The European Union was particularly hard hit by the sharp decline in world trade during the fourth quarter of 2008 and the first quarter of 2009. In 2009, the EU's merchandise exports into third countries shrank by 14.8 per cent in real terms, compared to 12.2 per cent for global exports. During this period, Germany also lost its lead position in world trade. The German share (in nominal terms) of global merchandise exports (including internal exports in the EU) fell behind China's share, with 9 per cent as against 9.6 per cent, respectively. The EU as a whole nevertheless remains by far the world's largest exporter: in 2009, the share in global merchandise exports (not counting EU internal exports) of its exports into third countries was 16.2 per cent, surpassing China's share - 12.8 per cent - by a wide margin.1
Figure 1 (back to the text)
The European Union's Position in World Trade1
1 World trade excluding internal trade of the EU 27.
Source: WTO; own calculation.
Internal and External EU Trade Linkages
As can be seen in Figure 1, the EU's share of world trade has been slightly receding for some time now. This is true for both its internal and its external trade shares. At the same time, its share of total world trade including EU internal trade is more than twice its share with internal EU trade not counting as international trade. This clearly indicates the outstanding weight of the EU's internal market in the world economy.
Figure 2 shows the significance of cross-border economic transactions in the EU in terms of their GDP (gross domestic product) shares. It refers to the three main "compartments" of the international division of labour, i.e. international goods trade, international trade in services and foreign direct investment (FDI). The first three quadrants of the figure reflect EU transactions with third countries, while the bottom right quadrant represents cross-border transactions within the EU, i.e. the "external economy" in the internal market.
Two observations stand out: First, in each of the three areas, cross-border transactions have grown faster than all transactions while the intensity of those transactions, i.e. their ratio to the GDP, increases from services to goods to FDI. Second, within the EU, the intensity of cross-border links is substantially higher even than those between the EU and third countries, which again is evidence of the paramount importance of the EU's internal market in the globalised world economy.
Figure 2 (back to the text)
Economic Significance in the EU of Cross-Border External and Internal Transactions
Share (in %) of EU gross domestic product, 2004 and 2008
Source: Eurostat, UNCTAD, WTO; own calculation.
Two Dimensions of EU External Trade Policy
The internal market and the policies related to it are a reference for the common trade policy of the EU member states towards third countries. This policy has a strong impact on the EU's external economic links, which is all the more true as the policy no longer just covers cross-border merchandise flows but likewise the exchange of services and FDI activities. Moreover, its reach also includes cross-sector issues such as the protection of intellectual property rights in international trade. In all these respects, external trade policy has been inspired by the EU's experience with its own internal market.
The common external trade policy takes place on three tracks - unilaterally, bilaterally and multilaterally - which interact with one another. In particular, the multilateral track defines the degree of freedom for policymakers on the two other tracks. The options of unilateral trade policy are also constrained by the EU's proceedings on the bilateral track. Together, the three tracks form the vertical dimension of trade policy or "competitive liberalisation", i.e. competition between different liberalisation or regulatory strategies within one and the same jurisdiction.2 In the present survey, the focus will be on the horizontal dimension of trade policy or "competitive regionalism". This notion refers to the competition between different jurisdictions which seek strategic advantages for themselves through the conclusion of bilateral agreements with "priority" trading partners. Together with a strong emphasis on "deep integration", it also reflects the main thrust of the EU's new trade strategy as announced in October 2006.3
Our analysis proceeds in two stages. First comes an outline of the new trade strategy in the light of the changes in EU external trade policy caused by the Treaty of Lisbon. The second step is a detailed "positioning" of the EU in the "geography" of international trade policy.
Impact of the Lisbon Treaty
The Treaty of Lisbon, in force since 1 December 2009, has an influence on the EU's external trade policy in three major respects4: First, it extends and clarifies the EU competences in trade policy. Second, it strengthens the role of the European Parliament in this policy area. Third, it makes trade policy an integral part of the EU's "unified external action".
According to the Treaty on the Functioning of the European Union, which is the former Treaty on the European Community as amended by the Treaty of Lisbon, the European Union now has exclusive competence (in contrast to shared competence among the member states and the EU bodies) in practically all trade-related policy areas. These areas comprise trade in goods and services as well as foreign direct investment and the transfer of technology to trading partners. The main "innovation" of the Lisbon Treaty in this regard was to add FDI to the exclusive EU competences, after the Treaty of Nice (2003) had already done the same with services trade and intellectual property rights (IPRs) or technology transfer. For the EU's new trade strategy, this in particular means that the European Commission and the Council of the European Union (or, by its old name, the Council of Ministers) are now free to negotiate and conclude bilateral trade agreements with third countries involving "deep integration" in areas as diverse as services, FDI and IPRs.5
The Treaty of Lisbon also enhances the role of the European Parliament in EU trade policy. First and foremost, the Parliament now shares power with the Council of Ministers in basic legislation on trade matters such as the adoption of regulations concerning antidumping and safeguard actions or the granting of trade preferences to developing countries. The pertinent legal instrument is the "ordinary legislative procedure", which replaces the former "co-decision procedure" and is now the standard decision-making method in trade legislation. Through the Lisbon Treaty, the Parliament also becomes more deeply involved in the process of trade negotiations conducted by the Commission, since it is now entitled to be informed about the progress of such negotiations on a regular basis. Last but not least, all trade agreements with third countries to be concluded by the Council now require the prior consent of the Parliament, which is also responsible for the final ratification of these agreements. Altogether, the higher profile of the European Parliament in trade policy augments its clout in shaping the new trade strategy of the EU.
The same is true with regard to the integration of trade policy into what is now called the EU's unified or coherent "external action". Accordingly, all EU policies with a bearing on relations to third countries should be guided by a common set of principles and objectives.6 Among these are general aims, such as to "consolidate and support democracy" or to "preserve peace", as well as more specific aims like to "foster the sustainable economic, social and environmental development of developing countries, with the primary aim of eradicating poverty" or to "encourage the integration of all countries into the world economy, including through the progressive abolition of restrictions on international trade".7 This raises the question as to whether trade and the conclusion of trade agreements are in the Lisbon Treaty in order to increasingly promote non-trade objectives (economic development, protection of the environment, etc.) and thus to further promote "deep integration" between the EU and its partner countries.
The EU's New Trade Strategy
When the EU ended its self-imposed moratorium (from 1999) on the negotiation of new bilateral trade agreements in 2006, one reason was due to the observation that other trading powers were becoming increasingly engaged in bilateral and regional trade policy. This in particular holds for the United States of America. The USA was the "champion of multilateralism for the first four decades of the GATT (General Agreement on Tariffs and Trade)"8 but started to change course in the late 1980s and early 1990s through the conclusion of free trade agreements with Canada and Mexico. Its turn to the bilateral and regional routes in trade policy received a strong boost by the failed WTO conferences on multilateral trade liberalisation in Seattle (1999) and Cancún (2003).9
Another major reason for the EU to re-embark on bilateralism in trade policy was its desire to enhance or "deepen" the substance of trade agreements. This motive also had to do with shortcomings of the multilateral trading system, evidence of which includes the failed attempt by the European Community to launch a wide-ranging "Millennium Round" in the WTO in the late 1990s and the apparent inability of the Doha Round to substantially advance the "new trade agenda" as symbolised by the "Singapore issues" (competition, foreign direct investment, public procurement and trade facilitation). The EU therefore decided to push "deep integration" via its bilateral trade policy track.
Accordingly, the "new generation of Free Trade Agreements (FTAs)"10, as envisaged in the EU's new trade strategy, is in essence a response to two challenges facing the EU:
The first challenge is the "race for markets" between the leading trading powers in the world, and thus for the EU to avoid falling behind major competitors in cutting bilateral trade deals. In this context, the European Commission points to "a growing risk of trade diversion detrimental to the EU in the most dynamic countries" as "several of our main trading partners and priority targets have been negotiating FTAs with our competitors (e.g. ASEAN (Association of South-East Asian Nations) members with Japan or Korea with the US)." Accordingly, "the rapid development of third countries concluding FTAs with the EU's main competitors such as the USA or Japan carries risks of marginalising the EU".11
The second challenge for the EU is to realise "deep integration" with its trading partners and, with regard to its competitors, to "seek full parity at least"12 in this respect. A case in point is the policy on foreign direct investment which, as noted above, has become part of the EU's exclusive competence in trade policy under the Lisbon Treaty.13 Other policy issues explicitly mentioned by the European Commission in this context include competition, the regulation and deregulation of the services sector, technical barriers to trade, sanitary and phytosanitary trade measures, government procurement policies, the enforcement of intellectual property rights and the elimination of all forms of duties, taxes, charges and restrictions on exports. Furthermore, the Commission holds that "future FTAs will also need to cover sustainable development concerns, by involving public participation, including in ensuring that social and environmental commitments are fully implemented".14 In general, the new bilateral trade agreements of the EU are meant to be "WTO plus", i.e. exceed the scope of the agreements already reached in the WTO through both the opening of new policy areas and the deepening or strengthening of existing provisions, obligations and disciplines.
For the selection of partner countries or "priority" partners, two criteria shall be decisive. One is the market potential, expressed by economic size and expected growth, of the prospective partner country. The second refers to the level of protection in the form of tariff, non-tariff and regulatory barriers against EU export interests. Hence the preferred partners are those countries where domestic markets are large and dynamic and where EU firms face comparative disadvantages in competition. Based on these criteria, the European Commission has identified ASEAN, South Korea and MERCOSUR (Mercado Común del Sur) as the highest-priority countries or country groups, followed by India, Russia and the Gulf Co-operation Council.15 Before the EU's bilateralism and regionalism is discussed in greater detail, some background on general tendencies in this area is in order.
Intra- and Inter-Regional Trade Flows
In 2009, the world economy experienced dramatic problems. As noted earlier, the global trade volume decreased by 12.2 per cent compared to 2008. This was the sharpest decline in global merchandise trade since the Second World War. It also exceeded the decline of the global GDP in 2009 (-2.3 per cent) by a multiple.
The slump in global merchandise trade followed a phase characterised by an almost steadily increasing integration of the world economy. The international division of labour has greatly deepened during the past decades. While the volume of global trade in 2008 was more than 30 times higher than in 1950, the corresponding volume of global production "only" increased by a factor of eight.
The dynamics of international trade reflects growing economic linkages among nations within an increasingly complex structure of intra- and inter-regional trade flows. For our purpose, it is useful to distinguish between the following regions: European Union; Non-EU-Europe (including the former Soviet Union); North America (USA, Canada, Mexico); Asia-Pacific (East and South Asia, Australia, New Zealand); Middle East (also including North Africa); Latin America (except Mexico) and Africa (except North Africa). Figure 3 shows the merchandise trade flows within and between these regions in the years 1990 and 2008.
Figure 3 (back to the text)
Matrix of International Trade
1990 and 2008
Source: IMF; own calculation.
The high and even increasing global economic importance of the EU, North America and Asia-Pacific largely results from economic integration within these regions, whereas the Middle East, Latin America and Africa do not focus on intra-regional trade to such a great extent. In view of the strong inter-regional trade linkages between the EU, North America and Asia-Pacific, the three regions can also be seen as the dominant triad in world trade. Figure 3 also indicates substantial inter-regional trade flows between the EU and Non-EU-Europe, the Middle East and Asia-Pacific as well as between North and Latin America. Two factors play a key role here: the dependence of the "North" on the natural resources of the "South" and the (physical and "cultural") distance between the regions.
Preferential Trade Agreements at a Glance
The intra- and inter-regional network of trade is matched by a worldwide web of Preferential Trade Agreements (PTAs).16 PTAs are bilateral or plurilateral agreements between countries or groups of countries within a region or across regions which provide for the removal of trade and other economic barriers between the partners of the agreement on a reciprocal basis.17 The liberalisation may thereby take on a broad variety of intensities, ranging from "shallow integration", which concentrates on the removal of tariff and other border barriers to trade, to "deep integration", where the emphasis is on reducing the cross-border competitive distortions caused by behind-the-border policies or regulations.18
Figure 4 depicts the proliferation of PTAs between the periods 1958-1989 and 1990-2010. It relies on the numbers of PTAs that have been notified to the GATT/WTO since 1958 and are still valid at present.19 It is shown that the overwhelming majority of bilateral and regional PTAs have come into force since 1990. In a historical perspective, this development in particular points to a "revival" of bilateralism in trade policy.
Figure 4 (back to the text)
Preferential Trade Agreements1
1 Number of PTAs notified to the WTO by date of entry into force (cut-off date: 19 July 2010).
Source: WTO; own compilation.
Bilateralism in international trade is an old phenomenon that preceded the emergence of a multilateral trading system by decades. In the late 19th and the early 20th century, bilateralism was dominant in trade politics. It took off with the Cobden-Chevalier Treaty (1870) between the United Kingdom and France, around which an intricate system of interlinked bilateral trade agreements in Europe was built. Later, the USA joined this trend through the Reciprocal Trade Agreements Act of 1934. However, unilateral protectionism had already gained the upper hand in trade policy at that time.
After the Second World War, bilateralism was for a while practically absent from the political scene, where multilateralism under the GATT now prevailed. In parallel, "old regionalism" arose in the 1950s and 1960s. Its geographic foci were Europe and Latin America. In substance, "old regionalism" was about "shallow" as opposed to "deep integration", and in Latin America it mainly consisted of import-substituting industrialisation ("closed regionalism").
With the advent of the "new regionalism" in the late 1980s/early 1990s, things again changed fundamentally: bilateralism was "resurrected" in a global frame, in a more open manner ("open regionalism") and with a new emphasis on "deep integration".
The proliferation of "Economic Integration Agreements" in the services sector since 1990, as displayed in Figure 4, is a strong indicator of the increasing significance of "deep integration".20 At the same time, the figure shows that in the goods sector, Free Trade Areas have clearly outpaced Customs Unions as the preferred integration type. At first sight, this finding appears to be counterintuitive, since Customs Unions (with a common trade policy towards third countries) are usually understood to represent a deeper form of economic integration than Free Trade Areas (where the member countries preserve autonomy in trade policy). In actual fact, however, a steadily increasing depth of integration in Free Trade Areas is observable over time.
Figure 5 (back to the text)
Preferential Trade Agreements by Partners' Stage of Development and Geography1
1 Number of PTAs notified to the WTO by date of entry into force (cut-off date: 19 July 2010).
Source: WTO; own compilation.
Global Bilateralism with "Hubs" and "Spokes"
Figure 5 highlights core features of the new bilateralism. Most striking are the sharp increase from 1958-1989 to 1990-2010 of PTAs among developed and developing countries (North-South agreements) and of PTAs transcending regional limits (cross-regional agreements).21 Both developments are closely linked, indicating that bilateralism or regionalism is indeed going global.
In line with these trends, a "hub-and-spoke" system of PTAs has emerged on a global scale and in the major regions of the world economy. It is characterised by a few "centres" building preferential trade links with a number of "peripheries" which for their part are rather loosely connected with one another through PTAs. Concerning such relations among the central regions, an asymmetric pattern can be observed: whereas presently no PTA links exist between the leading trading powers, PTAs are increasingly concluded or negotiated between larger countries or country groupings in one region and smaller ones in another region. Examples are Mexico, which has PTAs with the EU and Japan, and South Korea, which has (pending) PTAs with the EU and the USA.
An example of a regional "hub-and-spoke" system is the European-Mediterranean trading area. In this case, the bilateral trade relations between the EU and non-EU Mediterranean countries have been liberalised to a high degree while involving a stark measure of asymmetry: whereas the Mediterranean exporters depend heavily on the EU market, their respective home markets are rather negligible for exporting EU firms. At the same time, the Mediterranean countries typically still maintain relatively high barriers against each other's exports.22
In this context, two further observations are important. First, bilateral PTAs between countries from different regions in the world increasingly cut across - and thereby arguably impair - regional integration initiatives.23 Secondly, North-South agreements often contain more in the way of "deep integration" than do South-South agreements. A case in point is Latin America, where intra-regional integration schemes in some respects appear to be "shallower" than the bilateral trade agreements concluded with extra-regional developed countries. For instance, most intra-Latin American PTAs do not have specific services provisions.24
Europe as a Model
The European Union's place on the global map of regionalism and bilateralism and its change over time is unique. This is true in terms of substance, or of the guiding ideas of the development, as well as in terms of the sheer numbers of respective PTAs. It holds for both the old and the new regionalism.
The Customs Union, which the original European Economic Community had established as early as mid-1968, epitomised "old regionalism" and "shallow integration". Indeed, the Treaty of Rome (1957) is the oldest Regional Trade Agreement notified to the GATT and still in force today. Moreover, in the second half of the 1980s, the now European Community was the harbinger of "new regionalism" with "deep integration", for which the "single market programme" guided by the "Cassis de Dijon" principle served as a model.25
Looking at the PTA numbers, Europe was the undisputed dominant force of regionalism in 1958-1989. The European Union has been a big player in this field in 1990-2010, too, alongside two other major world regions, as defined in Figure 3, that have become its main competitors, i.e. North America and Asia-Pacific. This is also shown in Figure 4.26
Figure 6 (back to the text)
Reciprocal PTAs in and with Europe1
1 EAC=East African Community; EEA=European Economic Area; EFTA=European Free Trade Association; EPA=Economic Partnership Agreement; ESA=Eastern and South Africa; SACU=South African Customs Union; SADC=South African Development Community.
Cut-off date: 19 July 2010.
Source: WTO, EU, EFTA, own compilation.
The European "Spaghetti Bowl"
Figure 6 displays the pattern of EU PTAs forming the core part of the European PTA "spaghetti bowl".27 At its centre is the EU's internal market, representing the highest level of economic integration. Below this "internal integration", several layers of EU integration with third countries ("external integration") can be distinguished:
First is the European Economic Area with Iceland, Liechtenstein and Norway, which comes close to the internal market, since the three EFTA (European Free Trade Association) members named have essentially transposed the EU aquis into their national legislations (except agricultural policy).
The next layer of external integration comprises what may be called the "other European neighbourhood": Switzerland (the remaining EFTA member), Andorra, San Marino, member states of the former Soviet Union, Balkan and Mediterranean countries, Arabic Gulf states and the ACP (Africa, Caribbean, Pacific) group of countries. The respective agreements (in force or under negotiation) appear in a variety of forms such as Customs Unions, Free Trade Areas, Stabilisation and Association Agreements or Economic Partnership Agreements (EPAs). Notably, the Euro-Mediterranean agreements and the EPAs with different sub-groups of ACP countries contain strong elements of development and political cooperation in addition to the common trade provisions.
Finally, and most important in the context of the EU's new trade strategy, PTAs with geographically more distant trading partners in Latin America and Asia-Pacific form the third external integration layer. As noted, the EU seeks to make these bilateral "new generation" agreements significantly "WTO plus", while at the same time encouraging regional integration among its partner countries along similar lines. In fact, however, the bilateral strategy increasingly seems to dominate the regional route, owing mainly to weak institutions which stand in the way of "deep integration" in these regions (as among the ACP sub-groups mentioned above).
Table 1 (back to the text)
Share of Bilateral Preferential Trade in Total External Trade of the EU and its Partner Countries
Share in Exports
Share in Imports
|Bosnia and Herzegovina||0.3%||72.4%||0.1%||62.0%|
1 Antigua and Barbuda, Bahamas, Barbados, Belize, Dominica, Dominican Republic, Grenada, Guyana, Jamaica, Saint Kitts and Nevis, Saint Lucia, Saint Vincent and the Grenadines, Suriname, Trinidad and Tobago.
2 Including Liechtenstein.
Cut-off date for counting of agreements: 22 July 2010.
Sources: IMF, WTO; own calcuation and compilation.
Partners and Competitors
Tables 1 and 2 indicate the quantitative importance of bilateral preferential trade for the EU and its respective trading partners, and the degree of "overlapping" in preferential trade between the EU and its main competitors.28 First, it is shown that the individual preferential trade flows sum up to about one-fourth of the EU's total trade with third countries. For the average preferential trading partner, on the other hand, trade with the EU accounts for nearly half of the total trade (Table 1). This asymmetry mainly reflects the different sizes of the EU and its partner countries. Secondly, more than half of the EU's preferential trade overlaps with the preferential trade of other "hubs" (Table 2). This is most visible in the EU-EFTA relationship: whenever the EU or the EFTA signs a PTA with a third country, the EFTA or the EU signs a similar agreement.
Table 2 (back to the text)
Overlap of Regional and Bilateral Preferential Trade between the EU and its Competitors
Share in preferential EU trade1
1 Share (2008) of EU trade with "overlapping" partner countries in total EU preferential trade; 2 Including Liechtenstein.
Cut-off date for counting of agreements: 22 July 2010.
Source: IMF, WTO; own calculation and compilation.
Such competitive regionalism not only expresses itself in the numbers of overlapping PTAs, but also in the "substantive" strategies adopted by the main players in this field. This has been analysed in a painstaking comparative study of 14 EU and 14 US PTAs.29 The research highlights the incidence and legal enforceability of those (altogether 52) provisions in the agreements - typically of a regulatory nature - which either reconfirm or enhance existing WTO disciplines ("WTO+" provisions) or create qualitatively new commitments ("WTO-X" provisions). It is found that, in contrast to the US PTAs, the EU agreements contain a sizeable measure of "legal inflation", i.e. a high share of the provisions agreed to are not legally binding and thus not enforceable. This in particular holds for the "WTO-X" category. At the same time, the authors of the study also point to a major common element of the EU and US strategies in this area, suggesting that the bilateral trade agreements effectively serve as a means for the two "regulators of the world"30 to export their own regulatory approaches to their PTA partners.
Georg Koopmann, University of Hamburg and Hamburg Institute of International Economics (HWWI), Germany.
Marco Wilhelm, University of Hamburg, Germany.