Causes of Changes in the Structure of International Trade, 1960–85

Two key questions have dominated trade theory. These concern the gains from trade and the pattern of trade. Only the latter issue was dealt with in the IESG Conference.

What determines the commodity pattern of trade between countries? Why does one country export textiles and another export wine? This has long been the central question of trade theory.

This chapter surveys some of the broad patterns in world trade of manufactured goods over the last two or three decades. It discusses whether they arise from ‘natural’ economic phenomena — factor endowments and technological advance — or from conscious efforts on the parts of governments to control trade through policies such as tariffs and NTBs. While not denying the possibility that trade policy can be restrictive enough to matter, it concludes that, on the whole, factor endowments offer the principal explanation of what we observe. This is contrary to much popular discussion which implicitly attributes both apparently successful trade performance — such as Japan — and allegedly poor trade performances — such as the UK — to the respective determined single-mindedness and bumbling incompetence of the trade ministries involved.

Empirical evidence suggests that a significant proportion of trade (in particular, but not exclusively) in manufactures, in particular (but again not exclusively) between industrialised countries, is now intra-industry in nature. The focus of this conference is on causes of changes in the structure of trade during the period 1960-85. There is an obvious rationale therefore for considering the significance of intra-industry specialisation in the changing structure of world trade, how we might explain the occurrence and growth of this phenomenon, and its policy implications. This is the aim of our chapter.

In recent years there has been a coherent attempt to come to terms with the deficiencies in the traditional Heckscher—Ohlin approach to the determination of trade patterns. Kierzkowski (1984) provides a helpful bringing together of many of the individual contributions where the major feature is the introduction of a monopolistically competitive rather than a perfectly competitive world. Helpman and Krugman’s (1985) book helps to set out many of these ideas in a coherent framework. Nevertheless, existing theoretical explanations of trade patterns among developed countries still seem to fall rather a long way short of the actual patterns observed. It is not totally clear from basic factor endowments why trade has developed its current pattern even when appropriate adjustment is made for the major array of distortions that have occurred.

The structure of international trade changed considerably in the quarter-century following 1960. Consider first geographical patterns. The older industrial countries (ICs) continue to dominate and account for nearly two-thirds of world trade but their share has declined gradually with the advent of the NICs in manufactures, and oil-exporting countries in fuels. The relative importance of trade among developing countries has grown quite steadily since 1973, more or less in line with the slightly increased share of GDP in developing countries¹. As for the commodity structure, the most significant change has been the sharply increased share of manufactures in the total exports of developing countries, particularly the NICs. This is also true for the commodity structure of developing country exports to each other.

Although the proposition is sometimes called into question, there is substantial evidence that there is a close connection between economic development on the one hand, and export growth on the other. In looking to expanded developing country intra-trade¹ it seems difficult to share the optimism of some that trade between developing countries provides an alternative to growth in North/South trade². Furthermore, recent history has further validated the proposition that the accumulation of foreign debt is no long-term substitute for exports; and even if conceived as a short-term relief, accumulated foreign debt can, in fact, seriously impede the process of economic development due to changing external conditions not under the control of the borrowing country. Hence whatever solution to the present debt problem of developing countries may eventually be found, export growth will have to continue if development is to proceed. The envisaged reduction of foreign debt from present high levels calls for even larger export increases in the years to come.

*The author was until 1985 Professorial Fellow, Centre of Policy Studies, Monash University, Australia and Senior Fellow in Economic Policy with the Reserve Bank of Australia. While he is currently Senior Counsellor at GATT, the chapter was written while he was employed by UNCTAD. A special debt is owed to Sam Laird of the World Bank for his contributions to the modelling section. Rene Vossenaar of UNCTAD generated the data on NTBs. Helpful comments on an earlier draft by Robert Baldwin of the University of Wisconsin and Richard Eglin of the GATT secretariat are gratefully appreciated. The views expressed in this chapter are those of the author and not the institutions with which he is currently, or has been associated.

There is a reigning paradigm in international trade: the H—O—S theory of relative factor endowments. There are also a number of special theories: for example, the product-cycle of Vernon (1966) and the IIT model of Grubel (1970), but these do not aspire to explain international trade patterns in general, merely a part of them. This chapter will not concern itself with these.

The falling UK share of world trade in manufactures is one of the great empirical regularities of economics, worthy to stand beside other ‘stylised facts’ like the rising capital-output ratio. Those who have studied the matter empirically seem agreed that non-price factors- such as failure to innovate, poor marketing and after-sales service — should bear the brunt of the blame. Much anecdotal evidence supports this view. There is also quantitative evidence in the shape of the low unit values, relative to competitors, of many categories of UK exports which suggest relatively low quality or less sophisticated products (Posner and Steer, 1979; Stout, 1979).