Introduction

Universal service obligations (USOs) have been largely debated in the arena of public policy in most public utility sectors.1 In the United States, for instance, USOs were one of the major issues that were dealt with in the reform of the telecommunications sector. Indeed, the Telecommunications Act of 1996 devoted a large section to the reform of USOs and its regulatory agency, the Federal Communications Commission (FCC), has expended great efforts to implement the reforms. In other reforming countries, governments have also taken USOs seriously as a social commitment and implemented universal service policies in their regulatory reforms.

Even though most governments have committed to universal service policies to various degrees, the issue of USOs has also created considerable controversy in the reform process. Indeed, universal service policies raise many important practical as well as theoretical issues. Two important issues in the central debate of public policies are most relevant to developing countries. One is how the expansion of the network can best be promoted. Unlike industrialized economies, where penetration of basic infrastructure services is no longer an issue, developing countries have not yet created the infrastructures that are necessary for the provision of services to most people, particularly in rural areas. Expansion of the network therefore remains an essential dimension of universal service policies in LDCs. The other important issue in LDCs is the design of the pricing policy to implement USOs. In practice, since the necessary fiscal instruments are not available, policy makers often use pricing to achieve not only allocative efficiency but also a redistributive goal.

Introduction

Parts of the former natural monopolies for public services in telecommunications, electricity, gas, transportation, etc. are now viewed as potentially competitive, such as long-distance services in telecommunications, and generation in electricity or gas, and consequently open to competition. Conversely some elements, such as the transmission grid in electricity and the tracks in railways are still considered natural monopolies, and remain regulated, possibly with new forms of regulation like incentive regulation.

The management of the interface between the competitive and regulated sectors is crucial for the success of liberalization. The conditions under which competitors can access the regulated sector, which is an essential input for their activities, determine the profitability of entry and therefore the level of competition in the sectors opened to competition. They also dictate the efficiency of the utilization of the natural monopoly elements.

Despite its vital role for the success of liberalization, which is regarded as a key institutional change for development, no specific proposals of desirable access pricing rules for developing economies are currently available. The purpose of this chapter is to start to fill this gap. All along, we will keep in mind the main features of developing countries which may call for specific rules: high cost of public funds, poor auditing and monitoring facilities, low transaction costs of corruption, weak counterpowers, weak ability to commit, and inefficient tax systems.

Using incentive theory elaborated in the 1980s to model regulation as a problem of control under incomplete information, the new economics of regulation (Loeb and Magat, 1979; Baron and Myerson, 1982; Laffont and Tirole, 1986, 1993 (hereafter LT)) has provided a useful normative framework for the reforms of public services in developed countries. However, this literature has paid no attention to the specific characteristics of developing countries.

Simultaneously, the privatization, deregulation, and liberalization movement of the 1980s which started in the United Kingdom and the United States and then extended to Europe and some countries of Latin America (Chile and Argentina in particular) has provided a lot of useful experiences. Under the pressure of international banking institutions (IMF,World Bank), developing countries have been forced to liberalize their public services as the developed world had just done.

Advisers in LDCs could rely only on the experience of the developed countries and on an intellectual framework also designed for those countries. Not surprisingly they have essentially repeated the precepts designed for the developed world and paid little attention to the characteristics of LDCs.

Some economists at the World Bank were quite aware of the risk early on and have accumulated precious knowledge on some of the leading reforms in the world often hidden in internal reports but increasingly available in academic publications. In terms of countries, Argentina and Chile have generated more than their fair share of empirical assessments of regulatory problems resulting from reform.

Introduction

Following the pioneering application of incentive theory to regulation by Loeb and Magat (1979) and Baron and Myerson (1982), the 1980s witnessed the emergence of a new theory of regulation which emphasized the role of informational asymmetries between government, regulatory commissions, firms, and various interest groups. This work has been synthesized in particular in Laffont and Tirole (LT, Laffont and Tirole, 2000).

These theories have been developed at the same time as the large privatization–deregulation–liberalization movement was taking place in Europe, in the United States, and soon after in Latin America, and quite naturally focused on developed countries. Under the pressure of the International Monetary Fund (IMF) and the World Bank, this movement was imposed on LDCs in the 1990s with no attention paid to the specific circumstances of these countries.

In this chapter we recap in section 2.2 a canonical model of regulation built for developed countries. Then, in section 2.3 we examine the insights one can draw from this model in the light of the characteristics of LDCs. We use a unique data set of concession contracts in Latin America built by Guasch (2003) at the World Bank to test our theory. In section 2.4 we stress the weaknesses of this model for analyzing LDCs and make a list of the desirable extensions needed for a proper policy discussion of network industries in LDCs. Section 2.5 briefly concludes.

A simple model of regulation

Description

We consider a natural monopoly in charge of a public service such as water, electricity, telecommunications, or transportation.

Introduction

This quote from the Foreword of Kerf and Smith (1996) raises two types of issues in relation to privatization. (1) Is infrastructure privatization a valuable option for very poor countries? This question will lead us to review the available theory about the role of property rights in the performance of public utilities. However, the main emphasis of this chapter will be on a more positive question which must be elucidated by a theory of reform if privatization appears to be the direction to go. (2) When does privatization occur in countries where governments clearly have private agendas?

Our starting point will be what appears currently as an internal contradiction of many theories of privatization. Public ownership is criticized for allowing excessive political interference by governments who are not benevolent. Privatization is supposed to curb this influence; however, privatization itself is decided by politicians. It is therefore important to understand why the governing bodies find it advantageous to choose the path of privatization.

Introduction

This book focuses on public utilities, telecommunications, electricity, gas, water, transportation (roads, railways, buses, ports, airports, …) and the postal service which are sometimes referred to as “economic infrastructures.” It does not concern itself with the so-called “social infrastructures” such as education and health, or with financial infrastructures. This chapter will discuss the specific questions surrounding the regulation and liberalization of public utilities in developing countries. We first review the characteristics of developing countries that have a bearing on the analysis of regulation and competition policy.

An essential concept is the marginal cost of public funds – that is, the social cost of raising 1 unit of funds. This cost includes in particular a deadweight loss because governments raise revenues by means of distortionary taxes. It is estimated that this deadweight loss amounts to around 0.3 in developed countries, meaning that it costs citizens 1.3 units of account every time the government raises 1 unit. The inefficiency of tax systems in developing countries, coupled with the corruption that is sometimes also present, makes it extremely difficult for governments to invest in infrastructures and affects the cost of all types of public interventions, in particular, regulation and competition policy. According to World Bank data, the deadweight loss in developing countries is well beyond 1.0. It has been estimated at 1.2 in Malaysia and 2.5 in the Philippines, while in Thailand it ranges between 1.2 and 1.5 (Jones, Tandon, and Vogelsang, 1990).

Introduction

It is well recognized now that the design of proper institutions is crucial to development. Among the characteristics of governmental institutions, separation of powers stands as a cornerstone of democracy. Article 16 of the French Declaration of Human Rights of 1789 goes as far as saying “A society in which the guarantee of rights is not assured, nor the separation of powers provided for, has no constitution.” Indeed, since Montesquieu (1748), separation of powers has been explicitly recognized as vital:

Hamilton and Madison in The Federalist Papers (Madison, Hamilton, and Jay, 1788) referred to Montesquieu as “the oracle who is always consulted and cited on this subject.” They put these principles into practice for the American Constitution within a broader view of checks and balances. It is only recently that economists have started modeling the value of separation of powers.

A first reason for duplicating regulation agencies is yardstick competition. Using the correlation of the signals obtained by these agencies enables the principal to extract in a costless way their information rents. This idea was modeled by Shleifer (1985) in the case of perfect correlation and Crémer and McLean (1988) in the case of an arbitrary degree of nonzero correlation.

This book is only a sketch of what a book on regulation in LDCs should be. Ideally, such a book should provide a general model of regulation which fits the institutional characteristics of LDCs, give empirical evidence of the relevance of such a model, and derive policy implications.

Many obstacles need to be removed before such an ambitious goal can be achieved.

A more general model

Developing economies are often described as “economies with missing markets.” In the contractual world of regulation, this translates into incomplete contracts. Contracts are incomplete because of players' bounded rationality as in any economy, but also because of institutional weaknesses in enforcement, commitment, auditing, etc. Accordingly a theory of regulation viewed as a contract between the regulator and the regulated firm is bound to be more complex than a theory with complete contracts. Furthermore, as we have argued, institutional development proceeds by stages which call for different theories.

A given developing country is characterized by specific values of some crucial parameters such as the cost of public funds (which reflects the quality of the tax system), or the propensity to corruption (which reflects the lack of education, among other things), but also by the quality of institutions such as the quality of democracy, the quality of the judiciary, or the quality of auditing.

Policy recommendations for such a country should be based on a model which incorporates all these features. The work needed to obtain the mapping from the characteristics of the country to the policy recommendations is daunting, and probably beyond the capacity of the few researchers in this area.

Introduction

Regulatory contracts, like any other contractual relationships, suffer in developing countries from a severe lack of enforcement. Good laws and rules are rather straightforward to import from the developed world. A good set of lawyers can transfer this institutional knowledge quite easily (if not cheaply). It is much more difficult to enforce them, because of the lack of financial and technical resources, because of the corruption of enforcement institutions, and because of the weak bargaining power of regulators.

The vital role of enforcement for laws, rules, and contracts was first stressed by the Chicago School (Becker, 1968; Stigler, 1970; Becker and Stigler, 1974). They modeled economic agents as performing a cost–benefit analysis when breaching the law and they reflected on the role of punishments and their limitations due to corruption and limited liability.

In the law and economics literature (Posner, 1972; Polinsky, 1983), a lot of work has been done on breach of contracts and on the types of remedies which can be offered by the law However, the emphasis is not on how to react to renegations of contracts in fully anticipated states of nature, but rather on how laws can deal simply with circumstances arising from unexpected states of nature.

Introduction

The question of multiregulation in developing countries has many dimensions. First, one can think of geographical decentralization as one aspect of the problem. Should we have federal regulation in federal states or should we decentralize regulation in each state? For example, should we recommend federal regulation of telecommunications in Brazil or a two-tier system of state and federal regulation as in the United States or the European Union (EU)? Or, for water distribution, should we recommend regulation at the level of provinces rather than national regulations? These questions show that we must have a clear understanding of the pros and cons of decentralization to deal with our topic. Second, what is the desirable industrial scope of a regulator, or how many industries a regulator should supervise, is another question to answer when designing regulation. Should we have one regulator per industry, or one regulator for all industries, as in Panama and Jamaica, or should we recommend one for each industry, as in Mexico and Canada? Should the optimal design evolve over time, as the integration of gas and electricity regulations in the United Kingdom might suggest? Third, regulation has several functional dimensions, including regulation of prices, quality, environmental effects, entry and can be ex ante, as traditional regulation, or rather ex post, as competition policy. Should we have a single national body to deal with regulation and antitrust, as in Australia?

Introduction

In an elegant Econometrica paper, Hotelling (1938) provided the appropriate formulas to assess the social costs of marginal departures from marginal-cost pricing when the interrelations between commodities are taken into account. In so doing he generalized the work of Dupuit (1844) and Marshall (1890). But he went further.

Historical overview of electric industry in France: 1880–1946

Discovery and beginning use of electricity

France played an outstanding part in basic scientific and technical discoveries for industrial and domestic usage of electricity. However, industrial and commercial applications were long to come, compared to the United States, Germany, and Switzerland.

In August 1881, the First Exhibition of Electricity held in Paris revealed foreign competition's vitality in the scientific domain and the American and German significant lead in industrial and commercial applications. Convinced of the commercial interest of electric innovations, industrialists from those countries were rushing in a patent race.

French researchers were involved very early in hydroelectric industrial specialties: electrochemistry and electrometallurgy. The first French industrialists interested in electric power were paper manufacturers installed on the Alps rivers' banks. Using until then the mechanical power of running water (with watermills), and willing to improve their performance, they became the pioneers of hydroelectric industry in the Alps. However, their production was restricted to the needs of paper manufacturing.

Because of their huge electricity consumption, electrochemical and electrometallurgical industries started off the rise of the use of hydroelectricity in France. These two technologically advanced industries generated the first great hydroelectric installations in the Alps as early as 1890. They alone escaped foreign rule. Rapidly, electrochemical and electrometallurgical firms improved their knowledge in electricity production and transportation enough to be able to sell the energy surplus to nearby towns and villages (among them Lyon and Grenoble).