The perception that European integration has not proceeded sufficiently fast over the last fifteen years is an important driving force behind `1992'. The theory of customs unions suggests that when members of a trading bloc reduce trade barriers among themselves and adopt common barriers against the rest of the world, there should be two main consequences for trade. The first is trade creation. Members of the union should start importing some commodities from other members instead of producing them domestically, because suppliers in other member countries will now face lower trade barriers. There should also be trade diversion, as suppliers in other union member countries take market share from suppliers in the rest of the world, whose trade costs have not been reduced.
In Discussion Paper No. 367, Research Fellow Damien Neven and Lars-Hendrik Röller assess the progress of European integration by examining data disaggregated into 29 sectors on the shares of EC and of non-EC manufactured imports in the apparent consumption of France, Germany, Italy and the United Kingdom during 1975-85. The data show that the shares of both EC and non-EC imports in apparent consumption have tended to increase over time. Far from slowing down, Neven and Röller argue, integration of trade flows within Europe has continued to proceed, at the same time as integration with the rest of the world. Only in the food industry has European integration occurred at the expense of world integration, probably because of the protectionist effects of the Common Agricultural Policy.
The authors do find, however, that the share of non-EC imports in apparent consumption has increased faster than the share of EC imports. But the level of EC imports in apparent consumption is systematically higher than the level of non-EC imports, and the slower growth in the former might just mean the level of EC imports is already so high that such rapid increases can no longer be expected. In order to test this explanation, Neven and Röller estimate an econometric model that accounts for the shares of EC and non-EC imports in apparent consumption in terms of a series of variables, including a proxy for the level of non-tariff barriers. If barriers to trade do hamper integration, they argue, then the estimates should show that the level of barriers has a significant impact on the shares of EC and non-EC imports in apparent consumption. Other variables in the model represent the factors underlying inter-industry trade (capital, labour and human capital intensity), the factors underlying intra-industry trade (economies of scale and product differentiation), transport costs and tariffs.
The estimates reveal that non-tariff barriers have a significantly stronger effect on trade between the EC and the rest of the world than on intra-Community trade. This lends support to the view that Europe is already well integrated, and that the relative slowing down of the share of EC imports in apparent consumption is not a matter of great concern. If this is so, Neven and Röller conclude, larger benefits could be obtained by opening up trade with the rest of the world than by pursuing the single market programme.

European Integration and Trade Flows
Damien J Neven and Lars-Hendrik Röller

Discussion Paper No. 367, February 1990 (IT)