Empirical relevance of the Hillman condition for revealed comparative advantage: 10 stylized facts

The theoretically necessary and sufficient condition for the correspondence between ‘revealed’comparative advantage and pre-trade relative prices derived by Hillman (1980) is analyzed empricially for virtually all countries of the world over an extended period of time. This yields 10 stylized facts, including that (i) violations of the Hillman condition are small as a share of the number of observations, but quite substantial as a share of the value of world exports, (ii) violations occur relatively frequent in the period 1970 – 1984 while they hardly ever occur in the period 1985 – 1997, and (iii) violations occur foremost in primary product and natural-resource intensive sectors, for sectors in countries in Africa, the Middle East, Latin America, and Eastern Europe. The condition appears also to be useful for identifying erroneous trade flow classifications.