Distance and the Pattern of Intra-European Trade

Abstract

THE basic question to which this paper is addressed is What is the importance of distance in determining the pattern of Western European trade? The importance of distance in the pattern of trade has, of course, always been recognized. The assumption of no transport which has always been necessary in expositions of theories of international trade, is a recognition of the fact that transport costs, that is, the costs of covering distance, exist and are significant; so that abstraction from them has to be made quite explicitly in order to analyze other elements such as factor endowments.' First, it is necessary to abstract from the effect on the pattern of trade related simply to differences in the size of countries. Thus, Germany's imports from Greece are likely to be smaller than Germany's imports from France simply because the total exports of Greece are small relative to the total exports of France. Therefore, in order to explain the distribution of Germany's imports in terms of factors other than the relative size of the supplying countries, say, in terms of distance, relative prices, relative incomes, the data have to be corrected to eliminate the effect of differences in the size of the trade of the various countries. This may be done by adjusting every country to an (equal or basis. The framework used for this adjustment is as follows. To facilitate statistical work, Western Europe is defined as the member countries of the O.E.E.C. The intra-European trade of these countries is arranged in the form of a trade matrix. The data used are the f.o.b. export figures as recorded by the exporting countries. Then in order to examine first the distribution of each country's imports, all countries have been put on an basis. To do this the elements in all the rows (exports) of all the countries have been multiplied by the coefficients (one for each row) required to bring the row totals up to (or down to) ioo. Similarly, in order to examine the distribution of each country's exports, the data have to be adjusted to put each country on an basis by adjusting the columns relating to each country in such a way that they all total ioo. The principal tables are in the form of trade matrices; there are two tables for each year examined, one in which the countries are adjusted to basis, and one in which they are adjusted to basis. Three years have been covered in this study, namely, 1938, I948, and I953. The tables for I938 and I953 are shown in full in the Appendix. The tables for I948 can be supplied by the writer upon request. Concentration of trade. Before proceeding to analyze statistically the basic tables, certain features are immediately apparent. One of these is the abnormality of the distribution of the figures. Consider, for example, Appendix Table ia and examine the column for, say, Sweden. Fourteen other countries are listed as exporters to Sweden, so that if Sweden's imports were distributed evenly among them, each would account for 7.15%o of her imports. (The discussion will proceed in terms of the equal importer or equal exporter basis unless otherwise specified.) If the distribution were not perfectly even but varied more or less according to some normal distribution, one would expect an accumulation of figures round the 7 per cent level, with few figures at either very high or very low percentages. In fact, as can be seen from the table, the reverse is the case. The figures seem to accumulate at the extremes at high percentages (i6.7% from Iceland, i6.7 % from Norway, I2.0 % from Germany) or at low percentages (I.07% from Greece, 0.07% from Ireland, 2.4 % f rom' Turkey).