Abstract (EN)

For several years, preferential trade agreements have been oriented towards initiating a deep integration process. As a result, treaties include clauses concerning the harmonization of regulations, policies and standards. In order to join the European Union, a future member must be a stable democracy that respects human rights, the rule of law and the protection of minorities. The inclusion of such a clause can be interpreted as a political choice to promote democracy. But it can also be viewed as a pre-condition for implanting a "deep integration" process : only democracies are able to drive the harmonization process of rules and standards. If democratic constraints imply deeper integration, we can expect that such agreements will have a higher impact on trade inside the area concerned by the agreement. The aim of this paper is to verify that preferential trade agreements (PTAs), which are bound by a democratic constraint, have a higher positive effect on trade inside the area, than non-constraining agreements. To achieve this, we use an Anderson and van Wincoop (2003) type of gravity model, and we differentiate between PTAs according to whether or not they include a democratic clause. Whereas trade between democratic countries is higher than trade between autocratic countries, empirical evidence shows that the inclusion of a democratic clause fosters bilateral trade between the partners has no significant effect relative to no PTA, and is neutral relative to a non-democratic PTAs between South countries.