Competition Policy Analysis

Einar Hope

• Publisher: Routledge
• Year: 2000
• Print ISBN: 9780415226530
• eBook ISBN: 9781134736430
• Topic: Business
• Subtopic: Business General
• Index: Business

As markets become increasingly integrated and globalised, competition policy is facing new challenges. Contributions from leading international experts explore theoretical and methodological issues of practical relevance for the new competition policy order and give examples of practical policy adjustments.

1 Towards a competitive society?: The promotion of competition as a goal of economic policy

Agnar Sandmo

Introduction

The aim of competition policy is usually taken to be the achievement of efficient resource allocation through the promotion of effective competition. In fact, this is more or less a literal translation of the official objectives of the Norwegian Competition Authority according to the Competition Act of 1994. It is of considerable interest to note that it was not always like that. The Price Act of 1953, which provided the legal foundation for the Authority’s institutional predecessor, the Price Directorate, had among its aims the promotion of full employment, an efficient utilization of production possibilities and the achievement of an equitable distribution of income, and among its instruments the regulation of prices and dividends was accorded a degree of prominence at least equal to that of the promotion of competition. While the 1953 Act defined the objectives of the Price Directorate as being more or less equal to the goals of government policy in general, the new Act narrowed it down to the promotion of efficiency, while at the same time stating that the proper way to achieve this objective is to create conditions for effective competition.

It is probably fair to say that the new Act defines aims and objectives in a way which is much more congenial to economists who approach this area with a background in economic theory, but it also raises problems. Thus, by instructing the Competition Authority to abstract from distributional considerations, the new Act is based on some implicit assumptions about the separability of the goals of efficiency and equity which are not unproblematic. Moreover, by effectively excluding some parts of the market economy (e.g. agriculture and the labour market) from the scope of competition policy, the Act raises some fundamental questions about the justification for efficient markets in the rest of the economy. Basically, the question is whether the Competition Act represents a sensible policy of decentralization. In this chapter I propose to approach this question from the point of view of welfare economics, drawing in particular on the theory of the second best. Competition policy must fundamentally be judged in terms of its contribution to welfare, and the relationship between competition and welfare is one of the central concerns of welfare economics.

Before moving on to the substantial questions I should like to consider briefly the two central concepts used in the Norwegian Competition Act, namely, efficiency and competition. In a specific theoretical context the meaning of these terms is generally defined in a way which leaves little doubt about their precise content. However, theoretical concepts usually need a practical interpretation before they can be translated into the language of instruments and targets of economic policy.

Some history of thought

Economists are fond of quoting Adam Smith’s pronouncement on the invisible hand, by which individual economic agents in pursuing their own interests are led to promote the interests of society as a whole. While it may be tempting to see in this statement an early version of the first fundamental theorem of welfare economics, this would be an idealization of the history of ideas.

First of all, the concept of the public interest or the interests of society as a whole is a complex one, and Smith never gave a precise definition of it. It is possible to read into his statements a concept of the public interest as efficiency in production, as we would now say. Clearly, his conclusion that competition leads capital to flow to sectors of production where its rate of return is highest, is consistent with a view of competition as ensuring productive efficiency. But to go beyond this to a more general definition of efficiency would have required theoretical concepts that Smith and the other early economists simply did not have. To define efficiency in an economy with many individuals with partly conflicting interests was not possible before the introduction of the notion of Pareto optimality, which again was built on the concepts of preferences and utility. Needless to say, it took another hundred years for these concepts to be developed in the theoretical literature.

Second, Smith never gave a reasonably precise definition of what he meant by competition. Stigler (1965, p. 234) comments that the concept of competition did not in fact receive systematic attention in the theoretical literature until the 1870s, and that earlier it was ‘treated with the kindly casualness with which one treats of the intuitively obvious’. But it is clear that Smith identified competition with rivalry between independent agents, and that this rivalry would, at least in the long run, guide the flows of resources towards their most profitable uses.

Rivalry assumes at least two independent agents, i.e. the absence of monopoly, but the conditions for rivalry would be more likely to obtain, the larger the number of independent agents:

If this capital [sufficient to trade in a town] is divided between two different grocers, their competition will tend to make both of them sell cheaper, than if it were in the hands of one only; and if it were divided among twenty, their competition would be just so much the greater, and the chance of their combining together, in order to raise the price, just so much less.

(Smith, 1776 (1976), p. 361)

The formulation ‘just so much the greater’ probably should not be interpreted in a strict mathematical sense. In fact, the first mathematical formulation of a model of competition in the celebrated work of Cournot (1838) showed that such an interpretation would be incorrect. From his work one can easily derive an analytical expression for the degree of competition. For a monopolist the percentage mark-up of price over marginal cost will be equal to the inverse of the price elasticity of demand. In a Cournot duopoly equilibrium with identical cost functions it will be one half of the monopoly markup. With n oligopolists the mark-up will be 1/n times the inverse elasticity. As n increases, the equilibrium price approaches marginal cost, in which case ‘the effects of competition have reached their limit’.¹ So the mark-up, the deviation of price from marginal cost, diminishes with the number of competitors, but at a diminishing rate.

Cournot’s model was one of partial, not general equilibrium. The main step towards a model of general competitive equilibrium came in the 1870s with the work of Jevons, Menger and—above all—Walras (1874–77). When Pareto (1896–97) later laid the foundations for welfare economics, there is a direct line forward to the emphasis in contemporary theory on the two main theorems of welfare economics (first formulated by Arrow (1951) and Debreu (1951)) on the equivalence of competitive equilibrium and Pareto optimality. With Pareto optimality being the generally accepted standard of efficiency, it also became clear that the institutional framework that could be used to implement it was perfect competition, i.e. a market system in which no single agent was able to exercise market power.

Thanks to the precision of modern welfare economics it now also became clear to what extent one could set up an objective standard for the ‘public interest’ to be pursued by economic policy. Under ‘first best’ conditions, in which redistribution can be carried out by means of non-distortionary lump sum transfers, efficiency would always² be desirable as being in the public interest. However, when such transfers cannot be made, matters are more complicated, and a move towards greater efficiency would no longer necessarily be in the public interest in the sense of leading to unambiguously greater social welfare. The further implications of this conclusion will be discussed in more detail in the following.

This brief sketch of the history of economic thought on the connection between competition and efficiency over almost two hundred years shows a remarkable increase in the clarity and precision of the relevant theory. At the same time the theory also presents a competition authority with something of a puzzle. With perfect competition being necessary for efficiency, is any deviation from the competitive ideal a reason for interference? In order to answer this question one cannot rely solely on the formal theorems of welfare economics. All acts of policy interference are costly, and interference should therefore be based on a cost-benefit analysis. With limited resources on the part of the competition authority, priority should be given to interference in markets where the marginal efficiency gain, relative to the marginal cost of interference, is greatest. Such a conclusion fits well with the looser concept of competition in the work of Adam Smith and other classical writers: competition may be reasonably efficient even when it is not perfect. Most likely the prevention of monopoly is a more central task for the competition authority than the attempt to push the effect of perfect competition to its limit.

Another important distinction for competition policy is that between permanent and transient deviations of markets from the competitive ideal. Short-run variations in monopolistic mark-ups over the business cycle³ should probably not be a major concern for competition policy. Nor should the high price charged by a successful innovator during the early life of a new product be an argument for policy interference; in fact, the expectation of profits is an important incentive for dynamic efficiency. It is the more permanent deviations from the standard of competitive efficieny that should be the focus of competition policy, and it is accordingly those that I have in mind in the following analysis.

The scope of competition policy

The deviation of monopolistic from competitive prices has in principle two effects. (A) By raising prices above average costs, thereby generating pure profits, monopolistic pricing leads to a redistribution of income in society, away from that which would materialize under perfect competition. (B) By raising prices above marginal costs a monopoly distorts the efficiency properties of the competitive equilibrium, the normal implication being that the use of the monopolized commodity in consumption or production becomes too low. This is illustrated in Figure 1.1, which assumes that the demand function is linear and that unit costs are constant. The monopolist’s optimal output is x^m, which corresponds to the intersection of the marginal cost and marginal revenue curves, and this is to be contrasted with the competitive output of x*. With µ being the monopolistic mark-up (the difference between price and unit cost), the monopolist’s profit is π=µx^m, while the efficiency loss from monopolistic pricing is the triangular area e=(1/2)µ(x*-x^m). In general, both monopoly profit and the effiency loss depend on the elasticity of demand as well as on the elasticity of supply (which in the case shown is infinite). It is perhaps worth noting that in the special case represented in Figure 1.1 it will always be the case that the efficiency loss is exactly equal to half the monopoly profit;⁴ e=(1/2)π.

Figure 1.1 Monopoly profit and the social efficiency loss

While private agents who try to create a monopoly (sometimes with the support of politicians) are motivated by (A), competition policy is mainly motivated by the efficiency losses implicit in (B). An interesting question is now whether competition policy should be designed with both objectives in mind, just like tax policy has to be designed with a view both to efficiency and justice. I take the standard answer to this question, just as in the formulation of the Norwegian Competition Act, to be no, implying that competition policy should be formulated solely with regard to efficiency. Whether the answer is a good one will be discussed further below.

To which sectors of the economy should competition policy be extended? The view from welfare economics is clear: to all of them! The aim to strive for is clearly to have price equal to marginal cost in all markets. In the markets for consumer goods and services the implication of this would be the equality between consumer prices and the marginal costs of production, while in factor markets the rule would have to be formulated as equality between the prices of factors of production, as facing consumers, and their marginal value products. Thus, the general principle would apply to markets for primary commodities like agriculture as well as to labour markets. As a matter of fact, these are both examples of sectors of the economy which are fairly well protected from competition policy, and I shall maintain later on that this is basically because of distributional considerations. It follows that competition policy must limit its scope to certain —although large—sectors of the economy. In the next section I discuss some problems that follow from this limitation, even when efficiency is taken as the sole aim of competition policy.

Another area which traditionally has been well protected from interference by competition policy is the public sector. Some decades ago the prevailing view was that competition was not an issue in the evaluation of the public sector’s activities; in fact, in many cases it was maintained that private competition was harmful and should be forbidden. This has changed. It has increasingly come to be realized that actual governments are far from the picture of the night-watchman state which supplies a small set of public goods, including basic administrative infrastructure. To a large extent the public sector has become a supplier of prvate goods, like education, health, energy and communications, and in these areas private firms present government organizations with both real and potential competition. The prevailing opinion is now that this competitive pressure should be utilized to make the public sector more efficient. Increasingly, therefore, competition policy has had to concern itself with the interface between the private and public sectors of the economy.

Competition and efficiency

In its everyday work on implementation of competition policy, a competition authority must necessarily work on a market-by-market basis. Faced with monopolistic price-setting in a given sector, the aim of the authority should be to take measures which will lead to a reduction in price towards the competitive level.⁵ This will promote effective competition. But will it lead to a more efficient allocation of resources? The answer to this question depends crucially on what is assumed about the nature of market equilibrium in the rest of the economy. To begin with, I shall focus especially on the conditions in the parts of the economy which lie outside the domain of the competition authority. If some sectors of the economy are taken as being protected from interference from competition policy, how then should competition policy be designed inside its own domain?

This is a classic problem in the welfare economics of the second best. Indeed, it is one of the central applications of the theory in the original formulation of that problem by Lipsey and Lancaster (1956–57). ⁶ What they showed was basically that partial or piecemeal reform which appears to move the economy in the direction of efficiency, may not in fact do so. In particular, suppose that there is one market where there is an exogeneously given deviation of consumer price from marginal cost. This could be either because there is an institutional monopoly which cannot be removed or—perhaps more convincingly— a tax wedge which is motivated by overriding distributional concerns. Then it cannot in general be taken as desirable to have prices equal to marginal costs in other markets in the economy.

Why is this? There are two related explanations, one mathematical and one economic. To take the mathematical interpretation first, the classic conclusion about the efficiency properties of marginal cost prices can be seen as derived from the solution to a welfare maximization problem. In that problem the only constraint on the maximization of consumer welfare is production feasibility. If in addition it is assumed that there is one price which must be taken as different from the relevant marginal cost, this introduces an additional constraint into the problem. This constraint obviously prevents one of the first order conditions to be attained. But because of the interdependence of the variables in the overall maximization problem this also means that all the first order conditions will be affected by the additional constraint. Hence the conclusion.

This line of interpretation does not give us a good feel for the economics of the second-best problem. To achieve this, let us consider the problem within a specific model in which the structure is so simple that it can be analysed by purely verbal arguments. To be concrete, let us think of a model in which there are three goods—leisure, energy and a generalized consumption good which serves as the numéraire. For redistributive reasons there is a distortionary income tax, so that the consumer price of leisure, i.e. the after-tax wage rate, is below labour’s marginal productivity. The assumption is that this price distortion is one that competition policy cannot touch. I shall also assume that the substitution effects on leisure demand are larger than the income effects, so that the consumption of leisure is too high relative to the first best, i.e. labour supply is too low.

Suppose now that it is found that the consumer price of energy is higher than its marginal cost and that energy use for this reason is too low.⁷ This is within the domain of c...

Table of contents

• Cover Page
• Routledge Studies in the Modern World Economy
• Title Page
• Copyright Page
• Figures
• Tables
• List of Contributors
• Preface
• Introduction
• 1 Towards a competitive society?: The promotion of competition as a goal of economic policy
• 2 Competition law and policy*: Achievements and failures from an economic perspective
• 3 Competition policy and market dynamics
• 4 Who should be responsible for competition policy in regulated industries?
• 5 Implementation of second-best solutions : The judge or the bureaucrat? A lawyer’s perspective
• 6 Competition policy with a Coasian prior?
• 7 The Australian competition policy reforms
• 8 Competition policy in the information economy
• 9 Regulating manufacturers and their exclusive retailers
• 10 Deregulating Norwegian airlines
• 11 Resource allocation by a competition authority
• 12 European state aid policy: An economic analysis