Deep Trade Agreements and Trade in Value- Added: Does Heterogeneity Matter?

This paper examines the heterogeneous impact of Global Value Chains (GVC) integration on singing deep trade agreements. Following Fontagné and Santoni (2021), we contribute to the literature in two ways. First, we extend their analysis by examining not only the impact of GVC on signing an agreement, but on the likelihood of signing a deep one. Second, for GVC, we distinguish between backward participation (value added in exports that rely on foreign inputs) and forward linkages (value added that is embodied in exports of other countries). We also distinguish whether GVC occurs in goods or services. To do so, we combine the Eora dataset with the Deep Trade Agreement database (World Bank) using a gravity-type model. Our main findings show that GVC increases the likelihood of signing a deep agreement that focuses on goods and services. More particularly, an increase in GVC by 1% increases the probability of signing a PTA by 2.9 percentage points. More particularly, backward linkages (relative to forward) are more likely to a↵ect the depth of the agreement as they increase the latter by 3.3 percentage points. Our results remain robust after running a set of sensitivity tests. In addition, we show that the impact of GVC is greater on the depth of provisions related to NTMs followed by services. When the development level is considered, GVC flows create a new motive to deepen trade agreements between countries that belong to di↵erent income groups.