The sample consists of all firms for which data were available over the entire period under review (1965–82). These included all firms listed on the stock exchange or subject to the legal obligation to publish financial statements. Although the number of firms subject to this legal obligation in 1965 was close to a thousand, a number of them were merged, liquidated, or transformed into holding companies during the period under review. Some were also not legally obligated to publish data in every year. Thus, our sample consists of 450 firms (see Table 7.1).

The profit rate variable is defined as follows: (profit before taxes + interest payments)/total assets. Although it would have been preferable to use after tax profits, the amount of taxes paid by the firms was not available. One should note that in previous structure performance studies in France, the same profit rate definition was used and the econometric results were consistent with those of other countries and what standard theories predict.

The average profit rates of the 450 firms in the sample and the dispersion of profit rates varied over time, as can be seen from Table 7.2. The 18-year interval can be subdivided into four periods: profit stability from 1965 through 1968; high average profit rates from 1969 through 1973, which was one of sustained growth in France; a period of profit rate instability (1974–80) reflecting adjustment to the oil crisis; and a decline in profits in 1981 and 1982, possibly due to the measures taken by the socialist government that came into power in 1981 but also to stiffening competition among firms in a period of declining growth.

Two views of competition

Two views about competition exist. The first sees competition as a process for allocating resources to their optimal uses. The price mechanism is the instrument for achieving this goal, and when it functions properly, equilibria emerge with prices equated to marginal social costs of production. When it malfunctions, equilibria exist with some prices above marginal costs, and society suffers a welfare loss from the underconsumption of these goods. Such malfunctions are usually attributed to an insufficient number of buyers or sellers. Monopoly is seen as the antithesis of competition. Thus, under the first view, competition is seen as a process for determining prices and quantities, the allocation of resources for a given set of tastes and technological opportunities. At its zenith, competition produces an equilibrium set of prices that induce a Pareto optimal allocation of the economy's goods and services. Such equilibria are anticipated so long as monopolistic elements are absent.

The other view of competition sees it not as a process for allocating a given stock of resources but as a process for transforming these resources into new products and production techniques. Competition takes the form not of lower prices for an existing set of products but of new and improved ideas, and these in turn are the property of the individuals) who created them and his/her/their employer. In the first instance, competition for a new product is competition for a newly created monopoly. With time the monopoly disappears as other firms imitate and improve upon the new product.

Introduction

The structure–conduct–performance model has long attracted the attention of industrial economists interested in the empirical analysis of monopoly. The most notable characteristic of that model is that much of the theorizing underlying it is static, and virtually all of the associated empirical work has been cross-sectional in character. Standing at a slight distance to the structure–conduct–performance paradigm are evolutionary and Schumpeterian arguments. Focusing on innovation, imitation, and adaptation, they are concerned not so much with monopoly as with its persistence. Both of these alternative lines of thought are fundamentally dynamic in character, and both are concerned with analyzing competition as a process. To date, neither have generated much empirical research, much less a full-blown empirical methodology. However, as industrial economists gradually come to perceive more and more limitations in static models and cross-sectional empirical analysis, so the attractiveness of developing alternative methodologies will grow. Under the circumstances, it is natural to think that extending the empirical analysis of monopoly to examine its persistence will be placed high on the research agenda.

In this chapter, we shall make a case for extending traditional static, cross-sectional empirical models to include market dynamics. In the main, we shall examine a particularly simple type of dynamic model whose major characteristic is an autoregressive structure that emerges as the solution to a latent variables problem. Aside from any other virtues, the model does enable one to see clearly that latent variables are a major difficulty that must be faced when modeling market dynamics.

Introduction

That the recent performance of the U.K. economy has been poor is beyond dispute. One diagnosis of this state of affairs asserts that the cause of the problem arises from the fact that U.K. markets are, in general, rather uncompetitive. There are two steps to the argument. The first asserts that strong and vigorous competitive processes stimulate the kinds of innovativeness and general dynamism that generate rapid economic growth and development. Entry and fringe firm activity, it is argued, encourage the generation and diffusion of new products and techniques and provide the kind of pressure needed to keep all industry members alert, efficient, and flexible. The second strand of the argument asserts that the competitive process in the United Kingdom is, in fact, quite weak. Relatively low rates of innovation and productivity increases, sluggish price responses to cost and demand changes, and a poor balance of trade in medium and high technology products are all taken to point to a lack of dynamism. Further, U.K. industries are highly concentrated, and high gross entry flows in the United Kingdom appear to produce little more than high gross exit flows. That the two sets of facts are associated is suggested by studies that indicate that highly concentrated, low entry industries may be less innovative and less price flexible than others (e.g., Encaoua and Geroski, 1986; Geroski, 1987). Although not conclusive, the overall argument has at least prima facie appeal.

Are industries in the United Kingdom competitive enough to ensure satisfactory rates of innovation and growth? To answer this, one must define the adjective “competitive” and then measure how extensive the force of competition is in various industries.

In this chapter we develop and test a simple autoregressive model that describes the profits of industries over time. The model explicitly addresses the issue of incomplete adjustments of profitability and distinguishes longrun from short-run effects. The model is fit to census data on U.S. manufacturing industries over the period 1967–82. Industry-specific estimates of the speed of adjustment of the profit rate toward its long-run equilibrium level and hence measures of the degree of persistence in performance across the various industries are obtained. These estimates are then linked to structural characteristics that theory suggests might be pertinent in determining the rate at which the forces of competition erode excess returns. The results of the chapter provide support for the view that competition in an industry is rooted in its underlying economic structure. We find support for both the Chamberlinian hypothesis and the more recently advanced contestability conjecture that states that the performance of industries depends continuously on the degree to which they exhibit imperfect contestability.

A partial adjustment model

Let Πit denote the profit rate of industry i at time t. We postulate that the level of Πit reflects industry-specific characteristics determining both the internal conditions of actual competition among the established firms in the industry and also the external conditions of potential competition from outside firms. Among the internal conditions we include brand loyalties, cost and informational asymmetries, demand inelasticities, and so on. Prominent among the industry characteristics that determine the competitive pressure exerted by potential entrants is the height of entry barriers.

Introduction

The state of competition and pace of adjustment are frequent themes in public policy discussions regarding Canadian industry. Compared with other Western industrialized countries, Canadian industry tends to be highly concentrated, and domestic markets are generally small relative to efficient scale of production and subject to tariff protection. Over the past decade or so, however, the trends toward increased industry concentration have been less marked, tariffs have on average been reduced to half their pre-1970 levels, and import competition pressures have been exerted in a wide spectrum of markets (see Khemani, 1986). Negotiations for bilateral free trade between Canada and the United States are also currently under way.

In light of these developments, it is of some interest to examine the trend of corporate profits over time. To the extent that the competitive process is relatively fast, profits above or below the competitive norm should disappear (allowing for factors such as uncertainty, innovations, and changes in tastes). If, however, profits persist over time, they may be indicative of impediments in the competitive process such as barriers to entry and oligopolistic coordination of firm price–output policies.

In this chapter, we first measure the extent of persistence in long-run profits of large Canadian corporations over the period 1964–82. We then examine the determinants of these profits and the role played by entry and exit of firms in the adjustment process. Our conclusions are that the competitive process does work in pushing profit rates toward the competitive norm but it does not succeed in equalizing profit rates.