The proposition that imports may constrain domestic competition and ought to be encouraged on at least monopoly welfare loss grounds is neither new nor very frequently challenged. In this paper, we should like to present several reflections on the literature — empirical and theoretical — associated with this proposition, and to sketch the outlines of two further empirical exercises which we believe capable of yielding information on some aspects of this general view.

It is becoming increasingly accepted that industry conduct ought to play an important role in our theorizing about oligopoly, and that structural indices can be poor, and occasionally quite inaccurate reflections of the more complex reality of industry performance. In this section, we should like to explore this insight in the context of competition in an open economy.

The aim of this paper is to derive and test equilibrium relations between aggregated indices of market power, concentration and foreign trade variables. Encaoua and Jacquemin (1980) established such relationships between different measures of concentration and Lerner indices, but do not take into account international competition.

The first part of this paper provides a brief survey of previous theoretical work concerning the impact of foreign trade on domestic market performance, and derives a set of oligopolistic models for both homogeneous and differentiated goods in a open economy. The second part tests these models using a sample of 31 Belgian industrial sectors in 1973.

Because Belgium is a small open economy, Belgian producers are unable to affect prices on international markets for homogeneous goods (assuming there are no impediments to trade). Thus Belgian firms are expected to be price-takers enjoying no market power whatever the level of concentration. There are, however, situations in which firms in a small open economy could theoretically enjoy some market power. First, if there are impediments to trade, the domestic market can be separated from the foreign market so that domestic producers are able to charge higher prices at home than given world market price (price discrimination). Second, industries manufacturing differentiated goods for which no perfect substitutes are being produced in the rest of the world face a downward-sloping demand curve.

The aim of this paper is to clarify three issues relating to the empirical investigation of prices and import penetration when marginal costs are not increasing. The first is to demonstrate explicitly the simultaneous determination of industrial prices and import penetration. Second, to examine how oligopoly and/or product differentiation may lead to international trade between identical countries. Third, to emphasise the importance of the technology of demand when trying to explain price and trade. The technology of production is often mentioned in this context, yet the technology of demand, by which is meant the characterisation of the tastes of consumers, is usually ignored but is shown to be of some importance.

In order to simplify the discussion it is assumed that all firms in all countries have access to identical technologies and factor costs. Unlike in the usual textbook models, marginal costs are assumed to be nonincreasing. Furthermore, the consumption patterns in each country are taken to be identical. There is, however, one difference between selling on the home market, and selling abroad — there are trading costs. These may result from transport costs, tariffs, social, linguistic and legal factors, or anything else, but they are assumed to apply with equal magnitude to each country. These assumptions permit a clearer analysis of the effects of oligopoly and product differentiation.

The analysis of intra-industry trade, as it is understood now, was born almost by accident. In his study of the changes in the pattern of intrabloc trade of the Benelux Union, Verdoorn (1960) calculated the bilateral trade ratios of a sample of 121 products at comparable levels of international trade classification, for two different points in time. He noticed that the number of the extreme values of the ratios had been reduced while their median increased since the formation of the Benelux Union. From this he inferred that specialization, if it did accompany the intra-bloc trade, was to be found within rather than between the different categories of trade. He also argued that the rather large variance of the price differentials found for the Dutch and BLEU bilateral export prices indicated, in the absence of trade obstacles, a high degree of product differentiation and corroborated the trend towards intra-industry specialization. These findings, tucked away into seven brief sentences in the text of Verdoorn’s (1960) article did not immediately attract attention as “new territory”.

The theory of Industrial Organization has traditionally assumed that firms in an industry are alike in all important dimensions except for their market share. In this context, a large body of the literature posits that some long run structural features on an industry such as the size distribution of sellers, the nature of the product and of technology and the existence of barriers to entry generate market power. Above normal profits are the manifestation of this market power and firms are assumed to benefit from it in proportion to their market share.

This theory of industry-wide profit determination has recently been challenged by several authors (Hunt (1972), Newman (1978), Porter (1979), Caves and Pugel (1980), Encaoua and Franck (1980), Encaoua and Jacquemin (1981)). In particular, they have emphasized differences among firms’ strategies within an industry, indicating that these differences appear to be rather stable in the long run. Hence, strategic groups, i.e., groups of firms highly symmetrical in their strategies have been introduced as an additional element of market structure. The resulting hypothesis states that industries with substantial heterogeneous strategic groups should not fit the traditional structure-performance paradigm as well as industries with homogeneous strategic groups. Empirical support for this hypothesis can be found in Newman (1978) and Encaoua and Franck (1980).

The hypothesis that the U.K.’s domestic manufacturing industry is subject to significant foreign price competition has been frequently vaunted in both the academic and journalistic literature. Empirical examinations of this hypothesis have either related domestic to foreign prices — in times series analysis — or some measure of the domestic price-cost markup to a tariff or import share variable — in the cross-section approach. Since our concern is with allegedly increasing international competition over time, we examine a simple price formation specification in time series.

We proceed by discussing the theoretical and empirical literature. It is suggested that the existing empirical results may have several possible alternative explanations and that U.K. manufacturing industry may not be subject to significant foreign price competition. It is argued that the conventional hypothesis that imports are independent of the domestic oligopoly structure is not altogether convincing. We pursue a specification which is an adaptation of that adopted by Coutts et al (1978) and use this to question the conventional view of import price competition. The results obtained in the estimation of this model, while not conclusive, do give grounds for doubting the conventional hypothesis.

Assuming a certain extend of labour immobility between different industrial sectors, this paper shall test empirically the effect of monopoly power on labour earnings in 40 West German industries.

This investigation has been done before by several authors for different countries, but so far only one empirical analysis exists for West Germany, where a negative effect of firms’ concentration on wages was found. This is rather surprising, considering the opposite findings for the USA, the UK, France, Belgium and Italy, where in all cases positive relationships between firms’ concentration and wages have been revealed. Therefore another test for West Germany seems justified.

In general, the analysis of links between monopoly power and labour income contributes to the research in welfare loss through monopolization, since any empirical measure of this welfare loss is understated if some fraction of monopoly profit is transferred to labour.