INTRODUCTION

Collusive agreements can take different forms: firms might agree on sales prices, allocate quotas among themselves, divide markets so that some firms decide not to be present in certain markets in exchange for being the sole seller in others, or coordinate their behaviour along some other dimensions. Institutional arrangements to sustain collusion might range from a very well organised cartel-like structure where a central office (secret, if anti-trust laws exist) takes the main decisions, to situations where firms merely find some form of communication to sustain the agreement. Further, a collusive outcome might be sustained even in a situation where firms never meet to discuss prices or never exchange sensitive information (but I shall argue that in such a case, labelled “tacit collusion”, the law should not intervene.)

Collusive practices allow firms to exert market power they would not otherwise have, and artificially restrict competition and increase prices, thereby reducing welfare. Accordingly, they are prohibited by any anti-trust law, and a large part of the anti-trust authorities' efforts is devoted to fighting such practices. However, while any serious anti-trust authority would certainly attack a cartel or an explicit agreement among competitors to set prices, or share markets, there might be divergences as to the standard of proof required in less blatant infringements of the law, and as to the treatment of cases where firms manage to keep prices high without overtly colluding.

The main purpose of this chapter is to identify the main mechanisms behind collusion, to study the factors which facilitate it, and to explain which behaviour should be treated as an infringement of the law and which one should not. I shall also analyse what actions anti-trust authorities should take in order to deter and break collusion.

The chapter is structured in the following way: Section 4.1.1 briefly sketches the main features of collusion from an economic point of view. Section 4.2 investigates the industry features, contractual characteristics, and other factors that make collusion more likely to occur.

WHAT ARE VERTICAL RESTRAINTS?

In most markets, producers do not sell their goods directly, but reach final customers through intermediaries, wholesalers and retailers. Further, the final good is often produced in several stages, from raw material, to intermediate good, to final product. Very often, firms at different stages of the vertical process do not simply rely on spot market transactions, but sign contracts of various types in order to reduce transaction costs, guarantee stability of supplies, and better co-ordinate actions. These agreements and contractual provisions between vertically related firms are called vertical restraints. This chapter analyses the welfare effects of vertical restraints as well as of vertical mergers, that is mergers between vertically related firms.

To gain some initial insight on the topic, consider the classical example of the vertical relationship between a manufacturer and a retailer which distributes its products. In general, both the manufacturer and the retailer decide on different actions, and what is an optimal action for one is not necessarily optimal for the other. As a result, a party can try to use contracts and clauses so as to restrain the choice of the other and induce an outcome which is more favourable to itself. (To put it another way, each party's actions create an externality on the other. Vertical contracts might be used to try to control for these externalities.)

For instance, the manufacturer would like the retailer to make a lot of effort in marketing its products (such as advertise its products, put them in evidence on shelves, employ specialised personnel who assist potential customers, offer post-sale assistance and so on), but the latter might have a lower incentive to do so, as effort and services are costly to provide. The manufacturer might then decide to use contractual provisions (that is, vertical restraints) in order to induce higher marketing effort from its retailer.

Recently, competition policy (or anti-trust policy, as it is more often called in the US) has often made the first pages of newspapers. High-profile cases both in the European Union and in the US have attracted the attention of society at large. Among the possible examples, there are US v. Microsoft (see Chapter 7 for a discussion), where the Department of Justice at one point asked for such a drastic measure as the split of the software giant into two separate companies; a few cartel cases with an international dimension (such as those involving the producers of lysine, vitamins, or the famous auction houses Sotheby's and Christie's), that resulted in prison sentences for some of the firms' managers involved; and some EU merger cases, such as General Electric/Honeywell (see Chapter 6), which was followed by public opinion on both sides of the Atlantic (and most people were surprised when eventually the European Commission blocked the deal between the two American companies).

What Competition Policy Is and Why We Need It Rather than starting by defining competition policy in abstract terms, in the book I first provide the reader with an idea of what competition policy is about through a historical approach (see Chapter 1). Only after having briefly described competition laws in the US and in the EU, do I give a formal definition of competition policy (see Chapter 2) as “the set of policies and laws which ensure that competition in the marketplace is not restricted in such a way as to reduce economic welfare”.

In this definition two elements should be underlined. The first is that firms might restrict competition in a way which is not necessarily detrimental (for instance, this is the case for most vertical restraints, that is, restrictive clauses between a manufacturer and a retailer, see Chapter 6). The second is that economic welfare, a standard concept for economists (see Chapter 2) is the objective that competition policies should pursue.

INTRODUCTION

As we have seen, the concept of market power is central to competition policy. So far, we have dealt with this concept from a theoretical point of view. This chapter introduces the reader to the issue of how market power should be assessed in practice. Many competition law investigations will start with such an assessment.

Ideally, one would like to estimate directly the extent to which a firm has (or increases its) market power. In merger cases, for instance, one might want to understand whether the merging firms will be able to profitably raise prices above the current level. Some modern econometric techniques (briefly analysed in the technical Section 3.3.2) allow us to do precisely that.

However, in many circumstances these econometric exercises are not feasible for lack of reliable data, and even when they are feasible, it might be a good idea to complement their results with the more traditional approach that evaluates the market power of firms by analysing the market in which they operate. In turn, this requires defining the “relevant market”, that is the set of products and geographical areas to which the products of the merging firms belong. It is such a set of products (and areas) that might create competitive constraints to the firms under analysis.

In this perspective, the definition of the market (both from its product and geographical points of view) is a preliminary step towards the assessment of market power.

In this chapter, I first discuss how to define a market (Section 3.2) and then how to assess market power (Section 3.3) so as to mimic the sequence that cases often follow in many anti-trust jurisdictions. Nevertheless, it should be stressed first that market definition is not of interest by itself, but only as a preliminary step towards the objective of assessing market power.

OVERVIEW OF THE CHAPTER

The basis of competition policy is the idea that monopolies are “bad”. Indeed, Section 2.2 shows that a monopoly causes a static inefficiency: for given technologies, monopoly pricing results in a welfare loss. Further, there generally is an inverse relationship between market power (the ability of firms to set prices above marginal costs), of which monopoly power is the most extreme form, and (static) welfare.

Sections 2.3 and 2.4 show that by looking only at allocative inefficiency one might actually underestimate the welfare loss from market power. A monopoly (more generally, high market power) might also result in productive and dynamic inefficiencies. Not only does a monopolist charge too high a price, but it might also have too high costs and innovate too little, since – sheltered from competition – it is not pushed to adopt the most efficient technologies and to invest much in R&D.

One might be tempted to conclude that if having one firm (or very few firms) leads to welfare losses, then competition policy should try to increase the number of firms, which operate in the industry (for instance subsidising and protecting less successful firms). Section 2.3 shows that such a conclusion would not be correct, because keeping less efficient firms artificially alive would distort the allocation of resources and reduce economies of scale, thus reducing welfare. In short: (1) competition policy is not concerned with maximising the number of firms, and (2) competition policy is concerned with defending market competition in order to increase welfare, not defending competitors.

Although there is an inverse relationship between market power and welfare under static analysis, it is not clear that the same unambiguous relationship exists when productive and dynamic inefficiencies are considered. In any event, Section 2.4 argues that market power is certainly not per se bad. Indeed, the prospect of enjoying some market power (and profits) is the main incentive for firms to invest and innovate.

INTRODUCTION

This chapter introduces the reader to imperfect competition models that are used in the technical sections of the book (specifically, the intermediate technical sections, labelled*; the advanced technical sections, labelled**, will likely need a stronger background than the one which is offered in this chapter).

Of course, this is no replacement for more proper training in basic industrial organisation models, but the chapter should help some students who have already some background in economics and in simple mathematical analysis (I use little more than derivatives of real functions in the book and in this chapter), or those who want to refresh a knowledge acquired some time ago, to follow the formal arguments made in the book.

The choice of the topics analysed here is functional, with the objective of helping the reader follow the book. The chapter starts with a short treatment of monopoly (Section 8.2), then introduces the reader to elementary game theory (Section 8.2.2.3), which is indispensable for understanding modern oligopoly theory, which for convenience I divide into static models (Section 8.3) and dynamic models (Section 8.4).

MONOPOLY

This section offers an introductory treatment of monopoly pricing. First the case of a single-product monopoly and then that of a multi-product monopoly are analysed.

8.2.1 Single-Product Monopoly

The easiest possible model of imperfect competition is one where there is a monopolist that sells only one good. I will first solve the monopolist's problem with general cost and demand functions, and then offer some specific examples.

Denote demand for this good as q = D(p), where p is the price and q output, and assume that demand is negatively sloped: ∂ D/∂p < 0.

INTRODUCTION

This chapter deals mainly with exclusionary practices, that is, practices carried out by an incumbent with the aim of deterring entry or forcing the exit of rivals. By and large, such practices correspond to the legal concepts of monopolisation in the US and abuse of dominance in the EU (see Chapter I).

The identification of exclusionary behaviour is one of the most difficult topics in competition policy, as often exclusionary practices cannot be easily distinguished from competitive actions that benefit consumers. For instance, suppose that following entry into an industry a dominant firm reduces its prices considerably: should this be considered an anti-competitive strategy aimed at forcing the new entrant out of the industry (after which prices will be raised again, damaging consumers in the long-run), or is it instead just a competitive response that will be beneficial to consumers? Most of this chapter will be devoted to understanding how to answer this question.

Exclusionary practices by incumbents are certainly not a new phenomenon, but there are at least two reasons why such practices should receive fresh attention. The first is that in many countries there have been processes of liberalisation, privatisation, and deregulation that have resulted in several sectors having an incumbent facing potential entrants. This asymmetric structure creates strong incentives for potential exclusionary behaviour. The second is that a growing share of today's advanced economies is composed of sectors (for example computer software, Internet and telecommunications) that exhibit network and lock-in effects. In such environments, entrants might find it very difficult to compete with incumbents, and particular attention should be paid to possible exclusionary practices.

Section 7.2 focuses on pricing strategies, and Section 7.3 on non-pricing strategies, such as over-investment, tying and bundling, and incompatibility choices.

INTRODUCTION

Rather than starting with a long and abstract discussion of what competition policy is, this chapter aims at introducing the reader to competition issues by using a historical approach. Section 1.2 briefly describes the main features that competition policies have exhibited in the past in the US and in Europe. The historical review also shows that in the practice of competition policy a number of public policy considerations and objectives have been (and still are) used. Section 1.3 briefly discusses them, and indicates the possible conflicts between economic and noneconomic objectives. Armed with this discussion, at the end of the Section I also provide the definition of competition policy that I use in the book. Section 1.4 describes the main features of competition law in the European Union (EU), to provide the reader with further insight of what competition policy is about.

BRIEF HISTORY OF COMPETITION POLICY

This section briefly reviews the main historical events in the development of competition (or anti-trust) laws in the US and in the European Union. The purpose here is not to have a complete description of the history of competition laws, but rather to help understand the circumstances in which competition laws were created and enforced, as well as the objectives which they purported to attain.

1.2.1 Anti-Trust Law in the United States

The origins of modern competition policy can be traced back to the end of the 19th century, mainly as a reaction to the formation of trusts in the United States.

The Events Leading to the Sherman Act In the second half of that century, the United States experienced a number of events, which resulted in the transformation of manufacturing industries. Perhaps the most important events were the dramatic improvement in transportation and communication. The railways extended rapidly throughout the US territory, as did the telegraph lines and the telephone services.

Introduction

The literature on the nature and the structure of optimal tariffs is an old one in the theory of international trade. There are many arguments for the imposition of tariffs, the most well-known argument being the monopoly power in trade – the terms of trade argument (see Bhagwati and Ramaswamy, 1963): since a large country can affect the international terms of trade by imposing tariffs, it can use tariffs to maximize its own welfare. There are other reasons, such as monopoly power in production (Katrak, 1977), unalterable domestic distortions, the infant industry argument (see Corden, 1974, ch. 9), etc., for a country to impose tariffs in order to raise its welfare.

The above literature ignores one important aspect of real life, viz., the fact that often producers and sellers of a commodity are different entities. For example, Toyota cars are most often sold abroad by dealers that are nationals of the country where the cars are sold. Another example is the clothing industry where items are usually sold by big stores under their own brand names (e.g., St. Michael for Marks and Spencer) but are often produced not by the stores but by other domestic and/or foreign producers.

In deciding on the level of the optimal tariff on a commodity under this circumstance, the importing country clearly has to take into account its effect on the domestic sellers' profits. Can this new consideration change the sign of optimal tariffs?