02).

3.8. Penalties

Another significant point of discussion with regard to cartels has to do with the possibility of applying penalties, which can be highly dissuasive for the possible infringer. If a leniency programme is also established, this can be very effective in detecting and punishing this type of anti-competitive practice.

Penalties, however, have only been introduced successfully in countries with a long experience in the field, such as the USA. The culture of competition needs to be relatively well established to apply them. In other words, society must be clearly aware that this type of conducts is sufficiently harmful for them to be considered an offence. Otherwise, although this type of measure forms part of a particular legislation, it may never be applied. Furthermore, such measures require a fairly well-developed expertise on the subject in the Judiciary in order to reduce the possibility of committing an error to the utmost.

So although this type of instrument can be fairly useful, in a country such as Costa Rica and probably in various countries of the region, greater experience has yet to be gained in this sphere so that it can be applied.

3.9. The Leniency Programme

Leniency programmes implemented by different competition agencies have been one of the most effective weapons in prosecuting agreements among competitors, since they make it possible to discover practices that might otherwise be impossible to detect (OECD, 2001). These consist of the elimination or reduction of a particular sanction in exchange for voluntary cooperation by one of the infringers.25

In order for this programme to be successful, clarity and certainty for the person investigated are essential. That is, the persons or companies that cooperate should have full certainty as to how the possible advantages will be granted them. Also, these should be appropriate and attractive for the first to collaborate. Hence, programmes offering complete immunity tend to be the most successful. Obviously this last point is directly linked to the sanctions imposed by the authority, since if these are very weak or are applied infrequently; no one will have the incentive to cooperate with the authority (OECD, 2001).

In Costa Rica's case, this would be a programme that could improve cartel investigation processes, since it would make it easier to detect not-so-obvious evidence. Moreover, it could save financial and human resources that the agency uses in

25 One of the clearest examples is that of the US legislation, which grants total immunity in criminal matters. Since this country modified its leniency programme in order to extend the scope of immunity, the number of requests for immunity has surpassed 20 per year and the number of sentences and fines has increased significantly. Likewise, since the European Commission implemented its programme to reduce penalties in exchange for cooperation on cases, it has received more than 20 cases. Also, the European Commission submitted a bill with the aim of maximizing the capability to detect these conducts (ICN, 2006).

investigations. However, the Commission would have to increase the fines currently applied so as to really generate an incentive for persons or companies to collaborate. This factor is an essential requirement for the programme to work.

3.10. International Cooperation

Many competition experts believe the existence of cooperation agreements between competition agencies to be essential in investigating practices with cross-border effects. These types of agreements have been implemented largely by the agencies of developed countries. Nevertheless, developing countries or small economies should follow this same path. For example, if we analyse CPC's jurisprudence a number of cases registered have originated in Costa Rica with effects in other markets and conversely, cases originating in other countries with effects in Costa Rica. In this regard, the case of the airlines and travel agencies began in Canada and the decisions taken by the airlines had effects in Costa Rica, the rest of the Central American countries and Panama. In fact, an investigation was opened in Panama on these actions that were fairly similar to those of the Costa Rican agencies.26

This means it is indispensable for the Commission to establish agreements with Costa Rica's most important trading partners. Specifically, as regards cartels, it would be advisable to sign agreements with the agencies of neighbouring countries, especially if we bear in mind that in such small countries there are strong incentives to fix prices in the region, as well as to split up markets. The manner in which cooperation schemes are established will depend on each case and the resources available for that purpose.

Relative Monopolistic Practices

This section, like the preceding one, will analyse the legal framework for relative monopolistic practices and the most representative cases, in order to appraise the possible advantages and disadvantages of the current model and to contrast the analysis with the best practices in this field.

1. Regulatory Provisions

Competition legislation defines relative monopolistic practices as any acts, contracts, covenants, arrangements or combinations of these between economic agents whose purpose or effect is or may be the undue displacement of other agents in the

26 In this regard, see the newspaper "La Prensa," Panama, June 6, 2001. [online] http://mensual.prensa.com/mensual/contenido/2001/06/08/hoy/negocios/152364.html

market, substantial hindrance to their access or the establishment of exclusive advantages in favour of one or several persons.

Specifically, Article 12 of the legislation prohibits the following conducts: establishment or imposition of exclusive distribution of goods or services; imposition of resale prices; sale or conditioned transaction to buy or sell other goods or services; concerted agreement among various economic agents or inviting them to exert pressure on some client or supplier; and production or marketing of goods and services at prices below their normal value.

Subparagraph (g) of the above Article also sets forth as a relative monopolistic practice any deliberate act that leads to competitors leaving the market or prevents their entry. Although this latter prohibition allows for a certain amount of flexibility in order to adapt to circumstances that have not been expressly provided for in other hypotheses, it has also been criticized for the legal uncertainty it engenders for economic agents. Since this matter is governed by the Principles of Penalizing Administrative Law, the characterization of a conduct should be as detailed as possible.27

As a requirement to determine the illegality of this type of conducts it must be proved that the economic agent that committed the infringement has substantial power28 in the relevant market.29

In other words, according to Costa Rican legislation, three factors must be proved in order to determine the illegality of a relative monopolistic practice: (i) that the economic agent has substantial power in the relevant market; (ii) that one of the conducts characterized has been committed; and (iii) that the said conduct has anti-competitive effects.

However, it should be pointed out that the legislation does not clearly indicate the possibility of analysing the pro-competitive effects or efficiencies that these types of conducts normally have. In other words, the legislation does not lay down the typical rule of reason in which the anti-competitive versus pro-competitive effects of the practice are analysed. Moreover, it could even be interpreted as meaning that the legislation establishes a sort of per se rule, but with the requirement that the economic agent should have substantial power in the relevant market. Despite this limitation of LPCDEC, the Commission has assessed considerations of efficiency

27 It should be pointed out that the legislation does not characterize some vertical conducts and others related to abuse of position of control that are normally prohibited in other countries. Specifically, LPCDEC does not include cross-subsidies, discrimination among economic agents and denial of business.

28 Article 15 of the legislation establishes a series of criteria for determining the existence of substantial power on the part of the economic agent under investigation. Specifically, there must be evidence of the investigated agent's market share and the possibility of its fixing prices unilaterally, the existence of barriers to entry, the existence and power of its competitors, and its recent behavior.

29 Article 14 of LPCDEC establishes a series of criteria that serve as a basis for defining the relevant market, such as: the possibility of substitution of the good or service, distribution costs, its significant inputs, its complements and substitutes, the costs and the possibilities of consumers turning to other markets, and finally, national and international regulatory restrictions. Based on an analysis of these factors, the Commission must define a product market and a geographical market.

in this type of cases, i.e. it has been more flexible in interpreting the regulation established by the legislation.

Finally, it should be pointed out that the powers of investigation and sanction are practically the same as for cases of absolute practices. The only difference is that the fines are limited to 410 minimum wages.30

2. Significant Resolutions

Having described the legal framework in the preceding section, we will study CPC's experience in the analysis and investigation of relative monopolistic practices. However, it is worth noting that its experience in cases of relative practices is considerably less than in those involving absolute practices. Basically, there are two firm, representative cases in this area: the case of electrical components and the case of carbonated beverages and fruit juices.

The first of these investigations – on the electrical components market – was undertaken in CPC's early years. The second investigation – in the carbonated beverages and fruit juices market – was conducted several years later. Clearly, if these proceedings are compared there are major differences in the evidence obtained, the complexity of the analysis, and the fines imposed. Details of the most important features of the investigations are set out below.

2.1. The Case of the Electrical Components Market (CPC-9–95)
Case Description

The General Manager of the company Sigma Dam Accesorios Eléctricos de Centroamérica, S.A. (Sigma Dam) filed a complaint against the company Bticino Costa Rica, S.A. (Bticino), in 1995, for an alleged infringement of LPCDEC. The origin of the complaint was the text of some communications from Bticino to its distributors.

CPC opened the proceeding in order to analyse four possible anti-competitive conducts by Bticino: (i) a possible price reduction below normal value carried out by this company in order to deliberately cause the exit of its competitors from the market or prevent their entry; (ii) an alleged price imposition on its distributors, as well as the establishment of sanctions in case of failure to comply; (iii) a possible exclusive distribution by Bticino in order to introduce its imported products into the domestic market and; (iv) an alleged transaction subject to the condition not to acquire goods normally offered to third parties.

30 As with cartels, in especially serious cases it can impose fines of up to 10% of annual sales or 10% of the infringer's assets.

Investigation and Analysis of the Practices

In order to analyse the above-mentioned practices, CPC conducted a market study in which it requested information from the different participants in the production and marketing chain. Also, it arranged for two appearances so that the parties could present their pleadings.

The above with the aim of demonstrating the existence of the factors of the type provided for in Article 12 of the legislation in order to prove the commission of relative monopolistic practices. These were: (i) Bticino's substantial power in the relevant market; (ii) the characterization of the practices denounced; and (iii) the anti-competitive effects.

On substantial power in the relevant market

CPC set the limits of the relevant market as the market for low-voltage electrical appliances normally used in households. The rationale for this definition was based mainly on the fact that the possibilities for substitution of the goods were minimal, and that safety reasons almost invariably obliged consumers to use this type of appliance in households. The agency did not explicitly delimit the geographical market. Nevertheless, from the factors analysed in the resolution it can be inferred that the latter was defined as the national territory.31

As regards substantial power, the agency determined that the market share of the offending company was around 80%. It stated that although a high market share was not synonymous with substantial power, this factor constituted a strong indication. There is no record in the resolution as to how the existence of that share was determined.

The offending company, for its part, contributed a study on cross-elasticity which calculated that the latter was high (2.37, 1.32, 1.66) in three important lines of products in the range of low-voltage electrical appliances. The Commission, by contrast, considered that the said elasticity was not so high as to inhibit significant market power for Bticino. CPC, using the Lerner Index, showed that although the company did not have dramatic substantial market power, a value of 0.75 was significant and revealing.32

Thus, based only on some of the factors the legislation establishes, CPC concluded that Bticino had substantial power in the relevant market.

31 The following were among the entry barriers analysed in the resolution: distortions in the exchange rate, transport costs, international prices and tariff protection for inputs. Likewise, CPC pointed out that even though there were hundreds of retail distributors (hardware stores), the distribution channels were significantly concentrated.

32 This index adopts values of between 0 and 1: The value 1 is associated with a strong, aggressive market power, and 0 with a widely competitive market.

Practices investigated33

Having proven the existence of the offending company's substantial power in the relevant market, the Commission analysed the practices investigated and their effects on the market. The practices investigated by CPC were the following: (i) exclusive distribution; (ii) imposition of resale prices; (iii) transaction subject to the condition of not purchasing goods normally offered from third parties; and (iv) predatory prices.

Exclusive distribution

The first practice CPC analysed was possible exclusive distribution on the part of Btcino to its distributors. To that end, the agency ordered a market study from the Trade Directorate of MEIC. The study revealed that Bticino conditioned the sale of its products to exclusivity or awarded prizes in exchange for this condition. Despite the above, the Commission had to reject the study since the officials who conducted it failed to comply with the formalities required by the legislation.

Furthermore, the witnesses called upon to appear in order to determine the merits of the study all denied selling only Bticino products and in any event pointed out that if they did so it was by their own volition and due to the greater demand for that brand's products. By virtue of this and of the lack of additional evidence of the existence of possible illegal distribution, any sanction for infringement of subparagraph (a) of Article 12 was ruled out.

Imposition of resale prices

Second, CPC assessed the possible imposition of resale prices by the offending company to its distributors. The main evidence found by the agency in analysing this practice was a communication that the sales manager sent to its distributors, backed and acknowledged by the company's general manager, in which distributors were asked to calculate consumer prices according to a particular formula. The note indicated that if the distributor did not comply, sanctions would be applied, ranging from a reprimand to suspending delivery of goods.

The agency was unable to gather suitable evidence of the effects of the conduct on the market.34 Despite the foregoing, the Commission considered that the note

33 CPC also analysed the possible commission of practices that could lead to the departure of competitors or to preventing their entry. In this regard, the agency's interpretation was that in order to consider these acts punishable, it had to be a question of express, aggressive and evident acts that were not provided for in the other subparagraphs of Article 12 and that exceeded any company's logical and normal interest in achieving a greater market share or defending its current share, with strategies that enabled it to face its competitors. The Commission did not find evidence to suggest that Btcino's intent was to push Sigma Dam out of the market, but rather considered that its strategies sought to maintain its market position, which had been threatened by the entry of a new competitor.

34 There was evidence in the file to the effect that the resale price was not uniform in the retail market. However, CPC discarded the evidence, since it was gathered several months after the complaint had been filed. The agency did not succeed in collecting evidence for the months between July and September, which were the months when the notes were sent to the distributors.

was clear, sufficient and direct evidence that the company under investigation had the intent of obliging its distributors to charge consumers a particular price, thus eliminating intra-brand competition. The agency therefore considered that a violation of subparagraph (b) of Article 12 of LPCDEC had indeed taken place on the part of the offending company.

Transaction subject to the condition of not purchasing goods normally offered from third parties

The third practice CPC investigated was the conditioned sale of Bticino products, so that distributors should not acquire the competition products. Regarding this practice, the agency obtained evidence supporting the fact that the offending company conditioned the continuance of commercial relations with a hardware store to its refraining from acquiring goods offered by Sigma Dam. To demonstrate this, it based itself on statements and documentary evidence in which the general manager of the company under investigation and the owner of one of the country's most important hardware stores stated that distribution of Bticino products was conditioned to that company's not acquiring Sigma Dam products.

On the infringement investigated, the Commission established an important legal principle. Specifically, it pointed out that it was not necessary for the conduct to be generalized in order for the illegality of this type of conduct to be determined; rather the existence of one such case was sufficient to establish the pertinent sanction. It thus stated that the conduct prevented its competitor from gaining access to this channel and consumers to a greater variety of products. For this reason, the Commission considered that there had been a breach of subparagraph (d) of Article 12 of LPCDEC.

Marketing of goods at prices below normal price (predatory prices)35

The agency found evidence that Bticino lowered its prices in the product lines just when its market position was being threatened by Sigma Dam's entry. Specifically, there was an overall change in price policy in which the price of some products went up, essentially imported products, and others went down.

Nonetheless, according to the evidence found by the agency, the reduction did not set the prices at below-cost levels. The profit margin, although significantly lower, was always higher than the producer's cost, including administration and marketing overheads. Also, no price was ever lower than its competitor's, Sigma Dam. CPC therefore considered that the price reductions did not violate the competition

35 To analyse this practice, CPC stated that predatory prices could be defined in two ways: by measuring the marginal cost (stricter method), or the unit cost (less-strict method). According to the first yardstick, the concern is that no agent should sell at a price lower than that which provides it with reasonable profits. However, the Commission considered that the difficulty in calculating the marginal cost has led to the use of the second concept and to the use rather of rules presumptive of that level of profits that is acceptable in a particular economy.

legislation, since although Btcino could lower its profits, if they were kept above the costs, this was legitimate.

Measures and Sanctions Applied

To impose sanctions, CPC took into account various features of the legislation. It considered as an aggravating factor the express and careless manner in which the circulars to bring the market under control were issued. In this regard, it pointed out that the indications of intentionality were particularly strong, since the resolution's chief evidence was documents acknowledged by the head of the company in which express reference was made to the practices that should be sanctioned. The infringer's market share was considered another important aggravating circumstance. As a mitigating factor, CPC took into account that this was the first investigation on relative monopolistic practices conducted in the country.

Moreover, the agency made a distinction between the two practices analysed. On the one hand, it considered as an aggravating factor in establishing the sanction the fact that the imposition of resale prices had been done in a generalized manner in the market. On the other, it considered as a mitigating factor the fact that the only evidence found of the practice of imposing conditions was in regard to one distributor of these products.

CPC therefore ordered Bticino, in accordance with subparagraph (a) of Article 25 of Law N° 7472, to discontinue measures aimed at establishing resale prices and conditioning the sale of its products. Likewise, it proceeded to establish a fine equivalent to 60 times the amount of a monthly minimum wage for imposing resale prices and 40 for imposing conditions to acquire products normally offered from third parties, within the limit of 410 minimum wages.

Comments on the Investigation

The electrical components case was an important step forward for CPC. The agency was successful in carrying out its first investigation on relative monopolistic practices. It established an important precedent in terms of evidence, which was subsequently used in the case of carbonated beverages and fruit juices. That is, suffice it to find one case or piece of evidence of some conduct in order to meet the definition of the legal concept established in Article 12 of LPCDEC. Nonetheless, the agency did not carry out a thorough analysis of the anti-competitive effects of the practices investigated, nor of the essential factors to determine substantial power in the relevant market.

Due to the complexity of this type of case, several years elapsed before the Commission succeeded in establishing another investigation on relative monopolistic practices that had a significant impact on the behaviour of economic agents in the market. It was not until the investigation on the carbonated beverages and fruit juices market that CPC was once again able to set standards of conduct for agents with market power.

2.2. The Case of the Carbonated Beverages and Fruit Juices Market (CPC-19–04)
Case Description

In May 2001 the companies PepsiCo Inc., Pepsi-Cola Interamericana de Guatemala, S.A. and Pepsi-Cola Manufacturing Company of Uruguay S.R.L. filed a complaint against The Coca-Cola Company and Embotelladora Panamco Tica, S.A. for considering that the latter companies carried out relative monopolistic practices. In June 2001, the company Embotelladora Centroamericana filed a complaint against Embotelladora Panamco Tica, S.A. for the same reasons.36

CPC decided to deal jointly with the above-mentioned complaints, opening an ordinary administrative proceeding against the companies Coca Cola Inter-American Corporation and Embotelladora Panamco Tica, S.A., considering that there were sufficient grounds in the administrative file for this purpose. The practices investigated were the following: imposition of resale prices, imposition of minimum purchase volumes, exclusivity in the use of refrigeration equipment, exclusivity in vending machines, product exclusivity in stores, tied sales and price discrimination.37

Investigation and Analysis of the Practice

CPC carried out an extensive investigation on the market. To that end, it requested information from the competitors on the market's characteristics, products marketed, sales, distribution systems and their relationship with clients and consumers, among other factors. Likewise, it requested information from clients in different distribution channels on their relations with beverage producing companies as regards: price policies and marketing and sales policies, among other topics. It also arranged for the parties to appear in order to present their pleadings.

Once the investigation was completed, CPC handed down its final resolution. In this context and in accordance with the LPCDEC, proving the existence of the illegal conducts involved analysing three elements: (i) substantial power in the relevant market on the part of the agent investigated; (ii) execution of the practices characterized

36 Subsequently, other beverage producing companies joined this proceeding. Thus, Refrescos La Mundial, S.A., La Cruz Blanca, S.A. and Fábrica de Refrescos y Sirope La Flor, S.A. requested to be considered as additional parties in the complaint filed. These requests were admitted by CPC in October 2001 and on October 16, 2001.

37 CPC exonerated Embotelladora Panamco Tica, S.A. from all responsibility for the commission of anti-competitive practices of tied sales and price discrimination set forth in subparagraphs (c) and (g) of Article 12 of LPCDEC, since no indication was found that they were carrying out such acts. The agency simply reiterated the prohibition on conditioning the sale of a product to other products from different markets and ratified that price discrimination was prohibited.

in Article 12 on the part of the offending company; and (iii) anti-competitive effects of the conducts.38 A summary of the features analysed is set forth below.

Relevant market

The first aspect the Commission defined was the product market, establishing it as that of non-alcoholic carbonated beverages and canned or bottled fruit juices. This definition gave rise to a major controversy, since the complainants claimed that the market should include carbonated beverages alone. Nevertheless, CPC decided that all the products contemplated could be substituted among themselves, since the technical studies showed a high cross-elasticity of demand. That is, the increase in the price of one product determined a displacement of the demand for same towards demand for the other product. Furthermore, the agency based its decision on various studies on the characteristics, uses and prices of the beverages, as well as their production processes.

The second factor analysed was the geographical market, which was defined as the entire national territory, since: (i) the marketing companies had a national distribution network; (ii) the companies did not differentiate their sales and marketing policies according to geographical areas; (iii) there were no restrictions to limit supply within the country; and (iv) the conducts denounced were carried out irrespective of the geographical area in which their clients operated.39

Substantial market power

To establish the substantial power of the agent investigated, CPC determined as a first factor that this company had a 74% share of the relevant market as defined. In addition, it analysed possible barriers to entry and in this regard pointed out that since beverages were goods acquired on impulse or in a routine manner, intensive distribution was required so as to be able to place the product in the largest number of sales outlets possible. The Commission considered this fact a barrier to entry.

Similarly, it considered the strength of the established brands and the advertising investment required to enter the market as an entry barrier. In this connection, the offending company had a wide-ranging portfolio of brands of carbonated and noncarbonated beverages, with products differentiated into different presentations and flavours, some of which were leading products in their categories.

Finally, CPC considered that there were no greater legal barriers, since taxes were applied to domestic and foreign beverages alike. Moreover, the customs tariff was 15%, which was not considered a significant obstacle to importing products.

38 LPCDEC lays down that it must be demonstrated that the practices have or may have the purpose or effect of unduly displacing other agents from the market, substantially hindering access to same or establishing exclusive advantages in favour of one or several persons.

39 It should be pointed out that the parties submitted various studies on different aspects of the relevant market, which to a certain extent made the agency's work easier, since in general it is difficult to obtain so much quality information on the market. However, it should be noted that the agency depended largely on the documents contributed by the parties in taking its decision, and not on its own investigation, which would have been too costly.

Once the agency had concluded that the company investigated had substantial power in the relevant market, it proceeded to analyse the practices denounced.40

Imposition of resale prices

The first conduct analysed was the imposition of consumer resale prices on the part of the offending company to its distributors. It was proven in the proceeding that Embotelladora Panamco Tica, S.A. established clauses in sales agreements with its distributors suggesting consumer sale prices, and in some of them, clauses were even found referring to the obligation of keeping to the price lists.

In addition, it was proven that the company distributed resale price lists in all the businesses with an ambiguous wording, which lent itself to doubting the latter's nature as a suggestion.41

Having analysed the evidence, CPC found the issuance of suggested resale prices permissible, as long as it was clear that these were not obligatory in nature. The rule laid down in this case was that a company with substantial power should make it expressly clear that the client did not have the obligation of following these lists.

Under these circumstances, the agency penalized the company investigated, considering that the anti-competitive effect of this practice was reducing intra-brand competition, i.e. the conduct reduced competition in products of the same brand among the different establishments and limited retailers' freedom to establish their sale prices. Thus, a fine of 410 minimum wages was established for violation of sub-paragraph (b) of LPCDEC. Furthermore, it ordered the company to discontinue the practice, eliminate the provisions of the contracts that obliged clients to keep to the lists, and change the wording of the same to indicate these were suggested prices and retailers were the ones to decide the price at which they should sell the products.

Minimum purchase volume

Second, CPC analysed the possible imposition of purchase volumes to distributors. Regarding this conduct, it was proven that the company Embotelladora Panamco Tica, S.A. had signed written contracts with different types of businesses, mainly schools and colleges, in which a series of benefits were offered in exchange for a minimum purchase volume. However, the Commission decided to absolve the company investigated of this charge, since in the establishments in which the minimum volume was demanded there was also an obligation of exclusivity, and it was therefore not possible to determine in themselves the anti-competitive effects of fixing minimum purchase volumes.

40 On establishing responsibilities for the practices investigated, CPC held Coca Cola Inter-American Corporation and Embotelladora Panamco Tica, S.A. responsible. However, it imposed sanctions and measures on the bottling company, as this company was the one that directly executed the illegal conducts.

41 In addition to the contractual clauses obliging distributors to follow the price lists, the Commission criticized the fact that the offending company's route delivery men affixed these lists directly in commercial establishments, chambers or cafés within sight of consumers.

The Costa Rican authority was clear in pointing out that demanding a minimum purchase volume was not always a prohibited practice. It was so only in cases in which the minimum volume became an exclusivity either in fact or in complicity or else when such volumes were fixed in irrational quantities that surpassed the real cost of the investment made in the establishment. It also pointed out that fixing a minimum volume could constitute an entry barrier if it was established within a contract having a very long term or conditioned to new premises that the client might establish in the future.

Exclusivity in the use of refrigeration equipment

Another of the conducts investigated was the establishment of exclusive use of refrigeration equipment. During the investigation it was determined that the offending company granted such equipment to its distributors for exclusive sale of its products. CPC considered that although it was understandable, being an investment of the company's, that use of the equipment by the competition's products should be restricted, for a company with substantial market power, imposing this condition in small establishments could restrict competition and reduce consumers' possibilities to choose.

That is to say, the rule the Commission established was that companies could request exclusivity in the use of their refrigeration equipment as long as this did not become an entry barrier to the market. Determining whether or not the equipment was an entry barrier depended on whether or not there was sufficient space on the premises to install additional equipment. In case of lack of space, the condition of exclusivity in the use of refrigeration equipment did constitute a barrier to entry for competitors.42

In keeping with the above reasoning, CPC determined that the practice could hinder the access of new participants. Consequently, the company Embotelladora Panamco Tica, S.A. was sanctioned by ordering it to abstain from imposing, including or negotiating in any manner, exclusivity clauses with its clients with respect to refrigeration equipment in locales where there was no space to set up another refrigerator for products considered part of the relevant market.43

Product exclusivity in stores

The Commission also investigated the establishment of exclusivity agreements by the offending company. It was determined in the proceeding that Embotelladora Panamco Tica, S.A. had indeed established sales agreements in some marketing

42 CPC established a similar principle for the case of vending machines. Basically it pointed out that if the machine was owned by the trader, the offending company could not condition the type of products he should sell. Furthermore, it stated that if the machine belonged to the bottling company (and its use was provided to the client free of charge), exclusivity could be established only in cases in which there was a material impossibility of sharing the machine, whether for reasons of space or other duly proven ones. In either case, there could not be a contractual limitation to hinder the client's free decision.

43 The criticism levelled at this parameter established by the Commission was that it could give rise to some uncertainty among traders, since in some establishments it could be difficult to determine whether or not there was enough space for additional equipment.

channels, which included an exclusivity or preference44 clause for the business in the sale of the said company's beverages. The Commission considered that this type of actions reduced competition and led to the displacement of competitors while hindering the entry of new participants. It therefore considered that Embotelladora Panamco Tica, S.A. violated subparagraph (a) of Article 12 and the company was thus sanctioned with a fine. This fine was 410 minimum wages. Moreover, the agency ordered it to eliminate this type of clauses from current contracts, as well as to abstain from including them in future ones.

Although CPC had indicated that exclusivity admitted reasons of efficiency on some occasions, for this case it considered that in view of the offending company's market power, no product exclusivity contract was acceptable. In its opinion, when a company had that degree of power, no possible benefit it could offer justified a total block on the competition's products.

Comments on the Investigation

The jurisprudence established by CPC in the resolution on the carbonated beverages and fruit juices market set the guidelines on the manner in which companies having market power should behave. In this regard, the resolution laid down the notion for the first time in the country that the conduct's negative impact is related to the level of power of the agent investigated. The greater the market power, the worse the effects of a relative monopolistic practice and therefore the more the company's actions should be restricted.

The Commission also established as a principle that among all the possible alternatives to attain a commercial end, companies should choose the one that is least restrictive to competition. According to the agency, conducts are acceptable as long as they are carried out within a framework that allows the participation of different companies in the market, and when the restrictions for the benefit provided are reasonable. Finally, it should be pointed out that the agency's resolution was also criticized, specifically, for the time it took CPC to resolve the case45 and because the amounts of the sanctions were considered too low.

3. Conclusions and Recommendations

As in the previous section, having analysed competition legislation in the area of relative practices, as well as the most representative cases, we now go on to present the most significant conclusions of the previous sections in the light of best

44 In this context, "exclusivity" and "preference" agreements were considered equivalents, given that the effects of the contract were the same.

45 This was basically due to the fact that during the proceeding the parties lodged various amparo proceedings before the Constitutional Court, which to a certain extent delayed resolution of the case. In addition, the file was sizable and contained a large amount of confidential information, which required many UTA resources to properly appraise the documents that comprised it.

international practices. Specifically, reference will be made to the analysis of prohibited conducts, the investigation instruments used, the imposed remedies and sanctions, the participation of state institutions in the market and the judicial background in the field.

3.1. Analysis of the Legality of Conducts

Costa Rican legislation formulates a sort of rule of reason which by its very nature should be as flexible as possible in order to adapt it to each case. However, Costa Rica's legalist tradition, especially important in the area of sanctioning, tends to favour legal certainty over other values. For example, relative monopolistic practices are characterized previously, which provides considerable weight to proving a particular conduct or not.

Along these lines, one of the most recent criticisms of the European system is that Article 82 of the Treaty of Rome places greater emphasis on characterization of the practice than on assessment of its effects. Some experts have pointed out that ex ante characterization runs the risk of leaving some conducts without sanctions. In other words, economic agents could carry out a conduct not expressly provided for, with the same anti-competitive effects as the practice characterized (EAGCP, 2005).

In view of the above, rather than taking confirmation of the existence of a particular conduct as a point of departure, the competition agency should begin by detecting the anti-competitive effects. In point of fact, one same practice, depending on the environment and circumstances in which it is executed, can have pro-competitive or anti-competitive effects. If the agency concentrated on detecting the effects of the conducts, case analysis would be more consistent, since practices with similar effects would be treated in the same manner (EAGPC, 2005).

Great care must be taken in bringing these ideas into the context of the Central American countries. This discussion arises in countries with a considerably more developed culture of competition than in our own countries. That is, it would be premature to change the rules of the game in the Central American region, where economic agents are beginning to understand the concept of abuse of dominant position and the competition agencies are starting to deal with their first cases. Nonetheless, CPC and the other authorities of the area should not loose sight of this discussion, so that in analysing a practice of abuse of dominant position, the issue that should determine its illegality should be proving the anti-competitive effects.

3.2. Investigation Instruments and International Cooperation

As in the case of absolute monopolistic practices, in carrying out investigations on relative practices it is essential to rely on relevant information. However, in these cases the problem is of a different nature and is related to the fact that most of the information depends on a single economic agent who generally holds a significant market share. The main consequence of this situation lies in the fact that for the agency it is extremely difficult to compare the information it gathers in order to determine its truthfulness. On the one hand, because normally there is no other similar agent, and on the other because it stands to reason that the agent investigated has no incentive to submit the information.

Furthermore, the other economic agents normally depend on the agent with substantial power. For instance, a distributor of a particular product normally wishes to have a good relationship with its supplier, especially if the latter is dominant in the market. Hence, with a few exceptions, since it could signal the end of the business relations, a distributor is unlikely to file a complaint for relative monopolistic practices against its supplier, and nor will it collaborate in an investigation against it. Moreover, it should be pointed out that the investigated agents are frequently multinational companies whose strategic decisions on sales and marketing policies are taken at their head offices in other countries. This means that the information may not even be available within the country.

It is therefore essential for CPC to increase its investigation powers so that in addition to requesting public or confidential information, it can carry out searches and attachment of documents. It is also important for the agency to promote international cooperation with neighbouring countries, as well as with strategic countries, so that it can gain access to information outside Costa Rica.

3.3. Remedies and Sanctions

There are various types of remedies or sanctions with regard to practices of abuse of dominant position that can be applied to economic agents that commit this type of offence. In order to define the most appropriate remedy or sanction in a particular case, OECD points out that the competition agency must be very clear about the goal it wishes to meet and the consequences that the instrument chosen could bring about.

For example, one of the agency's objectives could be to ensure that the market in which the offence was committed comes as close as possible to perfect competition. However, if economies of scale are generated by the industry, fulfilling this objective could diminish productive efficiency in the market, which over the long term could reduce incentives to innovate. The agency could also promote the elimination of entry barriers. For example, a sector's tariffs can be eliminated with the aim of diminishing the firm's dominance through international competition. However, this type of remedy implies costs for the government (OECD, 1996).

In addition, some remedies seek to eliminate or alter an abusive conduct in the market. But this type of measures gives rise to costs for the agencies, since they have to use resources to confirm that the orders issued to economic agents are complied with. Another available remedy is the establishment of measures to change the market structure. Some experts believe that this type of instruments should be used cautiously, as once the remedy has been executed it is difficult to revert (OECD, 1996).

In any event, the important thing is for the agencies to have the widest possible range of options to discourage this kind of conducts, or else to correct their negative effects. That is, in cases of practices involving abuse of dominant position, in general the application of a simple fine is not sufficient to eliminate or correct a particular situation.46

3.4. State Institutions

There are still various state-owned enterprises in Costa Rica that have a dominant position assigned by Law. Like any other economic agent, state enterprises are exposed to committing relative monopolistic practices. In most cases, however, they are partially or fully exempt from the application of competition legislation. For example, the Costa Rican Electricity Institute is totally excluded from antimonopoly regulation.47 Another case is the National Liquor Factory, which has an absolute monopoly on alcohol distilling, but can grant concessions in the preparation of liquors and competes freely in marketing a finished product. In practice this means that in the relations with its competitors the competition legislation can be applied.48

Although one can argue that all monopolies granted by Law should disappear and allow competition in the market, this issue involves political and other kinds of considerations that are beyond the scope of this paper. Nevertheless, even if these monopolies were kept in place, at least the LPCDEC should be applied to these enterprises in order to prevent conducts that may affect the process of unrestricted competition.

3.5. The Judiciary

As in the issue of horizontal practices, training the Judiciary in the area of practices involving abuse of dominant position is essential. In fact, the analysis of such practices is generally more complex, since it requires the use of economic instruments to determine the legality or illegality of conducts.

At present, no final resolution is known in this matter on the part of the Judiciary, and it is therefore not possible to technically assess its work. However, there are various cases of which it will take cognizance in the near future. For this reason, it is important to emphasize the point that training sources should be placed at its disposal. In this regard, it should be borne in mind that the Judiciary is not only an active part of the regulatory process of competition, but also the final authority available in resolution

46 In this context, in the case of carbonated beverages and juices the measures that obliged the company to change its commercial practices were much more significant than the fines imposed.

47 In this regard, see Article 9 of LPCDEC.

48 In 2004 and 2005 two investigations were opened against the National Liquor Factory. See CPC files D-14–04 and IO-01–06.

of the cases referred to it. Therefore, the Courts play a vital, long-term role in defining the criteria and interpretations of competition legislation.

Concentrations

In addition to absolute and relative monopolistic practices, company concentrations are also a usual object of regulation by competition laws and Costa Rica is no exception. For the purposes of this study, the issue will be analysed in line with the same methodology as the previous sections. Initially, the applicable legislation and the Commission's experience will be reviewed, and subsequently an analysis of same will be made in the light of best practices in the matter.

1. Regulatory Provisions

In Costa Rica, concentrations are regulated by Article 16 of LPCDEC. In accordance with these provisions, a concentration is:

the merger, acquisition of control or any other act by virtue of which corporations, associations, shares, capital stock, trust funds or assets in general, carried out between competitors, suppliers, clients or other economic agents, with the purpose or effect of diminishing, harming or hindering unrestricted competition, with respect to the same, similar or substantially related goods or services.

Thus, if we analyse the above text literally, we could conclude that Costa Rican legislation would appear not to distinguish between concentrations that are harmful to the process of competition and those that are not. Rather, operations described as concentrations are those that have an anti-competitive purpose or effect. Article 39 of the Regulations of the said Law would seem to be drafted in much the same manner, since it states that a sanction will be imposed in cases where the existence of a concentration is determined.

Article 28 of the LPCDEC, which addresses the issue of sanctions, sets forth that CPC may order total or partial de-concentration "when there has been an undue concentration." Similarly, this Article sets a fine of 410 minimum wages for those who incur "in any of the concentrations prohibited in this Law." This wording seems to imply that there are concentrations that are not prohibited by Law. This second interpretation, as we shall see, is the one the agency has followed in its jurisprudence.

For an analysis of such practices, the aforementioned Article 16 refers to the same criteria for measuring market power as those used to assess relative monopolistic practices. LPCDEC definitively prohibits only concentrations with an anti-competitive purpose or effect. Thus, for a concentration to be liable to a sanction, it must be demonstrated that it tends to create or maintain a position of substantial power in the relevant market, and that it has an anti-competitive purpose or effect.

By contrast to much legislation in other countries, LPCDEC does not establish a prior notification procedure by companies for concentration operations. CPC's powers are limited to the application of a posteriori measures. The main limitation of this path is that it is often practically impossible to de-concentrate a merger that has already taken place, except at a high cost, giving rise to even greater inefficiencies in the competition process.

By way of exception, the Law on Workers' Protection N° 7983 of August 20, 2001, lays down a prior authorization procedure for concentrations between pensions operators. Such authorization is granted by the Superintendence of Pensions, which has the obligation to request CPC's opinion. The purpose of this provision, according to the text of the Law is to ensure that "the merger process does not harm the interests of beneficiaries or levels of competition."

Finally, regarding the investigation, LPCDEC does not differentiate between the proceeding to investigate concentrations and the one to be followed when investigating monopolistic practices. This also gives rise to major limitations, since the ordinary administrative proceeding is designed as an eminently punitive process whose objectives frequently differ from those pursued by investigations on concentrations.

2. Relevant Resolutions

Few cases of concentrations have been pursued by CPC, probably due to the limitations of the regulatory framework. Below we will comment on two cases that show the system's advantages and weaknesses. First of all we will analyse the merger between Kimberly Clark Costa Rica, S.A. and Scott Paper de Costa Rica, S.A., which was the first concentration dealt with by the competition agency. Second, we will analyse the concentration of the pensions operators Banex and Interfin, which was one of the few cases in which prior notification was submitted.

2.1. Merger Between Kimberly Clark Costa Rica, S.A. and Scott Paper De Costa Rica, S.A. (CPC-05–96)
Case Description

The investigation was conducted as a result of a complaint filed in February, 1996, by Procter and Gamble Interamericas de Costa Rica, S.A. The complaint accused the companies Kimberly Clark Costa Rica, S.A. and Scott Paper de Costa Rica, S.A. of entering into a concentration prohibited by LPCDEC, and therefore of infringing Article 16 of same.

The complaint was grounded on the possible negative effects on the structure of competition in the market for paper products for domestic use in Costa Rica. Consequently, the complainant requested CPC to take the necessary action to prevent the above-mentioned adverse effects.

Investigation and Analysis of the Concentration

As a result of the complaint a preliminary investigation was carried out, and by means of a decision on record in the Minutes of Regular Session N° 20 of May 1996, the opening of the ordinary proceeding to investigate the actions denounced was ordered.

During the proceeding the following evidential elements were collected: commercial information requested from different companies participating in the market; official customs information on imports and exports of paper-associated products and on the evolution of tariffs on same; official MEIC information on the evolution of paper product prices; and information requested from the companies under investigation on prices, sales volumes and other relevant company information. Also, two appearances were convened to furnish proof, among which certain competitors of the offending companies were summoned, among others.

In the first instance, the agency analysed the concept of concentration according to LPCDEC. Here, the Commission was of the opinion that the effects or consequences of the conduct, rather than its intention, should be considered. In this regard, it pointed out that the mere effect of diminishing competition, irrespective of the spirit of the companies that were concentrating, made the operation subject to the suspension, correction or suppression of the concentration.49

The product's relevant market was defined as including four products derived from tissue paper: toilet paper, paper napkins, kitchen towels and facial tissues. All these products were being marketed by both companies prior to the merger. Costa Rica was defined as the geographical market. This in terms of transport costs, tariffs and normal strategies of companies participating in other countries of the area, which are usually concentrated in local markets.

The offending companies' defence is centred mainly on making light of their market power. They also argued that the merger was between companies of foreign origin and that, judging by their previous behaviour, there should be no fear that in the future the market concentration should result in aggressive behaviour with regard to prices or marketing conditions. Another argument put forward by these companies was that the merger did not involve pooling productive assets but only administrative structures, and that the complainant had no investments in Costa Rica.

These arguments led CPC to assert that in accordance with Article 70 of LPCDEC, although there had been no concentration of assets or merger at the statutory and shares levels, the fact that the two firms would respond to a single economic interest group made it possible to consider them a single unit with regard to business activity. In order to conclude that the companies, even though legally being separate entities, operated as a single firm, recourse was taken to various elements of fact, such as the use of common stationery, single price lists, use of a single administrative head office and a single warehouse, among others.

49 In this connection, see CPC-05–96, Preambular Paragraph II.

Particular circumstances in Costa Rica would likely lead one to suppose negative effects in this market, since the merger would have led to a concentration of almost 100%. To calculate the concentration that would be reached the Hirschman-Herfindhal index (HHI) was applied. The result of applying this index gave a result very close to 1.0 (maximum concentration value). In other words, once the merger took place, competition in the market would disappear.

Added to the above was the analysis of different entry barriers identified by the Commission, such as the level of investment required, the positioning of brands in the market, the tariffs on imported products from extra-regional countries (both merging companies had subsidiaries at the Central American level).

The conclusion was therefore reached that the merger would cause harmful effects on the market, since the merged company would enjoy a highly dominant position in a market with few threats from new competitors, as a result of which it would be difficult to restore competition in the market in the near future. The agency therefore considered it necessary to take measures to create a level of competition similar to the one that existed prior to the merger.50

Measures Taken

In its resolution, CPC agreed to request the companies in the process of merging to present, within 30 working days, an action plan to revert the negative effects of the merger on the market for toilet paper, paper napkins, kitchen towels and facial tissues. Subsequently, within the following 30 working days, the Commission could coordinate the mechanisms it deemed appropriate to reach agreement with the offending companies on the measures to be applied. Should an agreement not be reached or the plan not delivered within the established time frames, CPC would impose the measures deemed pertinent to eliminate the effects of the concentration.

The plan submitted by the companies consisted in the sale of certain brands to third parties and was accepted by the Commission. However, there is no record in the files that the conditions imposed were actually complied with in the end.

Comments on the Investigation

This case was extremely important for CPC, since it clearly highlighted the weaknesses of the system of regulation of concentrations, among which the following can be cited: lack of a procedure for prior notification, non-existence of appropriate conditions and remedies to approve or reject a concentration and lack of clear criteria for in-depth analysis of operations. These weaknesses have been reflected in the small number of concentrations in which the Costa Rican authority has intervened.

In spite of the weaknesses detected, the second case, on the pensions market, pointed up the advantages of making an ex ante analysis of concentrations. It was

50 In this context, see CPC-05–96. Preambular paragraph XXII.

also demonstrated that it is possible to establish relations of coordination with the regulatory bodies in this field, as will be seen in the next section.

2.2. Merger of Pensions Operators Banex and Interfin (CPC-14–01)
Case Description

This investigation was initiated in April 2001, as a result of steps taken by the Superintendent of Pensions, who requested CPC's opinion on the merger of pensions operators Banex and Interfin. The consultation was made under Article 47 of the above-mentioned Law for Workers' Protection, which, as we saw, establishes a hypothesis of exception whereby prior authorization is required for pensions operators intending to carry out mergers.

Investigation and Analysis of the Concentration

As a result of the consultation, UTA drafted a report which CPC used as the basis for its decision. Below we detail the most significant elements analysed by the Costa Rican authority.

Proof furnished

The evidence used in this case consisted of the documentation provided by the companies that merged. Likewise, information was available on the rest of the companies in this market, provided by the Superintendent of Pensions.

CPC concluded first, that pensions operators Banex and Interfin were competing agents in the market for labour capitalization administration funds and for funds for the complementary pensions system. Second, that neither of these two companies had substantial power in the relevant market and the merger would not lead to an acquisition. Third, despite the fact that at the time of resolution the relevant market was highly concentrated, where one or two companies (of a total of ten) were considerably larger than the rest, the merger of two small or medium firms could well translate into benefits for competition. The foregoing because it would tend to balance the agents' market share.

Measures Taken

CPC determined that the merger between pensions operators Banex and Interfin would not have an anti-competitive effect in the market for Labour Capitalization Administration Funds and funds for the Complementary Pensions System, thus positively resolving the consultation.

Comments on the Investigation

Had the concentration posed problems in terms of competition, the regulatory framework would not have been sufficient, especially because appropriate conditions and remedies do not exist. Hence, the necessary amendments to the legislation should not only include a prior notification procedure, but also the characterization of suitable measures to be imposed on concentrations presenting some kind of limitation on competition.

On the other hand, it is important to reiterate the fact that this was one of the few cases in which a sectoral body requested CPC's opinion on matters of concentration. The proceeding showed that it was possible to establish good coordination in which both regulatory organs respected one another's spheres of competence. As in this market, this scheme could repeat itself in other financial markets, or even in public sectors.

3. Conclusions and Recommendations

As in the previous sections, having analysed competition legislation in the area of concentrations, as well as the most representative cases, we now go on to contrast the most significant conclusions of the previous sections with best international practices. In particular, reference will be made to the procedure for dealing with these formalities, to prior notification as an investigation tool, to the method for analysing cases, and to available corrective conditions and measures.

3.1. The Proceeding

The proceeding provided for by the competition legislation does not respond to the particular needs of concentration cases. In this regard, the scarce existing jurisprudence has clearly shown that the proceeding laid out in the General Law of the Public Administration, N° 6227 of May 2, 1978, is inadequate for approving, rejecting or conditioning a particular concentration operation.

For this reason, an authorization proceeding and not a sanctioning proceeding should be designed. Specifically, this proceeding should take the following into account: (i) the role of the parties (this should be different from a typical sanctioning proceeding); (ii) treatment of confidential information (above all, information related to the strategic elements of the operation); and (iii) resolution time frames (for an authorization request, the proceeding should be as expeditious as possible).

3.2. Prior Notification

Costa Rican regulations place serious limitations on instruments available for the investigation of concentrations. LPCDEC's principal shortcoming is the absence of a mechanism for prior notification by the companies involved in the operation, a mechanism considered the main tool available to competition agencies at international level (ICN, 2005b). Its absence therefore highlights the inadequacy of Costa Rican Law in dealing with this type of operations. This has been reflected in the absence of cases processed by the agency, a similar shortcoming as the one that formerly existed in the European Union prior to the passing of specific regulations on the matter (Goyder, 2000/2001).

One basic principle that should be followed by an appropriate prior notification procedure is flexibility in the definition of notification thresholds and requirements. Here, the aim should be to ensure that only relevant operations are submitted to the verification procedure. That is, CPC should invest its (usually scarce) resources in concentrations likely to cause an appreciable effect on the market so as to avoid incurring unnecessary transaction costs on economic agents and superfluous use of resources by the Commission. Such thresholds and requirements should also comply with the following criteria: (i) clear and understandable; (ii) based on objectively quantifiable criteria; (iii) established on the basis of readily available information; and (iv) laying down a reasonable period of time for the notification (ICN, 2002, 2005a, b).

3.3. The "Two-Stage" Method

Proper control of concentrations uses up the majority of competition agencies' resources. For this reason it is important to establish a proceeding that makes it as easy as possible to review these operations. This is especially important for small agencies with very limited resources, as in CPC's case.

In this regard, the European Union and other agencies in the world have established a proceeding commonly known as the "two-stage method" which makes possible a relatively expeditious analysis of concentrations. This method or proceeding is grounded on the fact that most concentrations do not pose any problems for competition. Thus, the concentrations notified are submitted to a preliminary examination carried out for a short period of time, submitting only concentration operations that do pose problems for competition to longer review periods.

Under this system, however, the operation notified may not be carried out until the applicable waiting or suspension period has elapsed. Therefore, suspension periods should be subject to definitive and easily calculable time frames so as to allow operations that pose no problems for competition to be analysed as expeditiously as possible. Likewise, if the operation goes on to a second stage, the time frames and dates should be well defined (ICN, 2005b). In short, the proceeding should not only be expeditious, but also create the greatest certainty for the interested party.

3.4. Criteria for Analysis of Concentrations and Corrective Measures

Generally speaking, mechanisms for the control of concentrations seek to prevent economic agents from securing or strengthening a dominant position in the market by acquiring control of competing companies (current or potential). To that end, the premise should be that the market's structure will affect the conduct of the economic agents that form part of it (Conrath, 1995).

Thus, it is normal for investigations related to concentrations to use tools that differ from investigations of monopolistic practices. This difference is absent in LPCDEC, which treats regulation of concentrations on a par with relative monopolistic practices. This is reflected not only in the absence of criteria to analyse this type of operations, but also in the lack of measures or remedies for solving the problems to which these give rise in the markets.

In this regard, the Costa Rican Law does not clearly lay down the criteria and specific technical instruments to be used at the time of approving a concentration or not. For example, it makes no reference of any kind to the possibility of using indexes to measure the degree of concentration of a particular operation. Nor does the legislation provide for the measures or remedies that the competition authority can impose in order to condition concentrations. Such remedies can include the sale or licensing of brands, the sale of assets, prohibition on participating in certain markets for a set time, the horizontal or vertical division of the company, among others. In any event, these mechanisms should not only be as wide-ranging as possible in order to be able to adapt to the circumstances of each particular case, but should be duly characterized in the legislation to ensure legal certainty for the economic agents.

3.5. Coordination with the Regulated Sectors

Coordination between the competition agency and regulatory bodies is a key aspect where concentrations are concerned. In this context, the pensions case reflected good teamwork between the two authorities. However, this was an exceptional case in Costa Rica. The rest of the regulated markets do not have this type of coordination. For example, under current legislation if a concentration takes place in the banking market, both the Superintendence of Financial Entities and CPC could take cognizance of the concentration, but the bank authority would do so ex ante, whereas the agency would exercise ex post control.

It is therefore essential to amend all the regulations that cause such conflicts with regard to competition. Furthermore, CPC should intensify its competition law activities in the regulated sectors in order to facilitate coordination between authorities. In this connection, a practice recommended by ICN is to establish such coordination with the aim of effectively applying the legislation on concentrations and obtaining consistent or at least non-contradictory results in the jurisdictions involved, as well as reducing overlapping and unnecessary burdens for both the parties and the authorities (ICN, 2002)–this regardless of which authority hands down the final decision. On this issue there is no uniform yardstick on whether it should be the competition agency or the sectoral body that determines whether or not a particular concentration is approved. This will depend on the type of regulated sector. In any case, the important thing is that one of the determining factors should be the criteria with regard to competition.

II. Conclusions

The fundamental purpose of this paper was to appraise the Costa Rican competition legislation and its application by CPC. To that end, we presented a detailed analysis of the agency's work in the investigation of horizontal and vertical practices and anti-competitive concentrations. For purposes of the study, the legislation applicable to the field was described, as well as the most representative cases, possible limitations faced by the agency, and some recommendations were suggested, in line with best international practices.

However, the experience with regard to cartels can be divided into two stages that may well be repeated in other countries beginning to apply competition legislation. During the first phase, CPC's investigations were relatively simple. This was on account of the fact that in most of the investigations the agency succeeded in obtaining direct evidence of the conducts being carried out, and in some cases the evidence was even of a public nature. To a large extent this was due to the investigated persons' ignorance of competition legislation. This undoubtedly facilitated the agency's work, since in general the main difficulty in this type of cases is finding the evidence to prove that a cartel has been formed, as these can be set up without a trace.

However, as economic agents gradually broaden their knowledge of competition legislation and CPC applies stiffer sanctions, this reality is likely to change. Economic agents will tend to act in a less conspicuous manner to carry out prohibited agreements. This means that investigations will become more complex and some agreements will probably remain undetected. In this context, the airlines case somehow shows that the agency has entered a second phase regarding investigation of cartels. In this case the evidence obtained was of an indirect nature and the investigation went much deeper in order to determine whether or not the practice had been committed.

This implies that the competition legislation should be amended so as to reinforce investigation tools and establish a leniency programme that facilitates detection of this type of cases. In parallel, CPC should apply a policy of severer sanctions with the aim of discouraging economic agents from forming cartels. Otherwise, as time goes by the current tools will be inadequate for investigating this type of conduct.

In regard to relative practices, the competition agency's experience has been considerably less than in the sphere of cartels. This has largely been due to the difficulty in setting up this kind of investigations. Nonetheless, the cases dealt with have highlighted CPC's good technical capability to analyse the economic and legal factors involved in such investigations.

The difficulties have to do more with the rigidity of the legal framework in the analysis of these types of practices. First of all, LPCDEC characterizes all the possible practices beforehand; second, the rule of reason fails to clearly lay down the possibility of weighing up anti-competitive versus pro-competitive effects. However, the authority solved this last point in analysing cases by accepting justification of efficiencies.

The above notwithstanding, it continues to be a legalist system that needs to characterize beforehand the anti-competitive conducts and effects, which runs counter to the very nature of the rule of reason. In this context there has been an important discussion in the European Union on the need to focus on analysis of the effects rather than on the commission of a particular conduct. Thus, instead of taking as the point of departure confirming the relative practice, the competition authority should begin by assessing the anti-competitive effects.

This discussion, however, arises in countries with a more developed culture of competition than our own countries. That is, it would be premature to change the rules of the game when economic agents are beginning to understand the rules of competition and the authorities are starting to investigate their first cases. Nonetheless, agencies should not lose sight of this discussion, so that in analysing a relative practice, the issue that should determine its illegality should be confirming the anti-competitive effects.

Finally, in the area of concentrations the fundamental problem is that the legislation fails to adequately regulate this type of operations. Here, regulation of concentrations is practically the same as for relative practices. This could be inadvisable in view of the differences between measures which by their very nature concentrate on market structure and not on the conduct of the economic agents themselves.

Specifically, the lack of an authorization procedure, non-existence of the mechanism for prior notification, and absence of appropriate remedies and sanctions have limited the authority's investigation of anti-competitive concentrations. There can be no doubt that these factors are essential for regulation in this field to work properly. Hence, amending these aspects in the legislation is a pressing need.

Bibliography

Comisión para Promover la Competencia (CPC) (1995a), Artículo cuarto del acta de la sesión ordinaria 09–95 del 28 de noviembre de 1995, Costa Rica.

Comisión para Promover la Competencia (CPC) (1995b), Artículo quinto del acta de la sesión ordinaria 09–95 del 28 de noviembre de 1995, Costa Rica.

Comisión para Promover la Competencia (CPC) (1996a), Artículo cuarto del acta de la sesión ordinaria 39–96 del 19 de noviembre de 1996, Costa Rica.

Comisión para Promover la Competencia (CPC) (1996b), Resolución de las 19:00 horas del 15 de octubre de 1996, File 05–96, Costa Rica.

Comisión para Promover la Competencia (CPC) (1999), Artículo quinto del acta de la sesión ordinaria 09–99 del 15 de junio de 1999, Costa Rica.

Comisión para Promover la Competencia (CPC) (2000), Artículo sexto del acta de la sesión ordinaria 32–00 del 12 de septiembre de 2000, Costa Rica.

Comisión para Promover la Competencia (CPC) (2001a), Artículo tercero del acta de la sesión ordinaria 14–01 del 2 de mayo del 2001, Costa Rica.

Comisión para Promover la Competencia (CPC) (2001b), Artículo noveno del acta de la sesión ordinaria 39–01 del 13 de noviembre del 2001, Costa Rica.

Comisión para Promover la Competencia (CPC) (2002a), Seminar: El rol de la política de competencia en la protección del consumidor y en el fortalecimiento de la competitividad de las pequeñas y medianas empresas, Boletín de Competencia, year 5, N° 51, San Jose, Costa Rica, September.

Comisión para Promover la Competencia (CPC) (2002b), Artículo quinto del acta de la sesión ordinaria 32–02 del 05 de noviembre del 2002, Costa Rica.

Comisión para Promover la Competencia (CPC) (2002c), Seminar: El rol de la política de competencia en la protección del consumidor y en el fortalecimiento de la competitividad de las pequeñas y mediana empresas, Boletín de Competencia, año 5, N° 51, San José, Costa Rica, septiembre.

Comisión para Promover la Competencia (CPC) (2003), Artículo quinto del acta de la sesión ordinaria 37–03 del 02 de diciembre del 2003, Costa Rica.

Comisión para Promover la Competencia (CPC) (2005), Artículo sexto del acta de la sesión ordinaria 19–04 del 25 de mayo del 2005, Costa Rica.

Conrath, C. (1995), Practical Handbook of Antimonopoly Law Enforcement for an Economy in Transition, Washington, DC.

Curiel, C. L. (2000), Diseño de un modelo de abogacía de la competencia en el contexto de la implantación de las políticas de competencia en América Latina, United Nations Conference on Trade and Development (UNCTAD), New Cork and Geneva.

General Law for Public Administration N° 6227 (General de la Administración Pública, N° 6227) (1978), May 2, Costa Rica.

Goyder, D. (2000/2001), Merger Regulation, online course, Graduate diploma: EC Competition Law, Module 6, King's College, UK.

Grupo de Asesores Económicos para la Política de la Competencia (EAGCP) (2005), An economic approach to Article 82. [online] (http://www.shh.fi/stenback/eagcp_82_2005_july_21.pdf).

International Competition Network (ICN) (2002), Recommended Practices for Merger Notification Procedures, September. [online] (http://www.internationalcompetitionnetwork.org/media/library/conference_1st_naples_2002/speech_gdz.pdf).

International Competition Network (ICN) (2005a), Defining Hard Core Cartel Conduct. Effective

Institutions. Effective Penalties. Bonn, Germany, June. [online] (http://www.international-competitionnetwork.org/media/library/conference_4th_bonn_2005/Effective_Anti-Cartel_Regimes_Building_Blocks.pdf).

International Competition Network (ICN) (2005b), ICN Investigative Techniques Handbook for Merger Review. [online] (http://www.internationalcompetitionnetwork.org/media/library/conference_4th_bonn_2005/Investigative_Techniques_Handbook.pdf).

International Competition Network (ICN) (2006), Enforcement Techniques. International Competition Network. Anti-Cartel Enforcement Manual, ICN, April. [online] (http://www.internationalcompetitionnetwork.org/media/library/conference_5th_capetown_2006/DigitalEvidenceGathering.pdf).

Law for Cooperative Association N° 5185 (Ley de Asociaciones Cooperativas N° 5185) (1982), February 1973, modified by the law N° 6756, May 7, 1982 and its reforms, Costa Rica.

Law for the creation of the National Rice Corporation N° 8285 (Ley para la creación de la Corporación Arrocera Nacional N° 8285 N° 8285) (2002), Costa Rica, May 30.

Law of the Regulatory Authority of Public Services N° 7593 (Ley de la Autoridad Reguladora de los Servicios Públicos N° 7593) (1996), August 9, Costa Rica.

Law to Promote Competition and the Effective Defense of the Consumer N° 7472 (Ley de Promoción de Competencia y Defensa Efectiva del Consumidor N° 7472) (1994), December 19, Costa Rica.

Law for the Protection of Workers N° 7983 (Ley de Protección al Trabajador N° 7983) (2001), August 20, Costa Rica.

López, R. (1995), La Ley Federal de Competencia Económica y algunas consideraciones desde la perspectiva del derecho internacional, Revista Jurídica, N° 2, Anuario del Departamento de Derecho, Universidad Iberoamericana, Mexico.

OECD (1996), Abuse of Dominant Position and Monopolization [online] (http://www.oecd.org/dataoecd/0/61/2379408.pdf).

OECD (1998), "Recommendations of the Council concerning effective action against hard core cartels" (http://www.oecd.org/dataoecd/39/4/2350130.pdf)

OECD (2001), Report on Leniency Programs to Fight Hard Core Cartels [online] (http://www.oecd.org/dataoecd/49/16/2474442.pdf).

Organic Law of Sugar Cane Agriculture and Industry N° 7818 (Ley Orgánica de la Agricultura e Industria de la Caña, N° 7818) (1998), September 2, Costa Rica.

This page intentionally left blank

Chapter 6
Banking Competition in Mexico

Marcos Avalos and Fausto Hernández Trillo

Introduction

One of the most widely debated sectors in terms of economic competition is the financial system – particularly banks. Although worldwide trends in this sector have been moving towards concentration, the available studies conclude that this is not impairing competition.

Such trends can be seen in Mexico, where the banking sector has tended towards apparent concentration resulting from mergers between financial intermediaries, with emphasis on national institutions being absorbed by foreigners. The country, however, has failed to enjoy the normal benefits of efficient financial intermediation. On the one hand, although the interest rate spread (main price in the sector) has diminished in recent years, it remains high by international standards; on the other, overall sector penetration as a proportion of gross domestic product (GDP)–30% in 2002–is still the lowest among the Organisation for Economic Co-operation and Development (OECD) countries.

The literature seeking to explain these circumstances abounds,1 including studies by multilateral agencies.2 Among oft-cited arguments is low sectoral competition levels, and it is this aspect our study aims to review.

Competition in the financial sector is not an easy subject to tackle, as a number of approaches can be taken from the theoretical point of view. Traditionally, the study of banking competition has focused on market-structure theory and concentration indexes. Recently, however, the evidence has shifted, suggesting that some markets may be suffering from low industrial-concentration indexes along with low levels of competition, as evinced by far higher prices for financial products than international norms. The situation looks paradoxical.

An alternative approach that appeared recently was based on the theory of contestability in multi-product industries, the main argument here being that certain industries offering a wide range of products may in fact not be incurring entry costs

1 These studies will be referred to throughout this chapter.

2 OECD's 2002 edition on Mexico (OECD, 2002) was devoted exclusively to the financial sector and the challenges facing it.

in supplying one or more such products. Once profit has been obtained, they can exit that particular relevant market cost-free. In other words, companies involved in certain markets do have to face some form of contestability on the part of one or more entities.

This theory can be applied when identifying whether such entities have to confront contestability on the part of other financial or non-financial enterprises within the bank market. If no contestability exists, the situation could perhaps be due to the presence of some barrier to entry which should not be in place on principle, since this could limit competition.

Therefore, this paper maintains that competition should be analysed from the standpoint of the entire financial system – or, indeed, examining it in a setting that includes other non-financial entities offering financial-type products and services. Again, a study on competition in the financial sector should essentially take all agents into account; further, it should include the full range of financial products and services. Here the well-known saying that "protecting a company from competition gives rise to inefficiencies that are paid for by the rest of the economy" (Rajan and Zingales, 1998) should perhaps be rephrased to read "protecting one industry from competition from other industries – whether inside or outside the sector – gives rise to inefficiencies paid for by the rest of the economy – more specifically, by consumers."

This study will include the two approaches to identify bank competition problems. First of all, we look at the type of market structure and concentration indexes, followed by a sectoral analysis based on the contestability theory.

The results suggest that from the market-structure point of view, the Mexican system's bank sector shows no sign of concentration, except for two relevant markets: the first is note issuance and second, earnings from credit card interests. However, our analysis of other market indicators such as sector price levels (financial margins, coupled with fees and commissions and rates) reveals that competition is lacking in certain relevant markets within the financial system, prompting a review of possible barriers to entry – some legal – in a number of financial products. To put it another way, competition from other financial and non-financial intermediaries in some of the sector's overall services appears to be inadequate in the banking system, compelling us to examine one such product – credit cards – using contestability as the yardstick.

Contestability theory indicates that barriers to entry – legal ones – actually exist in the all-purpose credit card market, since banks are the only institutions permitted to issue them. Although cards issued by chain stores also exist, their share vis-à-vis bank credit cards differ, the issue here being different relevant markets.3 This helps explain why financial intermediation margins for that particular service have ranged far above international standards. One product whose price has dropped nonetheless is mortgage loans, following market entry by other non-financial intermediaries. In other words, a contestable market open to non-bank competitors has led to the ensuing drop in interest rates in this area.

3 Chain store's credit cards cannot be used outside this business.

Commission levels in the credit card market have also remained high as a result of lack of contestability. The difference compared to more advanced countries where fees and commissions have risen in response to the drop in interest levels in other products is that these are constantly challenged by other entities, both financial and non-financial. But high commissions in Mexico – although comparable to other countries – are not offset by lower intermediation margins as they are in the developed nations.

A policymaking feature on which Mexico should perhaps focus more closely is the possibility of reducing certain legal barriers to entry in some niches in order to promote more competitive markets. Among the most important are those that could make it possible to relax restrictions on the issuance of all-purpose credit cards (i.e. those not restricted to particular commercial endeavours) by non-bank agents. Similarly, the Mexican financial authority should study the possibility of placing bonds directly without financial intermediation, which could be helpful in promoting competition throughout the sector.

This study is structured as follows: The second section presents a review of certain theoretical concepts on competition, the third section analyses the status of competition in the banking sector, bearing two approaches in mind: the traditional view of concentration and the alternative one of contestability. The fourth section shows some of our conclusions.

I. Theoretical Concepts of Banking Sector Competition

Not long ago it became more generally accepted that financial intermediation actually generates economic growth. Many articles have even studied the causal relationship, both theoretical and empirical, in which a sound, competitive financial system contributes to sparking economic growth through different channels.4 In the first place, the financial system channels savings by offering a diverse range of alternatives for surplus units5 and savings instruments. At the same time, intermediaries are better at using financial analysis tools to identify worthwhile credit recipients. Likewise, such institutions are more efficient in collecting and processing client/borrower information.

Second, the financial system covers credit risks in an isolated manner, since by accumulating amounts from individual savings, financial agents can diversify their portfolios by channelling the proceeds into a wide range of investment opportunities.

Third, the presence of financial agents offers the possibility of a greater liquidity, thereby allowing savers quick access to their funds while at the same time financing long-term projects. Finally, risk management is made easier through intermediaries who can follow up on debtors and monitor managers to whom credit has been extended.

4 For an account of this aspect, see Hernández (2003).

5 A surplus unit is a synonym of a saver.

Financial intermediation can be useful as long as an appropriate operating framework exists in which on the one hand, the financial system is competitive and on the other, adequate regulation and prudential supervision prevail.

This section aims to review and comment on the current literature and empirical evidence that suggests that economic competition in a banking system is difficult to appraise and also because no information on costs and cross-subsidies is available in this type of industry, which is by nature multi-product. Paradoxically, the argument that excessive competition has been one of many factors that exacerbate financial crises throughout the world has also been put forward.

One feature that could indicate or suggest the level of competition in a given banking sector is access to the industry, or to put it another way, the existence of barriers to entry, a factor which has not been the object of much study in Mexico's case. Claessens and Laeven (2003) argue that this alternative approach to the study of competition in the financial industry has been possible because traditional analyses have placed too much focus on market structure and market concentration indexes rather than on 'contestability' in the relevant markets.

For contestability to exist, markets should be fully open to competition, even in cases where they are dominated by oligopolistic or monopolistic forces. That is, the presence of an inefficient incumbent could result in excessive market price levels and absorption of the entire consumer surplus, but when market entry is open to potential agents whenever possible and profitable, then contestability will force the incumbent to become competitive. Formally, the theoretical conditions for "perfect contestability" to exist are: (i) the entrant does not incur sunken costs; (ii) the entrant can begin to market its product before the dominant agent is able to implement a price change; and (iii) the entrant faces post-entry costs.6

On account of its having very clear policy implications, the contestability theory can be most useful in promoting an increase in competition in the financial sector, in contrast to other theories on competition and economic regulation. Bailey (1981), for example, has documented how the theory has been beneficial to the design of regulatory measures and economic competition policy in the USA.

To illustrate the above, Table 6.1 shows some indicators for the structure of bank markets in the world which suggest that high concentration levels in markets dominated by three large banks do not necessarily imply low – or inefficient – competition. As we can see, no clear relationship exists between concentration and greater competition: Finland is a case in point, where the three main banks concentrate more than 97% of bank assets, while intermediation margins are among the world's lowest. This indicates that the competition faced by these few banks probably comes from other, different financial intermediaries (Allen and Santomero, 2001).

Moreover, the changes the financial sector has undergone over the past decade in terms of technological innovation and deregulation processes have given rise

6 The concept of contestable markets is owed to Baumol et al. (1982), who argue that certain types of industries are subject to constant contestability resulting from lack of entry barriers and also when entry and exit costs are so low as to allow "hit-and-run" profits.

Table 6.1 Structure of world banking systems (World Bank, International Banking Statistics, several years)

Country

Bank Assets
assets/
GDP(%)

Classification

Government
bank assets(%)

Classification

Bank assets
help by foreigners (%)

Classification

Bank Assets
of the three
largest banks (%)

Classifiction

Intermediation
margin (% of total Assets)

Classifiction

Countries with high income levels

Australia

146.27

19

0

39

17.1

27

62.95

15

2.41

24

Belgium

315.12

6

0

40

24.1

23

57.39

21

1.15

7

Canada

153.84

14

0

42

6.1

41

55.32

23

1.76

15

Cyprus

76.27

33

3.3

35

10.9

34

78

7

1.6

12

Denmark

121.41

22

0

43

3.7

51

73.56

8

1.86

16

Finland

75.25

34

21.9

18

7.8

38

97.17

1

1.56

11

France

146.8

17

8.7

28

11.6

32

42.43

37

1.08

6

Germany

313.29

7

42

9

4.2

49

17.66

52

1.19

8

Greece

100.21

25

13

24

5

44

59.2

16

2.48

27

Israel

146.67

18

45.6

5

10.7

35

72.1

10

2.18

19

Italy

150.46

16

17

22

5

45

37.1

42

1.93

17

Japan

164.13

12

1.15

38

5.9

42

22.66

49

1.32

10

Korea

97.7

26

29.7

15

0

54

39.2

41

2.1

18

Luxembourg

3,423.18

1

5.03

33

94.97

3

17.06

54

0.36

3

The Netherlands

357.6

5

5.9

32

3.8

50

79

6

1.6

13

New Zealand

153.82

15

0

48

99

1

58.99

17

2.42

25

Portugal

238.29

9

20.8

19

11.7

31

34.2

45

1.6

14

Singapore

801.86

2

0

50

50

8

17.2

53

 

 

Slovenia

66.13

37

39.6

10

4.6

48

51.41

27

4.04

40

Spain

155.75

13

0

52

11

33

43.99

35

2.23

22

Sweden

128.91

20

0

53

1.8

52

68.99

12

1.3

9

Switzerland

538.9

3

15

23

8.5

36

67.06

13

0.85

5

UK

311.08

8

0

54

52.6

7

16.2

55

2.2

21

USA

65.85

38

0

55

4.7

47

21.48

50

3.35

36

Average

343.66

 

11.2

 

18.95

 

49.6

 

1.85

 

Countries with middle to high income levels

Argentina

54.24

45

30

14

49

9

29.8

46

4.9

43

Botswana

28.92

50

2.39

37

97.61

2

91.8

3

5.2

44

Brazil

55.17

43

51.5

4

16.7

28

44.7

33

5.3

45

Chile

96.58

27

11.7

26

32

18

41.24

39

3.82

37

Czech Republic

124.9

21

19

21

26

21

46.26

32

2.65

28

Estonia

59.33

41

0

44

85

4

92.4

2

3.9

38

Lithuania

26.88

52

44

7

48

10

72.9

9

0.13

2

Malaysia

166.07

11

0

46

18

26

41.26

38

2.2

20

Mauritius

96.15

28

0

47

25.8

22

80.7

5

3.17

34

Mexico

30.48

49

25

16

77.1

5

66.6

14

5.55

46

Panama

385.68

4

11.56

27

38.33

14

28.3

47

2.3

23

Poland

54.45

44

43.7

8

26.4

20

39.7

40

4.04

39

Saudi Arabia

92.81

29

0

49

0

55

58

19

2.9

31

Venezuela

6.03

55

4.87

34

33.72

16

44.27

34

13.2

53

Average

91.26

 

17.41

 

36.89

 

54.35

 

4.23

 

Countries with middle to low income levels

Bolivia

65.61

39

0

41

42.3

12

49

30

5.6

47

El Salvador

62.39

40

7

31

12.5

30

58.2

18

4.89

42

Guatemala

27.65

51

7.61

29

4.93

46

25.67

48

6.12

48

Jamaica

74.34

35

56

3

44

11

89.9

4

6.9

51

Jordan

213.92

10

0

45

68

6

52.7

26

2.9

30

Morocco

88.74

32

23.9

17

18.78

25

48.12

31

3.33

35

Peru

36.35

47

2.5

36

40.4

13

59.9

15

4.8

41

The Philippines

90.78

30

12.12

25

12.79

29

19.36

51

3.08

33

Rumania

25.49

53

70

2

8

37

57.58

20

3.71

50

South Africa

89.78

31

0

51

5.2

43

57.34

22

2.43

32

Thailand

116.85

23

30.67

12

7.16

39

53.56

24

0.54

4

Turkey

67.35

36

35

11

66.3

6

35.06

43

6.51

49

Average

79.94

 

20.4

 

27.53

 

50.53

 

4.53

 

Countries with low income levels

India

47.55

46

80

1

0

53

34.6

44

2.78

29

Indonesia

100.79

24

44

6

7

40

52.8

25

–3.84

1

Kenya

56.06

42

30.6

13

29.4

19

43.4

36

7.5

52

Moldova

25.29

54

7.05

30

33.37

17

49.3

29

17.91

54

Nepal

32.02

46

20

20

35

15

69.21

11

2.44

26

Average

52.34

 

36.33

 

20.95

 

49.86

 

5.36

 

Note: Income groups according to 2003 World Bank Classifications are as follows: low income: US$745 per capita or less; middle to low income: US$746–2,085; middle to high income: US$2,976–9,025; high income: more than US$9,026.

to a series of circumstances that make competition analysis in the banking sector highly complex. Of special note here are disintermediation, removal of barriers in certain products (in advanced countries), an increase in international capital flows, greater financial integration and a new set of risks to be covered and minimized (see Allen and Santomero, 1998).7 Several studies have shown that financial systems are more efficient and competitive in countries with low sector entry barriers; making lower active equilibrium lending rates possible (see Bhattacharya and Thakor, 1993).

1. Bank Competition: A Brief Review of the Literature

Focus has been placed on banking from the point of view of individual countries and of comparisons among many countries. As mentioned, the traditional approach in this regard has been market structure and level of market concentration.

However, recent studies – an important example being Claessens and Laeven (2003)–have shown that in the banking industry this does not necessarily mean a lack of economic competition. Barth et al. (2004) document that in some countries (Australia, Denmark, Finland, Sweden, Switzerland, among others), the assets are concentrated in three major banks; their quotient is higher than 60%, as can be seen in Table 6.1, and the industry's operation and financial margins are at competitive levels which in some cases are even below international standards. Here, concentration can play a much more complex role in interpreting the banking industry and the financial sector in general, than in other industries.

This may be due to the existence of scale and scope economies in many of the products provided by these institutions. For example, one of the major functions assigned to the banking system is more effective compilation and processing of information on possible credit recipients' current financial standing. This is because economies of scale and scope in this service do exist and these increase in keeping with the size of the bank in question (Thakor, 1992).8 Thus, an increase in competition as a result of a larger number of smaller-sized banks with less market power entering the market discourages investment in the compilation and processing of information9 on companies considered opaque, which can be very costly for small banks. We maintain that this argument should be used with caution as it excludes the possibility of related loans in bank conglomerates with large-scale horizontal and vertical

7 Hence the need for a new set of prudential rules, established in the 2003–Basel Accords II.

8 In Mexico's case the authorization and creation of the concept of All-Purpose Banks at the end of the 1970s was an advantage in terms of economic efficiency, as shown by Chávez-Presa (1990) on finding economies of both scale and scope.

9 Although credit bureaus solve some of the problems of small borrowers, this is not so when large-scale investments are involved, especially with companies that might be more opaque in the sense used by Morgan (2002).

integration, as was the case in Mexico prior to the 1994–1995 crisis, documented by Laporta et al. (2003).

Ideally, a competition analysis should include a detailed study of the price-setting process to verify whether or not a competitive balance would be likely to be obtained. But the literature acknowledges that this industry has special characteristics due to inherent information problems regarding the products and services it markets. Even so, comparing intermediation margins is a generally accepted practice in checking that such margins fall within a range in line with the standards, although, again, caution should prevail in this type of analysis due to cost-structure differences in each country, such as legal frameworks, tax structures, risk factors (political) and other more idiosyncratic factors such as income distribution itself.

Equally important would be a competition analysis to look at substitute financial products and services in domestic markets; this, however, lies beyond the scope of this study. An important point worth considering is whether competition between different types of financial intermediaries actually exists, as it does in most of the developed countries. Claessen and Laeven (2003) and Beck et al. (2003) have found evidence in panel studies that banking concentration is neither negative nor statistically correlated to competition levels, although the authors acknowledge that further research is necessary to ensure better grounded findings. The remainder of this particular chapter is devoted to a case study of Mexico.

In any case, a review of the traditional approach to concentration,10 complemented by some other form of analysis such as barriers to entry or contestability, would undoubtedly be of benefit. This is the path we will take over the course of this study. It should be pointed out that in order to better appreciate the status of banking competition in Mexico, a review of recent developments should be made. At the outset, banking competition concentrated on setting the stage for the sale of the banks and subsequently, as a consequence of the crisis, on mergers and acquisitions. Therefore, although this study does review such aspects, it actually focuses on more recent scrutiny (2000–2004) of competition policy. Furthermore, the necessary information has been made available relatively recently, among other reasons because an accounting change took place in this sector towards the end of the 1990s and also because by 1997 the Federal Commission on Competition (CFC) had only just begun to publish detailed information on certain aspects of competition.

2. Banking in Mexico

Over the years, banking in Mexico has been one of the economic sectors most seriously affected by the recurrent crises that shook the Mexican economy. Thus, the debt crisis in the early 1980s marked the beginning of a new stage for the banking sector

10 Such as the one presented by García Rocha (2004).

upon its nationalization.11 Further ahead, the banks were re-privatized and a few years later became once again the object of a financial crisis (1994–1995) which ended in a rescue operation.12

Since then, the banking system has taken around 10 years to recover, with the aid of foreign bank inflows and improvements in the sector's legal system. Today we can say that Mexico's banking indicators point to a healthier situation from the point of view of financial strength. Nevertheless, the evidence points to the existence of low competition levels in the banking system, as shown by high intermediation margins and the commission levels charged for different services. Based on an empirical analysis of foreign bank entry, between 1997 and 2004, Shulz (2006) shows that: "FDI had a positive, but limited impact on banking sector development. The key contribution of foreign banks was the recapitalization of the banking sector following the financial crisis. But there is only limited evidence that banking sector efficiency increased as result of a transfer of skills, technology, or management know-how. The main reason for the limited impact of FDI was the low level of competitive intensity in the Mexican banking sector". It is this aspect we are about to look at.

II. Banking Competition in Mexico 2000–2004

For the purposes of our study, we should point out that in Mexico, as in most countries, the Federal Law on Economic Competition (LFCE in Spanish) monitors the financial sector with the same provisions as those that apply to any other economic sector, whether in a preventive manner (mergers, for example), or in its role as investigator of any absolute or relative practices that could arise as a consequence of a financial agent's market power.

The law has two underlying principles. The first seeks economic efficiency and the second is that the law itself be enforced so as to include all economic agents in the relevant markets. From this standpoint, LFCE should formally apply to the financial sector to guarantee effective competition and lower the cost of goods and services, much in the same way as in any other sector of the economy.13 With this in mind, we take a look at the bank mergers that have taken place in the country within the context of the worldwide situation.

11 For a fine analysis of the banking situation during the bank nationalization, as well as the events that preceded it, see Del Angel et al. (2005).

12 A vast number of interpretations of the causes and consequences of the "tequila crisis" are in print. This is not an object of study here, but an interested reader could look at Kaminsky et al. (1997) for a review of several crises. For Mexico, an earlier study warning the macro problems that lay ahead is Arellano et al. (1993), and among those seeking to explain it are Hernández and Villagómez (2000), Calvo and Mendoza (1996), Kaminsky and Reinhart (1996), McKinnon and Pill (1995), and Sachs et al. (1996).

13 For a wide-ranging discussion of the LFCE's principles and implementation, as well as its regulation of mergers and monopolistic practices, see Ávalos (2006).

1. Mergers and Relevant Markets: A Brief Economic Analysis

It would seem pertinent to begin this analysis with a brief review of the mergers that have taken place in Mexico, describing behaviour patterns within the financial sector as well as the competition authority's reaction to such mergers. This is important, since much of competition policy in this sector has been aimed at ensuring that the mergers have not unfavourably increased concentration.

The database for this section on mergers was built mainly from official CFC (in Spanish henceforth) resolutions published in the Gaceta de Competencia Económica (Economic Competition Gazette), from its annual reports and also from indirect sources. The period covered by the study runs from April, 1997, to December, 2001.14 Most of the cases were reviewed, but in spite of the efforts made, the information contained in resolutions was incomplete in certain situations in the sense of not including all cases, so we chose only those whose information was complete – or practically so; thus, the final sample included 96 cases representing most of the mergers that took place during the period covered by the study.

Article 12 of the LFCE establishes the following yardsticks by which to measure the impact of mergers on competition: the degree of substitution of goods or services, and barriers to entry such as distribution costs, new technology, regulatory restrictions of a federal, local or even international nature that hinder access to the good or service. Thus, establishing the relevant market varies from case to case. As regards financial activities, there is a wide range of products and services in which there may be a high level of diversification or combination of the same. Table 6.2 shows certain relevant markets within the financial sector on which CFC bases its decisions.

On different occasions we find that both bank and non-bank institutions offer financial products that could be savings- or credit-like substitutes operating in the same relevant market. For example, in the specific case of housing loans, different financial agents (banks, savings-and-loan associations, credit unions, Sofoles,15 car financing, etc.) compete in the market, so an analysis of mergers should include all the participating financial agents. Another example is the money market, which includes all brokerage operations (Cetes, 28-day Bondes, UDI-bonds, notes, TIIE futures and bank bonds) in which around 51 financial intermediaries participate, including banks and stockbrokerage firms; likewise for the foreign exchange market in which both domestic and international financial agents take part. Finally, diverse financial agents such as independent companies, banks and stockbrokerage firms compete in the mutual fund market.

14 CFC began publishing its resolutions in April, 1997. It should be mentioned that 2001 was the last year an important merger was registered. The database has therefore not been updated to 2004, since it would not be relevant to our analysis.

15 SOFOLES: Sociedades Financieras de Objeto Limitado (Limited-Purpose Financial Companies) linked particularly to the mortgage-loan market.

Table 6.2 Mexico: financial sector. main relevant markets regarding mergers, 1997–2001 (Prepared by the authors based on SICMACEM data)

Codification

Relevant market

No.
Mergers

Percentage

303

Intermediation services provided by financial institutions to savers and credit recipients in the national territory (all-purpose banks)

20

25.6

304

Insurance services covering life, accidents and sickness, and damages (insurance policies)

12

15.4

317

Leasing services for car purchases in the national territory (NT) (Leasing company, car financing)

1

1.3

348

Afores and Siefores. Administration of savings for retirement of IMSS-beneficiaries (workers) which are deposited in individual accounts.

16

20.5

356

Services provided through stockbrokers in the national territory

4

5.1

371

Factoring services in the national territory

1

1.3

398

Corporate banking

4

5.1

443

Mutual funds and exposure investment

1

1.3

452

Provision of reinsurance services in the national territory

1

1.3

472

Services for audits, analysis of financial statements, consultancy, tax assessment

1

1.3

534

Foreign exchange firms in the national territory

1

1.3

677

Provision of financial leasing services in the national territory

3

3.8

722

Insurance and insurance companies specializing in pensions

2

2.6

834

Mortgage credit services in the national territory

5

6.4

888

Mortgage financing service for the construction of low-cost housing in the national territory

1

1.3

928

Lending of transfer services from the viewpoint of acquiring bank to issuing bank; of transfer from ATMs to banks belonging to the RED (network) system

1

1.3

958

Factoring services in the national territory

1

1.3

 

Others

3

3.8

 

TOTAL

78

100.0

At sectoral level, mergers under the financial heading proved to be one of the most dynamic items.16 As can be seen in Table 6.2, the relevant market17 for savings and credit intermediation by financial institutions in Mexico (all-purpose banking) was the one that registered the largest number of mergers (codified under number 303 in Table 6.2). Another relevant market worth noting in which merger activities stand out is that of Afores and Siefores with 16 cases (codified under number 348).

16 See Avalos (2006), Table 7.

17 The level of disintegration implied by the notion of relevant market is at the product(s) level. The base for the Classification System for Economic Competition in Mexico (SICMACEM, in Spanish) includes the codifying of over 1,000 relevant markets covering economic activities. See Avalos (2006, p. 49–52) for a broader description of the building of the SICMACEM base.

To a lesser extent, the relevant markets for insurance policies (304), services offered through stockbrokerage firms (356), corporate banking (398), leasing services (667), and mortgage credit services (834) were also relatively dynamic. Furthermore, mergers were classified as pure horizontal and administrative horizontal mergers,18 pure vertical and administrative vertical mergers,19 and diversified mergers (or conglomerates), of which there are three types: diversified by product extension, diversified by market extension and pure diversified.20

From this point of view, Table 6.3 shows that of the total of mergers, slightly over half fall under the heading of "pure" horizontal mergers (54 cases), and second, diversified mergers of various types with 36% of the total (34 cases). Among this type, the most dynamic were those known as diversified administrative product extension and pure diversified mergers with 50% and 17.6% of the total, respectively.21

By contrast, administrative horizontal mergers and administrative vertical mergers were not as important in terms of cases. The former represented only 5.3% of the total (5 cases). It should be pointed out that this type of merger does not have a major impact on the implementation of competition policy as they do not have a direct effect on the relevant market and only involve an internal restructuring of the firm or corporation to which the merged company belongs. During that period not a single pure vertical-type merger was registered within the financial system. For the same type of mergers, but administrative, only one case appeared.

It should be noticed that the behaviour patterns in horizontal versus vertical mergers are not a distinctive feature of the Mexican banking system. The annual reports of certain competition agencies in the world – the USA and the European Union, for instance – confirm this trend. It should therefore come as no surprise that

18 Horizontal mergers are those in which companies compete for one same market. The potential danger to competition derived from these type of mergers comes from the reduction of the number of competitors in a given market. Nonetheless, horizontal mergers can, under certain conditions, actually increase economic efficiency. By contrast, an administrative horizontal merger has no effect on the market whatsoever since it involves a company's internal restructuring. For a wider-ranging discussion see Avalos and Melgoza (2002).

19 Vertical mergers take place between economic agents carrying out different stages of the productive process in one same industry. Such operations have no effect on market concentration, although they can favour relative monopolistic practices. As in the case of horizontal mergers, an administrative vertical merger has no effect on market concentration (Avalos and Melgoza, 2002).

20 Conglomerate mergers include three types: those known as product extension between companies producing different goods but using distribution and marketing channels or applying similar productive processes. The second type is that of market-extension operations between companies producing one same product but in different geographical markets. Finally, there are 'pure' conglomerate mergers between totally unrelated companies, i.e. not participating in related markets, or that do not share production technology or distribution systems (Avalos and Melgoza, 2002). In the same way as the above types, administrative conglomerates have no effect on the relative market. It should be mentioned that CFC does not establish any classifications in case resolutions. Nor does the paper by García Rocha (2004) present such a level of breakdown in merger analysis.

21 Murillo (2005) also presents a merger analysis, although not from the Industrial Organization viewpoint.

Table 6.3 Mexico: resolutions of the federal commission on competition by type of merger financial sector,. 1997–2001 (Prepared by the authors based on SICMACEM data)

 

Authorized

Conditioned

Objected

Resolution of the CFC

Cases

(%)

Cases

(%)

Cases

(%)

Pure horizontal

54

57.4

1

100.0

1

100.0

Administrative horizontal

5

5.3

0

0.0

0

0.0

Administrative vertical

1

1.1

0

0.0

0

0.0

Pure diversification

6

6.4

0

0.0

0

0.0

Diversification as

5

5.3

0

0.0

0

0.0

product extension

 

 

 

 

 

 

Administrative

17

18.1

0

0.0

0

0.0

diversification as product

 

 

 

 

 

 

extension

 

 

 

 

 

 

Diversification as market extension

5

5.3

0

0.0

0

0.0

Administrative diversification

1

1.1

0

0.0

0

0.0

as market extension

 

 

 

 

 

 

Total affairs

94

100.0

1

100.0

9

100.0

most of the formal literature focuses on topics related to horizontal-type mergers. Nevertheless, this inclination in economic theory has not been altogether well received by specialists in this area – particularly by competition agencies. Many of the most complex cases addressed by the different authorities have to do with vertical mergers or restrictions. Consequently, research in this area still has a long way to go if the gap between theory and professional practice with regard to the implementation of competition policy is to be reduced.

As for resolutions, virtually all financial sector cases reviewed by CFC were authorized, 1% was conditioned and only one case was objected to during the period 1997–2001. The latter figure, an extremely small one when compared to the decisions adopted by other authorities in the world, is particularly noteworthy. Competition authorities such as the Federal Trade Commission (FTC) and those of the European Union are backed by much stricter laws than in Mexico, and these have made merger processes more difficult, and consequently the process of industrial concentration, too. In the USA, for example, the evidence shows that since 1940, active implementation of US antimonopoly legislation has made it more difficult for mergers to acquire market power (see Andrade et al., 2001). In Mexico's case, it cannot be said that the LFCE has played a leading role in comparison to the legislation in that country.

In short, from 2000 to 2004, bank mergers have continued, as shown in Table 6.4. Whether or not such mergers have harmed competition is difficult to determine owing to the lack of information at the level of each relevant financial service. From the point of view of concentration indexes, this has apparently not harmed the competition process, as will be noted further ahead. However, we can point out that no pro-competitive effects resulting from the mergers can be perceived either, and furthermore, the effects of scarce competition are reflected by certain phenomena such as high financial margins and fees and commissions on bank services. These aspects are an underlying feature of the subsection that follows.

Table 6.4 MERGERS 2000–2003 (Instituto de Protección al Ahorro Bancario (IPAB ) )

1999

2000

2001

2002

2003

Afirme

Afirme

Afirme

Afirme

Afirme

Banamex-Citibank

Banamex-Citibank

Banamex-Citibank

Banamex-Citibank

Banamex-Citibank

BBVA-Bancomer

BBVA-Bancomer

BBVA-Bancomer

BBVA-Bancomer

BBVA-Bancomer

Banorte-Bancrecer

Banorte-Bancrecer

Banorte-Bancrecer

Banorte-Bancrecer

Banorte-Bancrecer

Banregio

Banregio

Banregio

Banregio

Banregio

Bital

Bital

Bital

HSBC

HSBC

Del Bajío

Del Bajío

Del Bajío

Del Bajío

Del Bajío

Inbursa

Inbursa

Inbursa

Inbursa

Inbursa

Industrial

Industrial

Industrial

Industrial

Industrial

Interacciones

Interacciones

Interacciones

Interacciones

Interacciones

Inverlat

Inverlat

Scotia-Inverlat

Scotia-Inverlat

Scotia-Inverlat

Invex

Invex

Invex

Invex

Invex

Ixe

Ixe

Ixe

Ixe

Ixe

Mifel

Mifel

Mifel

Mifel

Mifel

Quadrum

Quadrum

Quadrum

Quadrum

Quadrum

Santander-Serfín

Santander-Serfín

Santander-Serfín

Santander-Serfín

Santander-Serfín

 

 

Ve por más

Ve por más

Ve por más

 

 

Azteca

Azteca

Azteca

20

18

18

18

18

2. Concentration and Banking Indexes in Mexico

The purpose of this section is to analyse the impact of the above mergers on concentration indexes in the industry.

First of all, we should mention the assessment criteria established by CFC. Concentration analysis based on the Hirschman-Herfindhal Index (HHI) and Dominance (DI) indexes22 is probably CFC's best-known merger-assessment yardstick, having set the following index classification23:

• When the HHI, following the merger, increases by less than 75 points or is lower than 2,000 points, the merger is unlikely to be objected to.

• When the DI, following the merger, either drops or stands at less than 2,500 points, the merger is unlikely to be objected to.

22 For wide-ranging discussion and application of alternative concentration indexes, see Tirole (1997) and Church and Ware (2000).

23 CFC published these criteria in the Official Gazette of the Federation (DOF) on July 24, 1998. HHI has values ranging from 0 to 10,000 points (also measurable from 0 to 1) and is equal to the sum of the squares of market participations. HHI is widely recognized in the literature and used extensively by many competition authorities throughout the world. DI sizes a company in relative terms in comparison with others in the relevant market. Under this index, a merger does not necessarily increase the index as in the case of HHI. For instance, when two small companies merge the index may even diminish. For further details of this index and its properties see García (1990).

Table 6.5 Dominance (DI) and Hirschman Herfindhal Index (HHI) headings for the Mexican banking system, 1998–2004 (Authors' calculations based on CNBV)

 

1998

1999

2000

2001

2002

2003

2004

Interest income in favour from securities

DI

0.2640

0.3576

0.2306

0.2328

0.1880

0.3260

0.2762

HHI

0.1520

0.1761

0.1310

0.1294

0.1225

0.1444

0.1383

Commissions and rate

DI

 

 

0.3532

0.3695

0.3630

0.3360

0.3361

HHI

 

 

0.1987

0.1924

0.1853

0.1788

0.1769

Interest income from availabilities

DI

0.8883

0.3917

0.2599

0.2798

0.2304

0.3050

0.3178

HHI

0.3849

0.1912

0.1341

0.1411

0.1371

0.1603

0.1660

Assets

DI

0.2941

0.2740

0.3003

0.3581

0.3138

0.3206

0.3110

HHI

0.1639

0.1547

0.1485

0.1621

0.1537

0.1528

0.1458

Term deposits

DI

0.3090

0.2662

0.3180

0.4244

0.3590

0.3421

0.2998

HHI

0.1691

0.1488

0.1526

0.1703

0.1632

0.1638

0.1501

Invest in securities

DI

0.3700

0.3329

0.2442

0.3214

0.3876

0.4269

0.3624

HHI

0.1769

0.1660

0.1410

0.1653

0.1948

0.2143

0.1713

Demand deposits

DI

0.2805

0.3064

0.3135

0.3483

0.3180

0.3154

0.3246

HHI

0.1855

0.1891

0.1891

0.2088

0.1906

0.1844

0.1808

Bank bonds in circulation

DI

0.6462

0.7772

0.7172

0.2916

0.4569

0.3185

0.8853

HHI

0.2783

0.3831

0.3294

0.1948

0.2843

0.2408

0.5870

Current loan portfolio

DI

0.3017

0.2745

0.3063

0.3607

0.2947

0.2747

0.2972

HHI

0.1644

0.1592

0.1591

0.1804

0.1600

0.1515

0.1479

Income from financial intermediation

DI

0,2636

0,2588

0,3197

0,3606

0,3028

0,2718

0,2670

HHI

0,1553

0,1534

0,1638

0,1865

0,1686

0,1496

0,1394

The indexes for the period 1998–2004 in Mexico were drawn up on that basis. Table 6.5 shows the DI and HHI for a series of variables that constitute the relevant markets for the years 1998–2004: interest income in favour from securities, fees and commissions and rates charged; and interest income in favour for availabilities, assets, term deposits, investment in securities, demand deposits, bank bonds in circulation, current loan portfolio, and income from financial intermediation. As can be seen, in virtually all these variables both indicators – DI and HHI – suggest that a considerable bank concentration does not exist, except for bank bonds in circulation from balance sheets, where the market shows considerable concentration.24 This factor should be closely

24 Although in most of the relevant markets the DI exceeds the criteria established by the CFC itself (higher than 2,500 points), this is not indicative of a high level of concentration in the said markets as they do not exceed a symmetrical duopoly, i.e. above 0.5 (or 5,000 points).

followed up by Banco de México, since it may affect interest rates and thus monetary policy as well.

As we can see, for earnings in favour from availabilities, in 1998 the DI registered 0.88, an extremely high figure, which would apparently suggest a concentration level far higher than the criteria established by the Mexican authority itself (higher than symmetrical duopolistic case). It should nonetheless be made clear that the figures register zero for the two largest banks (Banamex and Bancomer) under said heading for that year, which the DI could have contaminated. It should also be appreciated that for 1999, when these two large banks published information, the DI dropped abruptly into acceptable limits.

Table 6.6, moreover, displays the market share of the system's three largest banks in each of the previously calculated variables for concentration indexes plus net profit, which in themselves do not make up a relevant market. As shown, the three largest banks generally take a high percentage of market share. This sounds contradictory against the concentration indexes.

Beck et al. (2002) and Claessens and Laeven (2003), on the other hand, find no evidence of a negative relationship between concentration and level of competition. Let us recall that Table 6.1 shows that countries with such market share indicators for the three largest banks did not suggest scarce economic competition, either. We reiterate that these indicators do not make it possible to determine competition levels. So far, the results for Mexico have been

Table 6.6 Ccontribution of three largest banks to various incomes of the financial system, 1998–2004 (%percentages) (Authors calculations on the basis of CNBV information)

Year

Incomes
linked to
stocks

Commissions

Incomes from
interest rates

Assets

Long-term
deposits

Investment
in stocks

1998

59.89

 

74.69

64.33

65.18

64.24

1999

65.82

 

69.56

61.90

60.22

64.42

2000

52.54

71.79

55.05

57.90

58.08

56.71

2001

51.38

69.87

55.85

59.97

60.21

63.41

2002

49.23

67.68

54.78

57.55

61.46

68.52

2003

59.92

67.70

61.58

59.97

62.40

71.01

2004

54.71

67.04

61.38

57.08

57.66

62.67

Year

Immediately
exigible
deposits

Bank bonds
in circulation

Loan portfolio
of active loans

Incomes
from financial
intermediation

Net profits
from sub
sidiaries and
associates
participation

Net profits
from three
largest banks

1998

68.37

75.75

64.68

63.02

97.67

85.79

1999

69.66

85.78

64.50

62.57

NA

NA

2000

70.06

81.07

62.07

62.93

74.60

72.30

2001

73.03

69.39

63.79

65.50

68.19

67.38

2002

68.33

85.57

62.71

64.87

83.01

69.73

2003

67.72

79.42

59.79

60.12

65.26

63.48

2004

67.37

100.00

57.98

57.20

40.50

48.99

ambiguous. Therefore, it is advisable to consider alternative paths.25 The next section aims towards that objective.

3. An Alternative Attempt: The Contestability Approach

This section introduces the contestability approach applied to Mexico's case. We must emphasize that this type of methodology is in its early days from the empirical point of view, although from a theoretical stance it is generally accepted in the literature (see Tirole, 1997). In Mexico's case the attempt is necessary, as it is important to explain why, from the point of view of concentration, no structural evidence of lack of competition exists; paradoxically, however, prices remain above international standards.

Most of the empirical studies aimed at measuring competition through price fixing have been based on some measure of profitability to try to find the relationship between concentration and the aforementioned variable. If the relationship were positive, it would indicate that price fixing would be less favourable to the consumer, thus harming competition. Berger and Hannan (1989) argue that rather than some measure of profitability, it is better to use the different interest rates paid to depositors rather than average deposit-taking rates. In a study for the USA, the authors found a negative relationship between different instruments' deposit-taking rate and concentration in the bank market. The difficulty they came across, but managed to overcome, was registering (colloquially: getting) information on the rates for different types of deposits and for different regions of the USA, without taking price discrimination into account. The virtue of their study lies in having obtained and processed the said rates, which were not available for most of the countries in a systematized manner.

In Mexico's case it is difficult to reproduce such an exercise, precisely because detailed information on every financial product is not available on a sufficiently long-term basis. Even so, it is worth looking at Pearson's simple correlation coefficient to determine, worldwide, the type of relationship (although statistically insignificant) between average cost of deposit-taking (CPC in Spanish), average term-deposit rates (DP, in Spanish) and average demand rates (DEI, in Spanish). These are shown in Fig. 6.1 below.

The figure suggests an apparently negative correlation, and this is corroborated by correlation coefficients which actually are negative: CPC against DP stands at −0.41, whereas CPC against DEI is at −0.36. Although we cannot conclude from this that there is a statistical relation, nor does it reflect a relation at sectoral level, it does suggest that there is evidence – weak–that the latter is determined unfavourably for the consumer even though, we insist, these rates represent an average and do not reveal the characteristics of each of the products – differentiated.

25 A study on competition for Mexico was made by Rojas (1997), who appraises whether banks "steal" clients from one another. This original article uses Markov's transition chains. Unfortunately, it is the only part of banking competition it takes into account.

3.1. Financial Margins and Competition

Another more heuristic way normally used to try to infer whether price fixing in the banking system is competitive is by comparing the intermediation margins of a series of countries in such a way as to verify whether the financial margin falls within international ranges, mainly regarding countries with developed financial systems.

Table 6.7 shows the percentage intermediation margin measured as the active bank rate minus the free risk rate. This margin is an average of active bank rate. As may be seen in Fig. 6.2, this margin has been gradually falling off for Mexico, suggesting an increase in efficiency, having dropped from 10.1% in 1995 to 8.5% in

Image

Fig. 6.1 Deposit-taking costs and deposit rates (CNBV, Banking System Financial Statements, several years, and CONDUSEF)

Table 6.7 Operative costs as a percentage of the balance sheet, 2000 (CNBV and Financial Statistics of OECD (2002). For Mexico, the active bank rate is the investment credit and the mortgage credit average)

Country

Operating cost

Active bank risk – Free risk rate

Czech Republic

16.5

NA

Mexico

5.8

7.5

Turkey

6.0

NA

Hungary

4.1

NA

USA

2.9

3.4

Great Britain

2.0

1.8

Germany

1.8

2.2

France

1.0

2.3

Japan

0.8

1.8

1996 to 7.5% in 2000. In 2003, the percentage was similar, which suggests that it remains high with respect to international standards. In short, although the decline in the spread may be attributed to increased competition, this may also be due to the system's greater administrative efficiency. This particular aspect follows.

One way to explain Mexico's high intermediation costs is the high level of its operational costs, which include staffing, property, and other costs. If we look at Table 6.7, Mexico's performance is only a slight improvement over the Czech Republic and Turkey in this regard for OECD sample countries. This means the Mexican banking system could do a little more in terms of efficiency, which would in turn result in lower intermediation margins, thereby benefiting consumers of these financial services in Mexico. At any rate, comparisons between intermediation margins are not simple, since these fail to take into account different legal frameworks, cost structures, as well as the economic conditions and uncertainty prevalent in the different countries. We now move on to Fig. 6.3, which makes it easier to see the size of financial margins.

Image

Fig. 6.2 Mexican banks' financial margin (OECD, 2005)

Image

Fig. 6.3 Financial margin banking active rate – free of risk rate (2001) (OECD, 2005)

Financial product prices vary considerably, and interest rates represent just one such factor. In general, the prices of other products that do not represent credit or intermediation as such fall within the framework of what are known as commissions and rates for different services. This issue has given rise to much worldwide debate in recent years, as it has in Mexico. As we can see, this heading is an important one, inasmuch as it provides us with guidelines regarding the existing de facto contestability in the sector, an alternative approach to analysis in the sector.

3.2. Fees and Commissions and Contestability in the System

The fact that banking in Mexico has increased the fees and commissions charged for the different products and services it provides has been extensively argued.26 Mexico's Banking Association (ABM, in Spanish) has countered with the argument that the rise in commission levels is a worldwide phenomenon (www.abm.org.mx). For example, this has occurred in the USA, Great Britain, France, Brazil and Chile, among others.27 But ABM rightly argues that international comparisons require caution, as different important aspects have to be taken into account, many of which have been ignored in many studies. Such aspects include: (i) characteristics of the product: for example, not all cards or accounts are the same, they do not have the same lines of credit for equal amounts, and they have different services, among other considerations; (ii) customers' habits in using them: balances, check issuance or window withdrawals, ATMs, etc.; (iii) regulation: diverse regulations can artificially raise the cost of certain services, whereas others do not do so; and (iv) costs, taxes and market size.

Although such difficulties exist, a study on competition should, heuristically, be made on the basis of international comparisons, in view of the lack of detailed market information at the domestic level, especially in the developing countries. As we have mentioned, it is difficult to obtain information on cross-subsidies among the different products and services provided by a multi-product industry such as banking (a product with market power in one relevant market can finance another product within the same industry).

An international study on competition between different financial intermediaries indicates the following:

First of all, in recent years banking units' earnings in connection with commission-taking in industrialized countries such as the USA, Great Britain, France, Germany and Japan, have actually risen. To keep it simple, we refer to the case of the USA alone; although the results obtained for that country have also been validated for France, Great Britain, and Japan (see Schmidt et al., 1999 for the case of Europe, and Miles, 1996, in Japan's case).

26 This has been stated by the current Governor of Banco de México, Mr. Guillermo Ortiz Martínez (in a speech given on the 30th anniversary of CIDE on October 3, 2004).

27 In September, 2004, CONDUSEF prepared a study (CONDUSEF, 2004) on bank commissions which also concluded that commissions had risen throughout the world.

The USA illustrates the point since it is one of the countries, together with Britain, which has developed its financial system the most. In the former country, following the 1929 economic crisis, the banking system confronted the severest crisis in its history, and extremely cautious action was taken in that sector. This helps explain the well-known "Q regulation," which prevented the banking rules from being extended nationwide and which placed restrictions on interest rates. For this reason the banks, in order to compete for customers, used cross-subsidies intensively within their own enterprises; in other words, they competed by means of commission-payment exemptions and further, compensated for this with earnings from intermediation. Among other factors, and having relaxed the Q regulation28 in 1982–1984, this favoured the entry of a series of competitors into the banking system, an aspect shown in Fig. 6.4.

This figure shows that banking lost a portion of its share of total assets in the US financial system. As stated earlier, this was the result of increased competition between different types of financial intermediaries, with emphasis on Pension Fund Management Companies and Mutual Funds, as well as stockbrokerage.

Moreover, for some products in certain market segments, banking has confronted competition from other financial and non-financial institutions offering a variety of services. One significant aspect is the credit card market, in which airlines, department stores and automotive corporations, among others, have issued credit cards managed by Master Card or Visa without the need for involvement by any

Image

Fig. 6.4 Distribution of the financial system total assets by intermediary in the USA (%) (Barth et al., 2000)

28 For an analysis of these regulations and their removal, as well as the consequences of these actions, see Kane (1981, 1984 and 1989).

bank. Figure 6.5 demonstrates this phenomenon; market share in this US market fell from 95% in 1986 to 23.8% in 2000. Clearly, competition from non-bank and non-financial companies has increased considerably in that market.

To sum up, banking in the USA has had to face not only interbank competition but also competition from other financial and non-financial institutions providing substitute products at better prices (see Allen and Santomero, 2001). That is, banks have been "contested" by other entities in this relevant market.

Therefore, the banks' response has been to eliminate part of the cross-subsidies in force under the Q regulations, which has increased fees and commissions and rates on products that still hold a significant market niche with few substitutes (Allen and Santomero, 2001). The following figures (Figs. 6.6 and 6.7) show developments in

Image

Fig. 6.5 Bank share in the credit card market in the USA (Allen and Santomero (2001) based on the Card Industry Directory, USA)

Image

Fig. 6.6 Income from fees and commissions (percentage of GDP) (Barth et al., 2000)

Image

Fig. 6.7 Earnings from interests (percentage of GDP) (Barth et al., 2000)

earnings from fees and commissions and interests in the USA. A negative correlation between these two variables can clearly be seen, which supports the above hypothesis. Much the same has happened in the case of mortgage loans, where the mortgage market has displaced the banking system in this segment. It should be stressed that the banking system's drop in earnings from interests took place over one decade (the 1990s), during which the US economy was experiencing an economic boom. This also occurred in at least France, Japan, Germany and Great Britain and other countries; this was documented by Schmidt et al. (1999).

To recapitulate, the loss of consumer surplus caused by a rise in commission levels is offset to a certain extent by an increase in the above-mentioned surplus derived from a reduction in intermediation margins.

Recently, in fact, competition agencies have begun to regulate the credit card relevant market in regard to fees and commissions, compelling the issuing companies to reduce the latter. This has happened in Great Britain, Australia and the European Union.29 In Mexico's case, developments as regards increasing fees and commissions have evolved differently (Fig. 6.10a). Although in this country the banking system has been the protagonist in the financial system, in the rest of the developed world banking has recently lost ground. Figure 6.8 again reveals the US financial system's distribution of total assets by financial intermediary, but adding the structure of the Mexican financial system in 2004 in the last column better illustrates the point.30 Here lies the difference between the US system and Mexico's. Whereas in the US banking market the banks' share has declined, in Mexico this intermediary continues to predominate (with virtually an 80% share of the financial system). In the USA, the evolution of the banking system has translated into lower intermediation margins, thus causing an increase in commission levels to compensate for the reduction.

29 The Economist (2005).

30 The structure in 2004 in the USA has remained essentially unchanged in comparison to 2002, according to Barth et al. (2004). This author does in fact suggest that a stable situation in the industry's structure had been reached.

Image

Fig. 6.8 USA (1950–2000) and Mexico (2004) structure of the financial system (Barth et al. (2000) and the authors' calculations based on CNBV data)

Image

Fig. 6.9 Development of charges for fees and commissions (Prepared by the authors with CNBV data)

In Mexico, on the other hand, whereas commission payments have increased much in the same way as in the USA, intermediation margins have remained constant in recent years; i.e. at high levels when compared to international standards. That is, even without competition from other financial and non-financial intermediaries in Mexico earnings from commissions have risen considerably, especially in the last five years. This particular aspect sets Mexico apart from the rest of the OECD countries.

Figure 6.9 shows the share of earnings on fees and commissions in Mexico. As may be appreciated, net commissions charged as a proportion of the financial margins

Image

Fig. 6.10 a Earnings from interests/total portfolio (CNBV, the banks' financial statements) b Earnings from interests (CNBV, the banks' financial statements)

are extremely high, fluctuating at around 50% in both cases, which is higher than international standards.

This analysis suggests that the Mexican banking system is facing practically no competition from other financial intermediaries. In contrast to banking in the USA, where fees and commissions have risen but earnings from interests as a proportion of current portfolio have dropped as a result of competition, in Mexico, as seen in Fig. 6.10a, earnings from interests as a proportion of current portfolio have not diminished, but have remained stable over the past four years. It should be noted that this happened despite the adverse macro setting in which the country was mired, especially between 2001 and 2003. Figure 6.10b shows earnings from overall interests as well as their breakdown. Emphasis should be placed on the fact that the former have increased in recent years.

Regarding the composition of earnings from interests, those from securities and housing loans have remained stable for the past five years, whereas those from commercial credit have even dropped, the most notable being the growth of consumer credit, which to a large extent represents the interests charged for credit card use, a market which is practically exclusive to Mexican banks and not open to other intermediaries (see Fig. 6.10b).

Another way of highlighting the problem is by reviewing the interest rates charged on the use of credit cards. Table 6.8 shows average rates for a series of banks. As may be seen, the average level has remained constant for the period 2001–2004.

This "suggests" that market power under the item credit cards is high. Certainly the financial margins (average active credit card rate less passive deposit-taking rate – CPC) under this item stand at around 30 percentage points, although it has declined slightly over time. Here we should make it clear that this type of credit has increased considerably in the last five years, partly due to better economic performance and partly to the fact that the banks became consolidated during that period. However, in comparison to other types of credit, it has increased unusually, so here it is pertinent to ask whether this relevant market is concentrated. We would like to emphasize that in the preceding section it was not possible to obtain the respective concentration index, since CFC does not consider the credit card market a relevant one. In our view, as the line of argument that follows will show, it is.

Mexico clearly has a market for credit cards issued by commercial chain stores. Although the consumer credit extended by these chain stores has gained ground over the past five years, they cannot actually be considered a perfect substitute for credit cards. This stems from the fact that the credit cards issued by banks are generally accepted in a very wide range of business establishments, whereas those of the latter restrict the wide range of consumption opportunities to a single business. Hence competition among these two types of cards concerns different relevant markets, since banking is funded, on average, at a 5% rate and lends to its clients at an annual 35%, whereas chain stores obtain resources at 12% on average, and charge a rate that may vary over a span of 40% to up to 80%, depending on the establishment, according to the figures put out by both banks and the businesses in question

In fact, according to the General Law on Credit Titles and Operations, the major stores are permitted to issue credit, for they comply with the requirement of being registered in the Public Registry of Commerce. So the major chain stores would have to be permitted to issue generally accepted cards. Thus, the financial margin would be reduced, since there would be greater competition in the credit card sector, as has already been the case in advanced countries.

We must stress that the financial services provided by commercial establishments are carried out through the banks themselves, and therefore such chains do not require authorization and their role is limited to extending their own services and linking up the customer with the operator, from which they obtain a commission,

Table 6.8 Loan rates evolutiona (%) (INFOSEL Financiero)

 

2000

2001

2002

2003

2004

2005

 

Monthly

Annual

Monthly

Annual

Monthly

Annual

Monthly

Annual

Monthly

Annual

Monthly

Annual

American Express

3.33

39.96

3.115

37.38

3.167

38

2.826

33.917

3.225

38.695

3.284

39.412

Banax

3.957

47.49

3.263

39.16

3.023

36.275

2.818

33.815

3.041

36.490

3.098

37.179

Bancomer

4.013

48.15

3.242

38.9

3.208

38.5

2.947

35.362

3.224

38.679

3.296

39.555

Bancreser

3.6

43.2

3.3

36.3

2.89

34.68

2.912

34.940

2.989

34.96

2.913

34.960

Banjercito

3.75

45

3.083

37

2.819

33.83

2.667

32.000

2.917

35

2.917

35

Banorte

3.729

44.75

2.083

25

2.083

25

2.083

25.000

2.523

25.98

2.115

34.900

Bilbao Vizcaya

2.9526

35.432

3.242

38.9

3.208

38.5

2.947

35.362

 

 

 

 

Bital

3.75

45

2.917

35

2.992

35.9

2.908

34.900

3.045

25.114

2.968

25.037

Citibank

4.1

49.2

3.3

39.6

3.3

39.6

3.300

39.600

3.512

40.132

3.311

39.600

Dinners Club

3.3

39.6

3.96

47.52

3.75

45

3.75

45

 

 

 

 

Industrial

3.729

44.75

2.417

29

 

 

 

 

 

 

 

 

Scotia Inverlat

3.27

39.24

2.961

35.538

2.798

33.573

2.591

31.091

3.074

36.893

3.075

36.894

Ixe

2.9665

35.6

2.455

29.46

2.798

33.573

1.997

23.96

2.111

25.33

2.283

27.39

Santander

3.27

39.24

2.719

32.63

2.563

30.76

2.727

32.718

2.994

35.456

2.887

34.646

Serfin

3.98

47.76

3.271

35.4375

3.222

35.328

2.959

32.718

3.197

32.971

3.045

33.258

a Rates without VAT

Image

Fig. 6.11 Dominance indexes credit cards (Authors' calculations based on information from CNBV, Banks' financial statements)

which, furthermore, raises the cost of the same. For instance, Liverpool's housing and automotive credit is backed by BBVA Bancomer, its Famsa life insurance is operated by Afirme and Viajes; Wall-Mart handles the Barceló agency.

Although concentration as such is not a sign of lack of competition, it is a phenomenon that merits investigation by the competition agency in order to determine whether concentration exists in that credit card relevant market. A series of indexes (HHI and DI) were obtained for earnings from consumer interests, as shown in the last series of bars in Fig. 6.11. This market, in comparison to other relevant markets, is one of the most concentrated, as it reaches levels of over 0.5 points, which suggest a sign of alert for CFC.

This also applies to other financial products in which intermediaries enjoy some kind of legal protection, i.e. barriers to entry. For example, in addition to the above-mentioned issuance of credit cards, the issuance of window cheques, payment orders, etc., in other countries is also more competitive by permitting access to other non-financial entities. These are areas in which it would be advisable to delve into further analysis in the future.

Much the same applies to minimum amounts for gaining access to the purchase of stock certificates. In Mexico, most financial institutions require minimum purchase amounts for these instruments, which range between 300,000 to 500,000 pesos. Small savers with lower amounts are compelled to purchase mutual funds with high fees and commissions charged. What also contrasts with Mexico's case is that in the USA one can buy US stock certificates directly without the need for a financial intermediary's charging a commission, on web page (www.savingsbonds.gov) with amounts upwards of US$1,000 (even for people from third countries as long as they possess a dollar account in that country). In this regard it would be interesting to

Image

Fig. 6.12 Average mortgage credit rates (%) (INFOSEL financiero)

study the rationale behind Mexico's current provisions, for they may be operating as barriers to entry into this relevant market.

The above may be exemplified by what happens in the automotive and mortgage credit market. Until recently, these sectors fell within the realm of the business of banking in Mexico. As of 1996, when the entry of Sofoles became generalized in this industry, bank interest rates have dropped considerably under these items, in other words, competition with other financial intermediaries compelled the banking system to reduce its profit margins. Figure 6.12 shows average rates for mortgage loans for the period 2000–2005. Clearly, the rates have dropped over time, due mainly to increased competition among the banks and also between various financial intermediaries, Sofoles being a case in point.31

A similar situation could be experienced in other products, especially those related to the payments system, which is where the banks are overprotected. Thus, the credit card, payment orders, window cheque segments, etc. could find themselves open to other participants, whether bank or non-bank. This would definitely reduce those products' margins in benefit of consumers and financial intermediation itself.

This could occur where other products are concerned, particularly those related to the payments system, which is where the banks enjoy excessive protection. Thus, credit cards, payment orders, window cheques, etc., could be opened up to non-bank and even non-financial intermediaries. This would definitely reduce the margins for these products for the benefit of consumers and financial intermediation itself.

To sum up, if we add earnings from fees and commissions and from interests in the USA, we find that the consumer surplus dropped slightly or did not change during the period 1986–2002 due to cross-subsidies between the two components.

31 It is true that this is partly the result of the country's greater economic stability, but the same conditions are valid in the case of credit cards, and the latter have not experienced the sale rate reduction.

By contrast, if we add these two items for Mexico, the consumer excess diminishes without any ambiguity. It is important to understand that if the fees and commissions are excessive; this could be negatively affecting financial intermediation and in fact, might be a factor that explains the financial sector's limited depth in the country.

Either way, a study of vertical – not merely horizontal – integration could also shed light on competition in the sector, and not only the portion of the sector covered by banking, but that lies beyond the scope of this study.

4. Epilogue

After this study was concluded, Secretaría de Hacienda y Crédito Público (SHCP) and Comisión Nacional Bancaria y de Valores (CNBV) approved the establishment of 13 new banks. Most of these included banks belonging to the popular commercial sector, although some are patrimonial banks of considerable international standing (see Table 6.9); on principle, this would make the market more competitive. One aspect that the country's financial authorities should heed is the fact that increasing the number of banks would not necessarily promote competition among the financial intermediaries. Thus far, it would be difficult to infer any results from such a phenomenon, but it may be that by constituting themselves as banks in order to compete, the transaction costs would be too high, and therefore any new banking entities would perhaps "adapt themselves" to the high current margins, without really entering into real competition with the existing banks.

Table 6.9 New banks, 2007 (CNBV)

New banks

Type of business

Niche/slot

Banco Autofin

Automobile distribution

Intermediate level

Barclays Bank, Mexico

Foreign bank and stock exchange

Corporate and high levels

Banco Ahorro Famsa

322 furniture stores in Mexico and 15 in the USA

Intermediate

Banco Monex

Stock exchange and Afores

Intermediate level and high levels

Banco Multiva

Stock exchange and Afores

Intermediate level and high levels

Banco Wall-Mart of Mexico

Supermarkets

Popular level

Bancofacil

Chedraui and Sherman Financial Group

Popular level

Bancoppel

Furnitures

Popular level

Banco Comercial del Noreste

Banco Amigo

Intermediate level

Prudential Bank

Stock exchange and Afores

Corporate and high levels

Banco Compartamos

Savings and micro-loans

Micro-loans

Banco Regional

Banregio

Intermediate level and high levels

UBS Bank Mexico

Foreign bank UBS

Corporate and high levels

III. Conclusions

This study analyses the status of bank competition in Mexico. As we have discussed throughout the paper, this is a difficult task as the very process of intermediation is a complex one. Morgan (2002), for example, argues that a country's most opaque sector is in fact banking. This is reflected by the work put forward here.

We can conclude that the traditional approach to analysis of structure and concentration, although useful in putting out valuable information, does not necessarily involve an analysis of competition. As an alternative, then, we have made use of the theory of contestable markets in different relevant markets throughout this study and reached the conclusion that Mexico needs increased competition in some of these markets to enable more companies to enter the market. The study includes an analysis of bank fees and commissions from the viewpoint of contestable markets. Furthermore, we argue that there is a future need to expand on the subject by means of studies embracing all non-bank and even non-financial competitors in the financial market, within the sphere of a multi-product industry covering different relevant markets.

Looked at in this way, the chapter suggests that the theory of market contestability may be a very useful tool in ordering, classifying and implementing competition policy and economic regulation in the Mexican banking sector. In particular, as regards competition-policy recommendations, the study proposes it should focus on dismantling entry barriers in the relevant markets where the banking system faces scarce competition – or none at all – from other financial and non-financial agents. One case in point is credit cards, the analysis clearly showing that there is evidence of market power in the said relevant market. Although a credit card market in which chain stores participate does exist, the latter does not constitute a perfect substitute for the credit cards issued by banks. Therefore, CFC should examine the impact on consumer well-being this could have in the credit card market.

The above aspect may easily be extrapolated to other financial products in which the banks enjoy full protection (legal barriers to entry). This occurs with issuance of window cheques, payment orders, or the acquisition of some types of stock certificates. There is no doubt that the entry of new non-bank and even non-financial agents in the above-mentioned market segments would reduce financial margins and thereby benefit consumers.

Bibliography

Andrade, G., M. Mitchell, and E. Stafford, (2001), "New Evidence and Perspectives on Mergers" Journal of Economics Perspectives, vol. 15, N°. 2..

Allen, F., A. Santomero, (2001), "What Do Financial Intermediaries Do?", Journal of Banking and Finance, vol. 25, N° 2, pp. 271–94.

Allen, F., A. Santomero (1998), "The Theory of Financial Intermediation", Journal of Banking and Finance, vol. 21, N° 11–12, pp. 1461–85.

Arellano, R., G. Castañeda, F. Hernández (1993), "La fragilidad del Sistema Financiero y sus Implicaciones Sobre la Cuenta Corriente", Economía Mexicana, New Epoch, Second Semester, vol. II, N° 2.

Avalos, Marcos (2006), Condiciones Generales de Competencia: el Caso de México, CEPAL, serie Estudios y Perspectivas, N°. 48,ECLAC/IDRC, Mexico, May.

Avalos, B.M., N.J. Melgoza (2002), "Concentraciones Internacionales y Aplicación de Leyes de Competencia", Gaceta de Competencia Económica, Comisión Federal de Competencia, year 5, N° 12, pp. 17–51., Mexico.

Bailey, E.E. (1981), "Contestability and the Design of Regulatory and Antitrust Policy", The American Economic Review, vol. 71, N° 2, pp. 178–183.

Barth J., G. Caprio, D. Nolle (2004), "Comparative International Characteristics of Banking". Economic and Policy Analysis Working Paper 2004-1, Auburn University, United States.

Barth, J., D. Nolle, T. Rice (2000), "Commercial Banking Structure, Regulation, and Performance: An International Comparison, "Modernizing Financial Systems, D. B. Papadimitriou, editors, Macmillan Press and St. Martin's Press.

Baumol W., J. Panza, R. Willig (1982), The Theory of Contestable Markets, John Wiley, United States.

Beck, T., D. Kunt, V. Maksimovic (2003), "Bank competition, financing constraints and access to credit", World Bank, [online] (www.worldbank.org).

Berger, A., T. Hannan (1989), "The price-concentration relationship in banking", Review of Economics and Statistics, 74(2): 376–79

Bhattacharya, S., A. Thakor (1993), "Contemporary banking theory". Journal of Financial Intermediation, 3, pp. 2–50.

Calvo, G., E. Mendoza (1996), "Mexico's Balance-of payments Crises: a Chronicle of a Death Foretold", Journal of International Economics, 41(3-4): 235–64.

Chávez-Presa, Jorge (1990). Scale and Scope Economies in the Mexican Banking System. PhD Thesis, The Ohio State University. Columbus, Ohio, United States.

Church, J.R., R. Ware (2000), Industrial Organization: A Strategic Approach, McGraw Hill, United States.

Claessens, S., A. Demigrut-Kunt, J. Huizinga (2001), How does foreign entry affect domestic banking markets?. Journal of Banking and Finance N° 25, pp. 891–911.

Claessens S., L. Laeven (2003), What Drives Bank Competition? Some International Evidence, World Bank and Federal Reserve Bank of Cleveland Conference on Bank Competition, Washington D.C.

Comisión Nacional Bancaria y de Valores, Boletín Estadístico de Banca Múltiple, several issues.

Comisión Nacional para la Protección y Defensa de los Usuarios de los Servicios Financieros (CONDUSEF) (2004), "Comisiones Bancarias", mimeo.

Del Angel, C. Gustavo, C. Bazdresh and F. Suárez (2005), Cuando el Estado se hizo banquero. Fondo de Cultura Económica, Mexico.

García, A. (1990) Un enfoque para medir la Concentración Industrial y su Aplicación para el Caso de México. El Trimestre Económico vol. 2, N°. 57, pp. 317–341.

García Rocha, A. (2004), "La Competencia en los Mercados Financieros", Competencia Económica en México, CFC, Porrúa-Comisión Federal de Competencia, Mexico.

Hernández Trillo, F. (2003), La Economía de la Deuda: Lecciones desde México, Fondo de Cultura Económica, Mexico.

Hernández Trillo, F., A. Villagómez (2000), "La Estructura de la Deuda en México", BID, Documento de Trabajo 123, Washington D.C.

Kaminsky, G., C. Reinhart (1996), The Twin Crises: The Causes of Banking and BOP Problems, Paper prepared for the IDB conference on Speculative Attacks in the Era of the Global Economy, Washington D.C.

Kaminsky, G., S. Lizondo and Carmen Reinhart (1997), "Leading Indicators of Currency Crises", IMF Working Paper 97/79, Washington D.C.

Kane, E. (1989), "Changing incentives facing financial financial-services regulators". Journal of Financial Service Research, 2(3): 265–74

_____ (1984), "Technological and regulatory forces in the developing fusion of financial services competition", Journal of Finance, 39(3): 759–72

_____ (1981), "Accelerating inflation, technological innovation and the decreasing effectiveness of banking regulation". Journal of Finance, 36(2): 355–67

La Porta, R., F. López de Silanes and G. Zamarripa (2002), "Related Lending", Yale University, mimeo.

Levy, S. (2005), Ensayos sobre desarrollo económico mexicano, Fondo de Cultura Económica, Mexico.

McKinnon, R., H. Pill (1995), Credible Liberalizations and International Capital Flows: The Overborrowing Syndrome, Stanford University, United States.

Miles, D. (1996), The future of savings and wealth accumulation: differences within the developed economies, Global Securities Research and Economics Group, Merrill Lynch, London, [online] (www.ml.com).

Morgan, D. (2002), "The Opacity of Banks". American Economic Review, vol. 92, N° 4.

Murillo, J. A. (2005), "La Banca después de la privatización", Del Angel, G., C. Bazdresh and F. Suárez, Cuando el Estado se hizo banquero, Fondo de Cultura Económica, Mexico.

OCDE (2002). Country Surveys, Mexico. Paris, France.

Rojas, M. (1997), "Competencia por Clientes en la Industria Bancaria de México", El Trimestre Económico, 64(1): 47–73

Sachs, J., A. Tornell, A. Velasco (1996). "The Mexican Peso Crisis: Sudden Death or Death Foretold?," NBER Working Papers 5563, National Bureau of Economic Research, Inc.

Schmidt RH, A Hackethal, M. Tyrell (1999), "Disintermidiation and the role of banks in Europe: an international comparison", Journal of Financial Intermediation, 8(1–2): 36–67

The Economist (2005), October 29, p. 76.

Thakor, A. V. (1992). Maturity Transformation, In "The New Palgrave Dictionary of Money and Finance". pp 678-680 MacMillan Press, London.

Tirole, J. (1997), The Theory of Industrial Organization, The MIT Press, Cambridge, Massachusetts, United States.

Chapter 7
Findings in an International Perspective

Simon J. Evenett and Claudia Schatan

Introduction

The purpose of this concluding chapter is to summarize and interpret the main findings for Central America and for Mexico of the research conducted for this volume and, more importantly, to relate those findings to the experiences of nascent and relatively newer competition agencies in other parts of the world. In principle, the following discussion may serve a number of useful purposes. First, it may help relate the Central American and Mexican experience to the broader dynamics that are affecting the intensity of competition in developing country markets more generally. For example, some of the important policy measures analysed in prior chapters, such as privatization of state-owned enterprises, have been implemented elsewhere in the developing world, notably in Central and Eastern Europe, Central Asia, and some parts of Africa.

A second potential contribution of this approach is to establish just how distinctive the Central American and Mexican experience is. The former are generally characterized as small economies and, given the belief of some experts that competition law and its enforcement should be different in these economies, it will be interesting to see just how distinctive the Central American enforcement experience is.

Comparisons with third countries may serve another purpose: the identification of alternative policy prescriptions that might be fruitfully considered by policymakers in the Central American region and in Mexico. Of course the policy lessons may flow in the opposite direction. Care is required here, not least because arguably it was the inappropriate transfer of certain policy reforms to the Central American region and to the Mexican economy that created many of these nations' current competition-related challenges in the first place.

The remainder of this chapter is organized into three parts. The first part summarizes the findings from this research project concerning the enactment and implementation of competition law in recent years in Central American and Mexican markets. The second part considers the findings from the analyses presented earlier of two sectors of development significance, namely, telecommunications and banking. In the third part, some concluding remarks are made and new outstanding matters highlighted.

1. The Enactment and Enforcement of Competition Laws

In recent years, there have been a number of potentially significant developments concerning competition laws and their enforcement in the Central American region and in Mexico. Between 2004 and 2006, El Salvador, Honduras and Nicaragua enacted competition laws for the first time and successfully set up enforcement agencies. What is particularly promising is that the laws and agencies in these two countries appear on paper at least to be stronger than others in the region (i.e. fewer exceptions from competition law were made; the maximum levels of fines that can be imposed are higher, etc.). This may reflect learning on the part of certain policymakers in the region and, if so, represents a positive development.

Three other jurisdictions (Costa Rica, Mexico, and Panama) have taken steps to revise or strengthen their competition law frameworks. Mexico, for example, enacted a new competition law in 2006. By and large, these initiatives have seen the ambit and tools available to the competition agencies clarified and improved.

Of all the countries studied in this volume, Guatemala has yet to enact and enforce a competition law. This reflects pronounced opposition to such legislation in that country and is a reminder of the importance of political economy factors in determining both the likelihood and terms of the enactment of competition law; a point we return to soon below. Overall, though, in the Central American region and in Mexico many political economy constraints appear to have been overcome and independent enforcement agencies, staffed by professional lawyers, economists and investigators and with clearly defined responsibilities, are now the norms.

The experience of other developing countries highlights the relatively supportive political environment confronting competition agencies in Central America. There are, for example, no cases like Thailand where the competition agency created remains both under direct ministerial control and where the overseeing commission is comprised of influential businessmen. This combination of outcomes was deliberately designed to ensure that the competition law is not enforced – in fact, the Thai competition agency has yet to issue certain fundamental guidelines or regulations that will enable it to enforce selected aspects of the competition law in the first place. Closer to Central America, in recent years, Peru's competition agency INDECOPI appears to have responded to political pressures and focused on consumer protection matters, effectively demoting the priority given to competition law enforcement. This latter example highlights the importance of national political support for sustained and effective competition law enforcement.

The studies in earlier chapters also reported that competition agencies in the Central American region and in Mexico were not independent of their respective national governments in every possible respect. Specific mention was made of the manner in which the competition budget is set and any interference concerning the use of the associated funds. Matters related to independence have been a significant concern in some other parts of the world too. At the least there is a potential tension between the effectiveness of enforcement (which is thought to be bolstered by various aspects of agency independence) and accountability (which requires political oversight of government officials). This tension exists in most democracies for good reasons and in many areas of regulation, and its existence suggests the (political) implausibility of advocating full independence for competition agencies. Yet is full independence necessary? Some competition law enforcement agencies, such as the Antitrust Division of the US Department of Justice, have tough enforcement records despite being part of a central government ministry.1 Other agencies are much "more" independent in a formal institutional sense, yet their enforcement records are much weaker. The relevant question ought to be what types of independence really matter? In recent years, some experts have come to believe that an independent agency is one with the power to open and close the file on a potential violation of the competition law (which could involve an investigation that may or may not result in an enforcement action) on its own terms, i.e. without political interference. On this view the ability to successfully enforce the law unhindered by political considerations is what generates private sector expectations about the predictability of enforcement and the deterrent effects of such enforcement. If this is the right perspective, then many of the competition agencies in Central America may already have the requisite form of independence.

Another important finding of earlier chapters, which recurs in other regions of the developing world, is that the anti-competitive acts facing a competition agency are diverse in nature.2 Horizontal and vertical agreements, abuse of market power, and trans-national anti-competitive practices have been found to be important in Central America and Mexico. This finding does not suggest that each competition agency faces exactly the same caseload. Nevertheless, other implications do follow. First, advice to focus on one type of competition enforcement action (notably cartels) is hard to reconcile immediately with the multiple perceived threats to competitive market outcomes in a national economy. Second, advice to nascent agencies to focus on competition advocacy seems particularly perverse in the face of significant number of alleged anti-competitive practices.

Third, the very fact that a number of anti-competitive acts will have to be investigated and potentially prosecuted implies that competition agencies face no alternative in the longer term but to build systematically expertise and capacity across the board. Internal training programmes and technical assistance received should be directed towards this goal. This recommendation does not sit well with one finding of the studies presented earlier in this volume.3 Those studies pointed out how

1 Even so the question does arise as to whether this agency's enforcement record would have been even better had it been established with comparable powers and budget outside a government ministry.

2 Evenett et al. (2006) presented similar evidence of a broad range of anti-competitive practices alleged to have taken place in sub-Saharan African countries. Clarke et al. (2005) report similar findings for Latin America and the Caribbean. Clarke et al. (2006) undertook a similar analysis for a broader sample of developing countries and confirmed the findings of the earlier two studies mentioned directly above.

3 A distinction ought to be made between the capacity of a competition agency to tackle a wide range of potential anti-competitive practices and the actual caseload of a competition agency over a given period of time. It may be that the latter is markedly affected by factors other than the former. For example, during a merger wave a competition agency may feel it has little choice but to allocate plenty of internal resources to merger reviews.

resource constraints were hampering enforcement activity (as in Central America) and that the better resourced Mexican agency specialized in one or two enforcement activities, notably merger review. Taken together, these findings raise the question as to what, if any, should be the appropriate enforcement priorities for different classes of developing countries. We do not presume that evidence on the prevalence of different anti-competitive acts alone settles this important question, even so competition agencies may wish from time to time to revisit the assumptions underlying their chosen enforcement priorities.

Many discussions of the impact of competition legislation on economic performance identify a nation's so-called competition culture as an important conditioning variable. Typically it is argued that a strong competition culture, adhered to by a large proportion of the population, can be the result of competition advocacy and may well reduce the need for certain enforcement actions. The experience of Costa Rica, however, identified a quite distinct relationship between national competition culture and enforcement outcomes. The absence of the former, it was argued, has led cartels to operate relatively openly in a fashion that has, perversely from the point of view of the firms involved, made competition law enforcement easier! How long it will be before those engaged in anti-competitive practices revisit the manner in which any agreements are concluded and maintained is another matter. Surely effective enforcement activity, including the imposition of substantial fines, will encourage those firms that engage in anti-competitive acts to take steps which will make the detection and prosecution of their illicit acts more difficult for the competition agencies. Greater efforts to hide evidence and resort in order to not formalized or implicit agreements between firms are likely to follow. Even so, the Costa Rican experience suggests that there may be some "low hanging" fruit for a nascent competition agency to pick and, in so doing, to help to build its reputation for seriousness as well as strengthening the deterrent value of the competition law.

A challenge faced by many competition agencies examined in this study was the obstacles deliberately or inadvertently erected by the local judiciary. Laws on economic matters are, of course, not new. However, judges may not be particularly well versed in competition law and the appropriate role that economics can play in enforcement cases and in helping to resolve matters in court. This challenge is arguably universal, i.e. not confined to developing countries. Competition agencies in many industrialized countries devote time and resources for meeting with and informing judges. Similar seminars are held in developing countries too. Judges, often being inherently conservative in nature, may well put some weight on the credibility of the competition agency and its track record. Reputation building could be very important in this regard.

One reaction to the difficulties experienced with judges is to consider adopting an administrative system of competition law enforcement, as is the case for example in the European Union. However, this option is not without its drawbacks. First of all, unless carefully designed an administrative law system may essentially result in the competition agency appearing to be both "judge and jury", which goes against what many view as a basic principle of justice. Second, most administrative law systems have some judicial oversight and so the need to inform judges about competition law still arises. Moreover, as European Union's experience at the beginning of this decade has shown, judicial oversight can be particularly tough even in an administrative law system. In sum, judges are an important constituency whose cooperation and understanding are essential for the smooth functioning of national competition law and competition agencies should recognize this fact and act accordingly.

Experience in Central America and Mexico also highlights the low level of fines generally imposed on violators of competition law. While this is discouraging, it is certainly not a feature unique to this region. For example, the total fines imposed on the corporate members of the international vitamins cartel, which distorted 12 vitamins markets worldwide for nearly 10 years, totaled approximately US$2 billion. Yet the estimated total overcharges made by the cartel in a group of 90 jurisdictions (which excluded the world's largest economy, the USA) exceed US$2.7 billion,4 suggesting that in aggregate the firms associated with this cartel probably benefited from their illegal acts. More generally from time to time, the OECD secretariat collects evidence on the magnitude of fines imposed on cartels and compares these fines with estimates of the gains to the cartel members. Depressingly, in many enforcement actions the level of imposed fines remains low, in particular when contrasted with the estimated gains and this finding calls into question whether current fining practice constitutes an effective deterrent to cartelization in many jurisdictions. Competition agencies and national policymakers should give further consideration to the adequacy of existing fines and to whether fines ought to be supplemented by other sanctions for anti-competitive acts, including incarceration.

Recent Central American and Mexican experience replicates another challenge facing many competition agencies worldwide, namely, the relationship between competition legislation and its enforcement and sectoral regulation. As is well known, competition agencies have typically come later to the field of economic regulation and it should not be surprising that some existing government agencies and ministries have, or claim to have, powers whose execution impinges one way or the other on the effectiveness of competition law. Aware of the potential future encroachment on "their turf", many established government bodies were often able to get whole economic sectors or activities exempted from the ambit of competition law. Others have found that they can ignore the recommendations of the competition agency at little or no cost.

In principle, in many cases there need not be any tension between a competition agency (with a pro-efficiency mandate) and a sectoral regulator (whose statutory goals are often non-economic in nature). Indeed, a competition agency may have some suggestions as to how the sectoral regulator might be able to meet its legal mandate with less distortion to competitive processes and ultimately to market outcomes. This beneficial relationship, alas, rarely comes to pass, not least because of factors which rose at the end of the last paragraph. What can be done to improve

4 For details see Clarke and Evenett (2003).

matters? One option is to legally mandate an institutional mechanism which effectively gives voice and clout to the competition agency in the regulatory process. For example, a competition agency may be given the right to comment on proposed regulation and legislation before the latter is implemented and, going further, an obligation on the regulator to explain publicly why it rejected any advice from a competition agency could be created.

At present very few nations, with Korea being perhaps the only serious example, give their competition agencies the voice and clout to effectively advance proposals for pro-competitive regulations. Without such clout the question arises (to paraphrase Stalin's inquiry about the Pope in World War II) "How many divisions does the competition agency have?" In short, even if the competition agency's arguments are correct, why should another government agency (or indeed politician for that matter) privilege those ideas over others that have the backing of more influential interests? Still, one should not get too depressed here as there have been some examples where competition agencies have claimed to have influenced the design and implementation of state-imposed regulation. What policymakers and researchers need to discern is whether these examples are extraordinary exceptions or whether any replicable lessons can be distilled from them.

In this section we have attempted to relate where possible many of the characteristics, challenges, and outcomes associated with competition agencies in Central America and Mexico to comparable features in the experiences of other jurisdictions, in particular those in developing countries. In this section we have attempted to relate where possible many of the characteristics, challenges, and outcomes associated with competition agencies in Central America and Mexico to comparable features in the experiences of other jurisdictions, in particular those in developing countries. Notwithstanding that Central American countries and Mexico have some features of their own regarding competition law-related experiences; arguing that they are particularly distinctive would be a little overstretched. The challenges they face are in many ways similar to those of many other countries and, therefore, further opportunities for cross-country learning and the exchange of experiences should be tapped.

2. Promoting Competition in the Telecommunications and Banking Sectors

If mines and ports represented the foundations of many developing countries' engagement with the world trading system in the early post-war era, then banks and telecommunications are two of their modern day counterparts. Both of the latter sectors influence, sometimes significantly, the cost of, and the time in, which local businesses can respond to the opportunities created by national reforms and globalization. Unreliable and expensive fixed telephone lines, cell phones, or Internet connections can seriously disrupt supply chains, preventing companies from developing countries from reaching even the first step on the ladder of their sector's cutting-edge international production networks. Inefficient and unresponsive banks can add substantial transaction costs to domestic and international commerce. Moreover, very poorly regulated banks can trigger financial panics and crises of confidence – all of which undermines support for markets and market reforms. The studies in this volume take these two sectors potential contributions to development as given and consider a slightly narrow matter: what role can competition principles and competition law play in the overall regulation of these sectors? Again specific reference was made to the experiences of nations and firms in the Central American region and in Mexico.

Interestingly the reason why competition principles have gained some prominence in discussions about policies towards these two sectors was the latter's prior unsatisfactory performance. In the telecommunications sector privatization and other sectoral reforms were supposed to result in a substantial increase in the quality of services, along with a reduction in the prices charged to all end users (and not just to large commercial buyers of telephony services). In banking the combination of a number of financial crises as well as a gap between national banking performance and internationally agreed standards triggered a rethink by some policymakers about the optimal mix of different types of regulation and competition. Digging deeper, failures in sectoral regulation appear to underlie the unsatisfactory outcomes in both sectors and a challenge facing policymakers and competition agencies is to identify which, if any, elements of current sectoral regulation should be retained.

The experiences of other regions, notably in East Asia, suggest that resort to competition principles in a second (or a third) round of regulatory reform should not be taken for granted. The question should arise as to why those principles were ignored in the initial round of reform. Strong commercial and often bureaucratic interests were often able to persuade policymakers to undertake liberalization (whether in the form of privatization or the removal of restrictions on bank activities) while arguing against taking additional measures to promote competition in the relevant markets that could, in certain circumstances, call for state intervention. Non-intervention was the mantra of the day. (Indeed one can imagine an impatient policymaker asking a proponent of a pro-competitive regulatory framework: "Well, do markets work or not? If so, then why bother with back-up regulations to cope with XXX problems?" Pithy answers to this line of questioning may not have been easy to come by.)

The supplementary question to ask is whether the unsatisfactory performance after the first round of reforms has sufficiently weakened those forces that initially opposed the potential introduction of competition-promoting state measures? If those forces still remain strong then there must surely be some doubt as to what a second round of potentially pro-competitive regulation will accomplish. These observations highlight the enduring importance of the relative strength of different political economy actors in the design of economic reforms and not just in their implementation.

With respect to telecommunications, a current and likely recurring problem in Central America and Mexico concerns the behaviour of privatized incumbent firms. In the landline market, for example, the incumbents in Mexico, Nicaragua, and Panama have been given a certain period of time to enjoy the exclusive (i.e. monopoly) right to supply their respective national markets. It was said by policymakers that this interval would be used to enable the incumbents to raise their competitiveness so that they could eventually be able to withstand and take on new competitors. These measures are tantamount to the infant industry protection measures, whose track record in Latin America and elsewhere has been disappointing. Worse, once competition is allowed, incumbent firms may be able to discourage new rivals by charging large fees to the latter for the use of the former's network infrastructure. The associated investigations into such interconnection charges tend to be very detailed (placing considerable demands on the competition agency) and controversial. Moreover, a strengthened private sector incumbent may also develop the political clout to delay, and ultimately to frustrate, the introduction of new competition. Competition agencies seeking to advance competition principles in the telecommunications sectors need to be cognizant of the incumbent firms' strategies towards lobbying and rivals.

Since the design of privatization schemes appears to critically define post-implementation incumbent performance, it should not be surprising that many have argued that competition advocacy could play a helpful role at this stage. Advocating for measures to simultaneously privatize and liberalize entry conditions in telecommunications is one recommended option. (Indeed, according to some of the studies presented in this volume such an option has delivered good results in El Salvador and in Guatemala.) A second option is to ensure that there is a clear, predictable post-privatization pro-competition enforcement regime with sufficient investigative powers and capacity to sanction. In its competition advocacy on such matters competition agencies may want to identify allies inside and outside government that are committed to seeking more pro-competitive post-privatization outcomes.

The promotion of entry by new suppliers has yielded substantial benefits in telecommunications. Nowhere is this clearer than in mobile telephony. The ability of new entrants to provide related services without having to use or invest in the incumbent's existing infrastructure has effectively reduced an important barrier to entry. The impact on prices, access (especially to the poor), and innovation has been impressive. The lessons for policymakers are not just to oppose barriers to entry but also to be vigilant about any rearguard anti-competitive actions that the incumbent firms may take.

Turning now to the banking system in Central America and Mexico it is important to appreciate the diversity of experience among these countries in terms of regulatory structures and outcomes. Costa Rica, for example, exempted its banking sector from the application of its national competition law, whereas Panama long ago stripped away barriers to entry into banking and has developed an internationally competitive banking sector with appropriate regulatory oversight. Some countries in this region have suffered from financial crises, but not all.

An interesting difference between the Central American and Mexico banking system concerns the manner in which international factors are altering the industrial organization of respective national banking sectors. In Mexico, for example, many foreign banks have taken over domestic banks but the degree of inter-firm rivalry appears little unchanged. (Earlier in this volume it was argued that Mexico's competition agency has done little to effectively alter these outcomes.) In contrast, Central American banks have expanded into their neighbours' economies and competition has risen accordingly, though recently large international banks have begun to enter the region too. Available choices have grown and downward pressure on commissions has intensified. It will be interesting to see, as happened in Mexico (and examined in this book) and in Europe, whether the greater expansion of banks eventually leads to cross-border consolidation, a reduction in inter-firm rivalry, and an increase in spreads between deposit and loan interest rates.5 Should such consolidation start to occur also in Central America, competition agencies in the region should be particularly vigilant and take a more active role in the financial sector than they have had up to now.

3. Concluding Remarks

In this sector we discuss two recurring and important themes that have been discussed in many places throughout this book. The first is that competition agencies should not take as given either both market outcomes and industrial organizations they face or the policy environment confronting current and potential suppliers. Many authors have stressed the importance of so-called competition advocacy, the support for state measures (including, in some cases, decisions to liberalize and decisions not to intervene) that promote competition between firms and more efficient market outcomes. While it is important to recognize that advocacy can complement the enforcement activities of a competition agency, this volume's studies do not support the notion that advocacy can entirely substitute for the later.

Yet, arguing that there is a role for advocacy is not to suggest that it is necessarily easy or effective. Indeed, one of the outstanding questions at the end of this study is to better understand the circumstances under which advocacy by competition agencies makes a long-term contribution to promoting markets. Making progress on this question requires thinking through the potential role and consequences of the competition agency as an actor in national political processes, and it is not just the technocratic matter of getting the policy recommendations and analysis correct. Part of this discussion could also include an evaluation of the relative merits of introducing government mechanisms to give competition agency's the right to comment, possibly to redraft, and perhaps even to veto, suggestions for new societal regulations.

A second theme in this book is the growing importance of regional factors in shaping international deliberations and outcomes on competition-related matters. On a number of occasions it was argued that Mexican experience provided useful pointers to countries in the Central American region. For Mexico, having a cooperation agreement with the USA on competition matters has probably helped. A similar

5 For evidence on the later in Europe see Evenett (2003a).

cooperation between all Central American countries and the USA might be very useful too. Better understanding of the mechanisms by which policymakers and competition experts learn from other countries' experiences and which, if any, international fora actually facilitate such learning would be useful. This is in addition to the processes developing between competition agencies that promote the exchange of information about investigations and associated anti-competitive acts and strong technical knowledge and best practices. Much more work needs to be done here by competition agencies and researchers alike, to understand how best to strengthen the appropriate enforcement of competition law by relatively smaller nations operating in an increasingly globalized economy.

Bibliography

Clarke J. and S. Evenett (2003), The Deterrent Effect of National Anti-Cartel Laws: Evidence from the International Vitamins Cartel. Published in Antitrust Bulletin.

Clarke, J., S. Evenett, and K. Lucenti (2005), Anticompetitive Practices and Liberalising Markets in Latin America and the Caribbean. Published in The World Economy. July 2005

Clarke, J., S. Evenett, and K. Lucenti (2006), Extracting the Most Out of Trade. Investment and Regulatory Reforms: What Role for Competition? Working paper. University of St. Gallen.

Evenett, S. (2003a), The cross border mergers and acquisitions wave of the late 1990s, in R. E. Baldwin and L. Alan Winters (eds). Challenges to Globalization. University of Chicago Press for the National Bureau of Economic Research.

Evenett, S. (2003b), Links between development and competition law in developing countries, October [online] (http://web.idrc.ca/uploads/user-S/10710933481Evenett_paper.doc).

Evenett, S. (2003c), Promoting competitive markets in developing economies: details of the research project: section C of the application for an IDRC grant, February, [online] (http://www.idrc.ca/en/ev-32149-201-1-DO_TOPIC.html).

Evenett, S. (2003d), Study on issues relating to a possible multilateral framework on competition policy, Working Group on the Interaction between Trade and Competition Policy, World Trade Organization (WTO) WT/WGTCP/W/228.

Evenett, S. (2006a), Competition in developing countries other than Latin America: lessons to be learnt?, PowerPoint presentation [online] (http://www.eclac.cl/mexico/competencia/index.html).

Evenett, S. (2006b), The effectiveness of technical assistance, socio-economic development and the absorptive capacity of competition authorities, serie Estudios y Perspectivas N° 60, ECLAC/IDRC, Mexico, October.

Evenett, S. (w/d), Competition policy in small developing economies: reflections on current literature, PowerPoint presentation.

Evenett, S., F. Jenny, and M. Meier (2006), A Database of Allegations of Private Anti-Competitive Acts in Sub-Saharan Africa. Presented at the IDRC-sponsored pre-ICN conference, Cape Town, May, 2006.

Segovia, A. (2006), Integración real y grupos centroamericanos de poder económico. Implicaciones para la democracia y el desarrollo regional, ECA, estudios centroamericanos, ISSN 0014–1445, N°. 691–692, pp. 517–582.

Silva, V. and A. M. Alvarez (2006), Cooperación en políticas de competencia y acuerdos comerciales de América Latina y El Caribe: Desarrollo y perspectivas, Serie de Comercio Internacional N° 73, ECLAC, Santiago, Chile.

Stewart, T. (2005), Special cooperation provisions on competition law and policy: the case of small economies, in P. Brusick, A. M. Álvarez, and L. Cernat (Eds.), Competition Provisions in Regional Trade Agreements: How to assure development gains, United Nations Conference on Trade and Development (UNCTAD), New York and Geneva.

United Nations Conference on Trade and Development (UNCTAD) (2002), Competition, competition policy and economic development: some thoughts about their relations and the implications for competition policy in developing countries, presentation of the Mexican Delegation to the Global Competition Forum, Paris, France, February 14 and 15.

United Nations Conference on Trade and Development (UNCTAD) (2004), Model Law on Competition, [online] (r0.unctad.org/en/subsites/cpolicy/docs/Modelaw04.pdf).

This page intentionally left blank

Index

A

Absolute monopolistic practices, 23, 27, 34, 146–162, 174

Abuse of a dominant position, 23, 174–176

Accountability, 224

Agency of International Development, 114

Airlines, 40–42, 149–153, 158, 162, 210

America Movil, 3, 56, 73, 76–79, 81–83

Amparo, 14, 27, 37, 38, 66, 70

Andean Community, 4

Anti-competitive practices, v, 1, 4, 8, 10, 13, 14, 16, 22, 23, 28, 33, 34, 36, 37, 39, 40, 42, 63, 146, 153, 158, 161, 225, 226

Anti-trust, v, vii, xii, 1, 4, 23, 27, 225

Australia, 12, 193, 196, 212

B

Banking sector, vi, 3, 5, 39, 98, 102, 106, 119, 120, 128, 189, 191–198, 228–231

Bank law, 100, 103–105

Basel standards, 98, 106, 115, 117, 133

Bean market, 147–149, 158

Bellsouth, 54, 71, 73, 76, 83

C

Cable / Wireless, 54, 78, 83, 92

Canada, 12, 52, 53, 86, 162, 193

Carbonated beverages, 164, 168–173

Cartel, 156–162, 225–227

Cell phones, 228

CELTEL, 82, 91, 92

Central America, v, vii, 1, 2, 4, 5, 7, 8, 12, 15–18, 21–35, 39, 41–42, 49–61, 75–92, 98–104, 107, 112, 114–116, 118–120, 123, 125, 128, 129, 133–135, 139, 140, 150, 223–231

CFC. See Federal Commission on Competition

Chile, viii, xii, 38, 70, 83, 106, 194, 209

CLICAC. See Commission on Free Competition and Consumer Affairs

COFETEL. See Federal Telecommunications Commission

COMCEL. See Communications Celulares

Commission for the Defense and Promotion of Competition, 24, 29

Commission on Free Competition and Consumer Affairs (CLICAC), 18, 19, 24, 29, 32–34, 36, 40

Commission to promote competition (CPC), 4, 145–147, 149–160, 162, 164–175, 177–184, 206, 215

Competition

advocacy, 13, 28, 34, 225, 226, 230

agencies, 1, 2, 5, 7, 12, 14, 19, 24, 28–34, 39–43, 50, 62, 64, 73, 99, 102–104, 133, 146, 152, 154–156, 159–162, 174, 175, 178, 183, 184, 201, 202, 212, 217, 223–231

culture, 16, 22, 28, 226

law, iii, v, xii, 2, 7–11, 16–29, 33–39, 62–65, 102, 103, 154, 158, 177, 184, 223–230

policy, v, vi, vii, 1–5, 7–15, 18, 20, 22, 33, 35, 97, 134, 192, 197, 199, 201, 202

Competitive prices, v, 1, 41, 155

CONATEL. See National telecommunications commission

Concentration, v, 2, 3, 11, 15, 16, 20, 23, 25, 26, 33, 74, 97, 98, 100, 102, 104, 106, 111, 117–123, 140, 145, 146, 177–184, 189–192, 196, 197, 199, 202–206, 215, 217

Conglomerate, 15, 41, 196, 201

Consumer protection, 19, 20, 24, 25, 29, 32, 36, 145, 224

Contestability, 4, 98, 104, 113–115, 189–192, 197, 206–219

Contracts, 1, 25, 60, 73–75, 87–89, 91, 110, 132, 146, 162, 171–173

Costa Rica, vi, viii, xi, xii, 2, 4, 5, 7, 9, 14, 17–19, 23, 25, 27, 28, 32–34, 36, 40, 42, 43, 49–52, 61, 62, 79–81, 83, 89, 90, 92, 98, 100, 105–111, 113–117, 122, 124, 127–133, 135, 139, 140, 145, 153, 158, 161, 162, 164, 174–180, 184, 224, 226, 230

Costa Rican Electricty Institute (ICE), 49–52, 79, 90, 95

CPC. See Commission to promote competition

Credit card, 190, 191, 210–212, 215, 217, 218

D

Developing countries, v, vi, 1, 4, 5, 8, 12–15, 33, 35, 70, 156, 162, 209, 223, 224, 226, 228

Digicel, 74, 82, 83

Dominance, 26, 77, 175, 203, 204, 217

Dominican Republic – Central America Free Trade Agreement (DR-CAFTA), 21, 50, 51, 123

E

Economic commission for Latin America and the Caribbean (ECLAC), vii, viii, xi, xii, 39

Economic reforms, 9, 12, 13, 17, 43, 229

Electrical Components Market, 164–168

El Salvador, vii, 2, 7, 9, 17–19, 21, 23–30, 32, 34–36, 40–44, 49, 52, 56–59, 62, 63, 67, 73–76, 78–84, 86, 91, 93, 99, 100, 102, 103, 106–108, 110–112, 115, 116, 119–121, 124–127, 129, 133–136, 138–141, 194, 224, 230

Empresa Hondurena de Telecomunicaciones (HONDUTEL), 59, 60, 77–79, 82, 89

Empresa Nicaraguense de Telecomunicaciones (ENITEL), 56, 73, 77, 78, 83

ENITEL. See Empresa Nicaraguense de Telecomunicaciones

Entry barriers, 10, 11, 13, 40, 94, 104, 105, 111, 170, 172, 175, 180, 196, 220

European Union, 8, 19, 42, 183, 186, 201, 202, 212, 226, 227

Exclusivity period, 52–55, 70, 73, 76, 77, 93

Exports, viii, 13, 16, 18, 22, 25–27, 53, 56, 100, 153, 179

F

Federal commission on competition (CFC), 29, 65, 66, 70, 197, 199, 202, 203, 215, 217

Federal Telecommunications Commission (COFETEL), 65, 66, 70, 86

Financial crisis, 101, 121, 134, 198

Financial intermediation, 99, 114, 117, 142, 189–192, 204, 205, 218, 219

Financial margins, 190, 196, 202, 207–209, 213, 215, 220

Financial market, 5, 97, 111, 128, 182, 220

Financial sector, 18, 97, 98, 100, 102, 104, 111, 133, 142, 189, 190, 192, 196, 198–200, 202, 219, 231

Fixed telephony, 3, 49, 52, 54, 56, 57, 67, 68, 72, 74, 76–81, 83, 85–95

Flour market, 40, 41

Foreign direct investment, v

Foreign exchange, 107, 114, 199, 200

Free trade agreements, 18, 21, 45, 50–52, 95, 123

G

General Law of Telecommunications, 52, 65, 66, 70, 74

Globalization, 113, 117, 123, 228

Guatemala, vii, xii, 2, 7, 8, 17, 18, 21–30, 32, 40, 45, 49, 52, 56–59, 62, 63, 67, 73–76, 78–82, 84, 86, 88, 91, 101–103, 105–107, 109, 112, 115, 116, 119–122, 124–129, 131, 133–136, 138–142, 169, 194, 224, 230

H

Herfindahl-Hirschman Index, 118, 120–122, 141, 180, 203, 204, 217

Honduras, vii, viii, 2, 7, 9, 17, 18, 20–30, 32, 35, 36, 38, 40, 41, 43, 44, 49, 52, 56, 59–62, 64, 75–84, 86, 88, 89, 91, 92, 102, 104, 106, 107, 112, 115–117, 119, 120, 122, 124–127, 129, 133–136, 138–142, 224

HONDUTEL. See Empresa Hondurena de Telecomunicaciones

Horizontal monopolistic practices, 1, 44, 146

I

ICE. See Costa Rican Electricty Institute

Imports, 1, 12, 25, 26, 40, 41, 53, 110, 155, 179

Incumbent, 13, 50, 59, 61, 64, 69–72, 74–76, 78, 80, 81, 83, 93, 94, 103, 192, 229, 230

Industrial organization, 50, 75–83, 230, 231

Insurance companies, 102–104, 200

INTEL. See National Telecommunications Institute's

Intellectual property rights, 25

Intendency for the Protection of Competition, 24

Interbank markets, 100, 107, 109, 110, 114, 115, 132

Interconnection contracts, 74, 75

Interest rates, 1, 99, 100, 105, 133, 189, 190, 205, 206, 209, 210, 215, 218, 231

International cooperation, 4, 45, 101, 162, 174, 175

Internet, 51, 94, 95, 228

Investment funds, 141

J

Judicial system, 4, 9, 14, 16, 35–39, 44

Judiciary, 50, 94, 103, 158, 161, 176, 177, 226

M

Markets

competition, v, 9–11, 25, 43, 91

distortions, v, 1, 39–43

power, v, 3, 10, 15, 22, 27, 45, 63, 88, 104, 165, 168, 170–173, 177, 179, 196, 198, 202, 209, 215, 220, 225

MEGATEL, 56, 77, 83, 91, 92

Merger, v, 2, 12, 15, 23, 26, 34, 44, 94, 98, 102–104, 106, 115, 117–124, 133, 178–181, 189, 197–203, 226

Mexico, v, vi, vii, xi, xii, 2, 4, 5, 7, 9, 14, 17, 18, 22–45, 49–62, 65, 67–70, 75–86, 89, 93, 95, 97–104, 106–109, 112–116, 118–120, 123, 129, 132–135, 138–142, 189–220, 223–225, 227–231

Millicom, 82, 83

Minimum efficiency scale, 11, 12

Mobile telephony, 3, 49, 54–56, 67–70, 72–74, 79–83, 85, 90–95, 230

Monetary board, 102, 103

Monetary policy, 98, 100, 107–110, 142, 205

Monopolistic practices, 1, 23, 27, 34, 42, 44, 45, 145–165, 168, 169, 173–178, 184

Monopoly, 22, 23, 33, 43, 49, 52–55, 58, 62, 67, 73, 78, 81, 95, 100, 105, 114, 176, 230

Multinational corporations, 12, 42, 45

N

NAFTA. See North American Free Trade Agreement

National Consumer Protection Authority, 19, 24, 32, 36

National telecommunications commission (CONATEL), 59, 64, 77, 83, 88

National Telecommunications Institute (INTEL), 52–55, 57, 76, 89

New banks, 104, 112, 113, 219

New Zealand, 12, 14, 193

Nicaragua, 2, 7, 17, 21–23, 25, 27, 28, 32, 40, 41, 49, 52, 53, 55–57, 62, 64, 66, 67, 72, 73, 76, 77, 80, 83, 100, 103, 104, 106, 107, 116, 119, 120, 122, 124, 127, 131, 133, 135, 138–140, 224, 229

Nicaraguan Consumer and Competition Defense Commission, 24

North American Free Trade Agreement (NAFTA), 50, 53, 76

O

OECD. See Organisation for Economic Co-operation and Development's

Offshore banks, 98, 123, 128, 129

Organisation for Economic Co-operation and Development's (OECD), 53, 86, 106, 156, 157, 161, 175, 189, 207, 208, 213, 227

P

Panama, vii, 2, 4, 7, 9, 17–19, 23, 25, 27, 28, 32–36, 40–42, 44, 49, 52, 53, 56, 62, 65–67, 76–78, 80, 81, 83, 89, 92, 93, 95, 100, 101, 105–108, 112, 114–116, 121–124, 126–130, 133, 135, 139, 153, 162, 224, 230

Pension fund, 102, 103, 130, 132, 210

Practices restricted by their nature, 23

Predatory prices, 166, 167

Price fixing, 147, 206, 207

Prior notification, 33, 34, 44, 178, 180, 182, 183, 186

Privatization and liberalization, 3, 49, 50, 52, 55, 81, 93

Prohibited practices, 37, 145

Protectionist practices, 22

Prudential regulation, 98, 99, 101, 106

R

Regional group, 111, 124, 126, 127

Regulatory authority, 51, 66, 90

Relative monopolistic practices, 23, 27, 162, 164, 165, 168, 169, 174–177, 184, 201

Relevant market, 27, 65, 66, 150, 163, 165, 166, 168–172, 177, 179, 181, 190, 192, 198–201, 204, 205, 209, 211, 212, 215, 217, 218, 229

S

Sanctions, 8, 27, 35, 38, 63, 64, 146–148, 151–152, 155, 156, 158, 159, 161, 164, 166–168, 172–175, 177, 182, 185, 186, 227, 230

Shipping Companies, 153–154

Small and medium-sized firms, 2, 16

Small developing economies, viii, 9, 14–16, 232

Stockbrokerage firms, 102, 113, 115, 132, 141, 199, 201

Stock market, 55, 98, 111, 113–115, 132, 133

Subsidies, 74, 83, 85, 90, 94, 102, 192, 209–211, 218

Superintendence, 63, 103, 178, 184

Supply chains, 228

T

Tariff policies, 13

Taxation, 98, 128

Telecommunications

sector, v, vi, 4, 59, 60, 62, 65, 80, 229, 230

services, 51–53, 55, 58–61, 63–66, 77, 83, 86, 88, 90, 92, 93, 95

Telefonica, xii, 3, 46, 53, 57, 68, 69, 74, 76, 78, 82, 83, 91, 94–96

Telmex, xii, 3, 46, 53, 59, 65, 66, 68–70, 76–79, 81, 82, 85, 94–96

Tequila crisis, 99, 134

Transportation costs, 27

U

UNCTAD. See United Nations Conference on Trade and Development

Unefon-Iusacell, 82

Unfair competition, 23, 25

United Nations Conference on Trade and Development (UNCTAD), 5, 45–47, 187, 233

USA, 28, 41, 42, 44, 45, 52, 53, 69, 70, 88, 89, 91, 123, 124, 129, 161, 192, 193, 202, 206–214, 217–219, 227, 231, 232

W

Wireless telephony, 94