Abstract
The objectives of this paper are to test the relaltive version of the law of one price in the short and the long run at various levels of aggregation for traded goods. The use of an error-correction model is made to test the validity of the hypothesis in the short run with a built-in tendency to one price in the long run. Using unit value trade data at the aggregate, 2-digit and 3-digit levels of the Standard International Trade Classification, the law of one price is generally rejected in the short run, although a long-run proportional relatlionship between prices of fairy homogeneous products are used the hypothesis is rejected both in the short and long run. This suggests that non-price changes which are likely to be incorporated in unit values are probably gradual and affect EEC countries in a similar manner. Price changes, however, are somewhat erratic and hence with unitvalue data, the law of one price in the long run is not contradicted, while it is refuted at the disaggregate level when price data are used.