Three essays on shallow and deep regional trade agreements

Authors: Sawadogo, Eli
Advisor: Larue, BrunoGaigné, Carl
Abstract: This thesis consists of three essays dealing with issues related to the effects of regional trade agreements and border effects on trade flows and welfare in member and non-member countries. It adapts and develops econometric methods to empirically analyse hypotheses on the heterogeneous distribution of trade creation gains, welfare and border effects. The first chapter analyses the asymmetry of trade creation effects by considering both within-agreement and between-agreement types on the one hand and by decomposing the explanatory factors contributing to the heterogeneity on the other. The second chapter focuses on the general equilibrium analysis of the relevance of deep integration dynamics in a hypothetical perspective by studying the case of the African Continental Free Trade Area. The third chapter focuses on the heterogeneity of border effects by looking at the effects of different types of regional trade agreements. In Chapter 1, we start from the observation that the level of integration and the size of RTAs may not suit all members equally and one would expect the benefits of integration to be distributed asymmetrically across member countries. Smaller countries generally benefit more from trade liberalisation because they enjoy a greater improvement in the terms of trade and lower trade costs because their consumers spend a greater share of their income on imported goods. RTAs remove uncertainty about market access by making it more difficult for large countries to engage in opportunistic behaviour. In addition to country size, the benefits of RTAs are likely to vary from country to country due to initial trade policies, institutions, infrastructure and geographical determinants. Given that deeper integration entails greater losses in policy flexibility, one would expect membership selection to be stricter for deeper RTAs and trade gains to be distributed more symmetrically among the members of these RTAs. We draw on the two-stage econometric approach of Baier et al. (2019). Our results show that there is considerable heterogeneity in the distribution of trade gains within RTAs, with some countries even experiencing losses, highlighting the second-best nature of RTAs. Specifically, we find that the pairwise directional effects of an FTA are more heterogeneous than the pairwise directional effects of CUs, which in turn are less heterogeneous than those for MCs and MUs. This means that countries in deeper RTAs tend to experience more similar trade creation effects. Our conclusion is consistent with our theoretical expectations. Furthermore, we find that many factors explains heterogeneity in trade creation between pairs of countries belonging to a given RTA. In the second chapter, we use a general equilibrium structural gravity model to generate four scenarios for the implementation of the AfCFTA from a hypothetical perspective. The first scenario concerns the effects of the AfCFTA as an FTA. In the second scenario, we assume that the AfCFTA becomes a CU with a common external tariff. The third scenario concerns the CM status and the fourth experiment is about of MU status of AfCFTA in relation to the first general objective of Article 3 of the AfCFTA texts. In all four scenarios, we are able to stimulate the impact of the four types of RTAs on member countries and their main trading partners on trade flows, multilateral resistance indices, factory prices and welfare. Our findings indicate that countries gain differently depending on the level of integration. Some countries reach their optimal level of integration in CU. Other countries maximise their trade gains when AfCFTA is a CM. Our results also show a non-monotonic situation for a group of countries, notably the majority of West African countries. We have used manufactured goods and considered that production has zero elasticity. Further research could carry out analyses with several sectors considering endogeneity of output. In Chapter 3, we extend the econometric approach of Larch et al. (2019) to investigate empirically the incentives of governments to make adjustments in external tariffs after different types of trade agreements. The economics of regionalism provides explanations for the decreases and increases in external tariffs induced by participation in a trade agreement. DUs and deeper forms of integration lead members to set common external tariffs. This is certainly true for EU members, but deviations from this rule are observed in other cases. Leaving aside the possibility of deviations, coordination has important implications that clearly distinguish CUs, CMs and MUs from FTAs. FTA members may set different external tariffs due to rules of origin. Weak enforcement and/or low domestic content would encourage FTA members to set fairly similar external tariffs. Otherwise, countries with higher external tariffs would lose tariff revenues on external trade to FTA members with lower external tariffs, when transport costs between FTA members are low. Thus, how the thickness of the border adjusts after the implementation of different types of RTAs is an empirical question. Our results show that, on average, importers in an FTA impose 19.52% additional barriers on non-members, while those in a DU impose 56.54% additional barriers. This result seems to reflect a certain reality since DU members tend to impose more barriers (external tariffs and non-tariff measures) on non-members. Importers from a MC tend to facilitate about 27.63% worth of barriers for non-members. This result could be explained by the size effect due to the presence of the 26 EU member countries. As the number of members increases, the common tariff tends to decrease. Importing countries that are members of the CU tend to impose 26.41%more barriers on non-members.
Document Type: Thèse de doctorat
Issue Date: 2021
Open Access Date: 20 December 2021
Grantor: Université Laval
Collection:Thèses et mémoires

Files in this item:
Description SizeFormat 
37492.pdf1.7 MBAdobe PDFThumbnail
All documents in CorpusUL are protected by Copyright Act of Canada.