Intra-industry versus inter-industry trade: Relevance for adjustment costs

Abstract

their path-breaking work, Grubel and Lloyd (1975) observed that intra-industry trade (IIT) is a high proportion of total trade among industrialised countries. This was not expected on the basis of existing theories of international trade. The Grubel and Lloyd work then led to the development of a number of theoretical explanations of HT (see Helpman and Krugman 1985 for a review). Apart from the challenge that Grubel and Lloyd provided to trade theorists, a second important implication of their work concerns adjustment costs. On this issue, Hamilton and Kniest (1991), Greenaway et al. (1994), and Brulhart (1994) suggest that adjustment costs associated with trade liberalisation depend on the extent of HT. If trade expansion is predominantly of the intra-industry variety, then adjustment costs are likely to be lower than if most of the expansion is inter-industry. Both streams of the literature (trade theory and adjustment costs) have used various measures of intra-industry or matched trade. If mentioned at all, measures of inter-industry or unmatched trade have been treated merely as a residual: the difference between a measure of total trade and the favoured measure of IIT. The focus on measures of IIT is appropriate for the first stream of the literature. However, when we are dealing with adjustment costs, the focus should be on the measure of inter-industry or unmatched trade. In the next section, we examine Briilhart's measures of matched changes in trade. These overcome various problems associated with earlier work by Hamilton and Kniest and Greenaway et al. We then suggest a measure of unmatched changes in trade (UMCIT) and argue that it is this measure, rather than measures of matched changes in trade, that is relevant for studies concerned with adjustment costs. We