(en Anglais seulement pour le moment)The study analyses sectoral specialization as a response to trade liberalization for three major trade areas, the Mediterranean Partner Countries (MPC) as a part of the EuroMed free trade area, the ASEAN countries and the Mercosur countries. In all cases, we analyze trade with the European Union to ensure that our results are not due to different characteristics of the trading partner.For the MPCs, a very clear pattern emerges from our analysis: First, the production of beverages and tobacco is expanding as a consequence of trade liberalization. Second, the same is true for everything related to the metal manufacturing value added chain, including the production of metals and even of metal working machinery. Third, our results indicate that rubber production and the manufacturing of rubber has also been stimulated by trade liberalization. Since a large part of the manufacturing of rubber is the production of tires, this connects nicely to metal production aimed at manufacturing parts of road vehicles.
On the other hand, trade liberalization in MPCs seems to have hit chemicals and related products. There is clear evidence of negative specialization, i. e. of a comparably smaller role for domestic production of these products. This may be the consequence of a highly price elastic domestic demand which shifted to cheaper imports. Also, with the exception of metal and rubber, other crude materials and their manufacturing seem to have shrunk in relative terms as a response to freer trade. The same is true for food production with the notable exception of fish.For the ASEAN countries, our analysis has revealed a trade block in a more mature phase of adjusting to world market conditions. While in the MENA states we see broad sectors of the economies moving in the same direction (e. g. an expansion of metal and rubber industries at the expense of chemical and food products), such developments are absent in ASEAN states. It may well be that these have taken place in earlier phases of ASEANs history, i. e. in the twenty years preceding our sample, i. e. in the years pre?1988. As of 1988, ASEAN economies seem to be roughly in line with world market conditions, remaining levels of protection vis?à?vis the EU not withstanding. Consequently, further trade liberalizing measures do not result in more or less uniform reactions across broadly defined industries. Rather, we observe finer adjustments where some specific industries shrink and other, closely related industries expand.This result has important consequences for labor market policies. Laid?off workers with industry?specific human capital will have great trouble finding a suitable new job at a similar level of compensation if whole industries are affected in the same way by trade liberalization. Hence the surge of unemployment in the shrinking industries may be very protracted, because the unemployment which develops is deeply structural due to factor specificity. Hence, labor market policies are well advised to focus on groups of unemployed with low qualifications or unspecific skills ? or offer extensive (and expensive) retraining.On the other hand, in the response of a mature free?trade area such as ASEAN expanding and shrinking sectors may be in the same broad value added chain. For instance, we find that in ASEAN countries the production of crude cork and wood is expanding while the manufacturing of these crudes is shrinking. Workers laid off in the latter should have less trouble finding an appropriate new job in the former and labor market policies can actively support the transition of people with specific human capital to a related production sector.For the Mercosur countries, we find a single industry which seems to be broadly expanding (production of machinery, in particular office machines, data processing, telecommunication, sound recording, electrical machinery). Other responses are heterogenous within a single broad industry. This result is very much in line with our previous interpretation: The Merecosur agreement entered into force in 1999 (roughly ten years earlier than the Association Agreement with the MENA countries), but much later than the ASEAN treaty. Hence, the finding for Mercosur states is the finding one would expect for a trade block which is in the transition between the initial and the mature phase of a free trade area.This interpretation is good news for the MPCs. If realignments in a country’s industry structure and their repercussions on unemployment are the main problem associated with freer trade, then our results indicate that the MPCs have walked a good way of the rough part of the road already. Industries have undergone some broad adjustments and it is fairly clear to identify the winners and the loosers. Future developments ? as far as they are related to trade liberalization ? should be easier to manage in terms of domestic policies than what the MENA countries have already gone through. In particular, deeply rooted structural unemployment due to the competitive pressures of the world market are less likely to occur in the future and labor markets should, in principle, find it easier to clear than they may have done in recent years.