Modelling of Small Countries in Economic Integration Processes: An Extension of the Armington Approach

The focus of this paper is a process of ‘deep’ economic integration. Economic integration is not merely a removal of trade barriers but also increased competition because goods become more homogeneous due to, for example, harmonization of standards. To accommodate this ‘deep’ integration process we changed the preference structure for consumers in such a way that products from EU regions are considered as closer substitutes. The enlargement of the European Union in Eastern direction has therefore profound implications for Eastern European countries. Next to the reduction in trade barriers their commodities will become closer substitutes in the long run. The empirical analyses shows that the welfare gains for Eastern European countries of increased substitutability of their products with EU products are much larger than the welfare gains from removing the trade barriers. A second focus of this paper is the magnitude of the terms of trade effects in an Armington world that are often considered as being excessive. An Armington world implies national product differentiation and implicitly monopoly power for all countries. The latter is also true for small countries. Unilateral Trade liberalization experiments in such a world have often a negative welfare impact due to these excessive terms of trade effects. By increasing the degree of competition within the EU Union the new preference structure reduces this problem. The total impact of the new structure is ambiguous Trade liberalization experiments show that the values of the Armington elasticities in the new preference structure are crucial for production and welfare effects. In case of higher substitutability within the EU and lower between EU and non-EU products on EU markets, multilateral trade liberalization improves terms of trade for EU countries and deteriorates it for other countries. The opposite is true in case of increased competition ‘only’ within the EU market. In this paper we made some ‘simple’ assumptions about the values of the Armington parameters. A better estimation of these parameters is crucial if one performs policy analyses with this model.