Intra-Industry Trade

The phenomenon of intra-industry trade was empirically observed and presented with some emphasis in studies carried out in the 1960s which dealt mainly with the effects of economic integration on specialisation in trade. This generally recognised¹ origin of the study of intra-industry trade relates in particular to the effects of the Benelux founded in 1948² and the European Economic Community founded in 1958.³ These studies found that the increase in trade among the members of the integration scheme had taken place largely through specialisation in production and export of products of the same industries rather than of different industries. Such a finding of increased intra-industry specialisation was unexpected. The traditional trade theory predicted that it was specialisation of the inter-industry type that would follow trade liberalisation.⁴ This prediction was reflected also in empirical studies attempting to forecast the developments in trade that would result from the mutual tariff concessions intended by the countries of the proposed European Economic Community.⁵ These expectations were founded on the standard economic theory explaining specialisation in trade: the Heckscher-Ohlin theory. As is well known, this theory assumes industries characterised by production processes with constant returns to scale, producing perfectly homogeneous goods in perfectly competitive markets.

This chapter offers some reflections on the recent evolution of intra-industry trade theory. These are selective and mainly relate to the interpretations of intra-industry trade in factor proportions models including some recent continuum models.

Intra-industry trade, i.e. the simultaneous imports and exports of the same statistical product group, has become an increasingly important part of world trade, in particular in the exchange of goods among developed countries (for a survey of findings see Tharakan, 1983). This fact has initiated empirical research on the causes of intra-industry trade. Attempts to explain differences in the share of intra-industry trade of total trade among different industries or product groups in terms of characteristics of the product or the market have been made, e.g. for the UK by Greenaway and Milner (1984), for the US by Toh (1982) and Bergstrand (1983), and for a sample of developed economies by Finger and De Rosa (1979), Loertscher and Wolter (1980) and Caves (1981). The explanatory variables used in these studies are generally assumed to capture some aspect of the concept of product differentiation. They include measures based on the statistical classification itself (e.g. subdivisions of the SITC or the BTN), as well as R&D costs, advertising expenditures, product age and measures of concentration and economies of scale. The basic hypothesis is that the higher the degree of product differentiation in an industry, the more intra-industry trade there will be.

Introducing their chapter on ‘Differentiated Products and Economies of Scale’ Grubel and Lloyd (1975) distinguish various groups of products, differing from each other according to two criteria: similarity of input requirements and substitutability in end use. The products serving as examples comprise final consumer goods such as cars and cigarettes, but also intermediate products as yarn and steel bars. Later on in their chapter when discussing the ‘nature of product differentiation and trade’, examples offered refer to consumer goods only.

In this paper we focus attention on the bilateral intra-industry trade (IIT)¹ in manufactures between selected Developed Market Economies (DMEs)² and the Developing Countries (DCs).³ In spite of the rapidly expanding literature on various aspects of intra-industry trade, the study of such trade flows of the DCs and the theoretical and policy implications thereof remain a rather neglected area. Nevertheless recent research suggests that the IIT of some of the Newly Industrialised Countries (NICs) has increased more rapidly than that of the DMEs (Lee, 1987). Further, there is some evidence of the prevalence of intra-industry trade in manufactures between countries with clearly different factor endowment patterns (Tharakan, 1986), although an extensive analysis of such trade at highly disaggregative levels in terms of products and countries is yet to be undertaken.

The present Uruguay-round of GATT-negotiations includes trade in services on its agenda. Yet, in spite of its importance and the widespread interest that has been accorded to it by policy-makers, international trade in services has as yet been the subject of only limited theoretical and empirical work.

Trade in goods which are similarly classified in production but are not perfect substitutes in consumption is apparently a growing component of world trade. Reduced tariffs enacted by customs unions have resulted in increased gross trade, while net exports and imports within commodity classifications change very little. Trade theory which assumes constant returns to scale and homogeneous goods generally predicts increased specialisation and inter-industry trade.

The effective tariff concept is well established in the theoretical literature in the analysis of inter-industry trade flows. This analysis is based upon assumptions such as product homogeneity, non-increasing returns, arms-length trade, and small open economy conditions. Relaxation of some or all of these assumptions has direct implications for effective protection analysis under any type of trade flows.¹ As is now widely recognised however, relaxation of these assumptions is also likely to be associated with intra-industry specialisation and exchange. There is a large literature on the effects of nominal tariffs in this context.² There is, by contrast, a paucity of studies on effective protection in the context of intra-industry trade.³ This chapter, therefore, seeks to redress this balance, and to employ effective protection analysis under conditions of ‘within-industry’ specialisation (horizontal and vertical) and of imperfect competition.