Report on:Trade Liberalization and Jobs in the Mediterranean:Towards a New Generation of Trade Agreements
FEMISE and the Center for Mediterranean Integration (CMI),International Trade Centre (ITC) and Economic Research Forum (ERF), are pleased to announce the publication of their joint report on “Trade Liberalization and Jobs in the Mediterranean: Towards a New Generation of Trade Agreements”.
In 1995, policymakers from around the Mediterranean region met in Barcelona and issued a key communiqué, the “Barcelona Declaration”.The document defined the framework for initiating renewed dialogue among the nations of Southern Europe, North Africa and the Levant, focusing on their common objectives of economic transformation and trade exchanges. This resulted in the EuroMediterranean Partnership. Since then, regional integration has been pursued primarily in the form of trade agreements, including the association agreements (AAs) signed between the European Union (EU) and the four countries studied in this report: Egypt, Jordan, Morocco and Tunisia (the Southern and Eastern Mediterranean Countries– SEMCs) as well as Algeria, Israel and Lebanon. Liberalization has been progressive, as seen in the SEMCs having reduced customs duties on industrial products from the EU while benefiting from a gradual transition period of 12–15 years. Intermediate goods from the EU were the first to be liberalized, followed by final products at a later stage.
This two-step liberalization was intended to provide SEMCs with time to improve their efficiency and production capacities and to adjust. The economic rationale was that trade would benefit all its partners through the exploitation of each nation’s comparative advantages, better access to markets, technological and know-how transfer, economies of scale and increased competition. By cutting tariffs on imported final products (“output tariffs”), a competitive effect was expected, even leading to an increase in productive efficiency. On the other hand, by cutting tariffs on imported intermediate products (“input tariffs”), a productivity increase was
expected, attributed to a direct cost reduction effect and an indirect effect through the introduction of new imported products, both of them considered favourable for industry growth. Since then, SEMCs have considerably advanced in reducing import tariffs.In terms of this liberalization’s effects on trade flows,even if the opening/reduction of the SEMCs tariffs has privileged products from the EU, the trade deficit vis-à-vis the latter has not increased as much as the trade deficit vis-à-vis the rest of the world.The evidence presented in this report indeed shows that SEMCs have undoubtedly faced a significant pro-competitive effect, though one probably more pronounced from the rest of the world than from the EU.
In terms of job creation, the implementation of AAs was also expected to have positive effects on SEMCs, although such an objective was not explicitly recognized. The existing theoretical literature and empirical evidence are not unanimous on these issues. Perhaps the most important theoretical finding is that the relationship between trade liberalization and job creation cannot be analysed in isolation from other factors affecting both trade policy and labour markets. Since Melitz (2003), some of the most relevant theoretical frameworks for understanding the effects of openness are those that admit the heterogeneity of firms. The consideration of imperfectly flexible labour markets in these theoretical frameworks has shown that, following openness, one could expect in developing countries in particular: (i) an increase in wage inequality between workers across firms, (ii) an increase in unemployment, and (iii) an increase in the demand of skilled and educated workers.
Globally, the empirical literature focused on developing and emerging countries points to
potential negative effects of trade liberalization on overall employment. Empirical studies have
specifically shown the following two links. Firstly, trade liberalization leads to labour reallocation across firms, across industries and less frequently across geographical regions. These reallocations not only take time but can also be further slowed down by the presence of market frictions and the costs involved. The evidence shows that, in developing and emerging countries, job creation may not offset job losses as a result of trade liberalization. Secondly, the causal mechanism between trade and jobs, which can lead to sectors being “winners” or “losers”, may not be linear and depends on numerous other variables. In any case, the effects of trade liberalization on employment remain specific to the country and sector.
As the main change for SEMCs following the introduction of the AAs was the decline in their
import tariffs vis-à-vis European countries, the report puts into perspective the changes in the EU’s effective preferential margin and the changes in the sectoral shares of total manufacturing employment.
In terms of the employment effects of the AAs signed between SEMCs and the EU, our analysis
shows that the sectors in which the shares in total employment have decreased the most (“losers”) are: basic metals (for all our SEMCs), textile and tobacco (except for Tunisia), coke and refined petroleum (except for Egypt), food (for Jordan and Tunisia), non-metallic mineral products (for Morocco and Tunisia), and chemicals and furniture (for Tunisia).
In terms of the sectors that experienced employment creation (winners), the electrical machinery and apparatus sector is a clear “double winner” (simultaneously benefited from an increase in value added and an increase in employment following the AAs) in all countries but Jordan. Meanwhile, fabricated metal products are a double winner in Jordan and Morocco. The wearing apparel sector and the leather products sector are also double winners in Jordan and Tunisia. These results would indicate that, in these sectors, an improvement in the preferential margins on imports from the EU went hand in hand with an increase in employment.
Nevertheless, a causal link cannot be identified and in no case can it be inferred that better access to the European products explains these variations in the sectoral structure of employment in SEMCs.
At the country level, in the case of Egypt and Tunisia we find a concomitance between the improvement in the effective preferential margin in favour of the EU and a decline in industrial employment shares. In Jordan and Morocco, however, this relationship appears to be positive, which means that an increase in the preferential margin for European products has gone hand in hand with an increase in the sectoral share of industrial employment.
Overall, SEMCs are currently lagging behind peers in non-oil international trade flows, still struggle to attract foreign direct investment (FDI) and, most importantly, face persistent unemployment, informality and low female labour-force participation. In view of these outcomes, we try to address the following question: What have been the effects of the AAs signed between SEMCs and the EU in terms of trade flows and key labourmarket outcomes?
Overall, it should be noted that there are important limits to available data, which have had considerable implications on the methodology of analysis. Employment data are not available by country at a sufficiently disaggregated level, which hinders empirical measurement of the effects of AAs on employment creation in SEMCs. Given the data limitations, an econometric and descriptive analysis of the causal link between the AAs and the results in the SEMCs, in employment terms, could not be carried out. Nonetheless, the analysis throughout the report is as exhaustive as possible, drawing on theoretical and empirical knowledge from the literature on several determinants affecting the trade and jobs link and making descriptive use of available data.