Trade in Textiles and Clothing: Comparing Multilateral and Regional Free Trade Agreements
Richard Avisse 1 and Michel Fouquin 1 2
2
Associate Researcher, Centre d'Etudes Prospectives et d'Informations Internationales (CEPII). Deputy Director, Centre d'Etudes Prospectives et d'Informations Internationales (CEPII).
Summary The last fifteen years have seen the conclusion of significant negotiations on the multilateral level – with the Marrakesh agreement and China’s accession to WTO – and the regional level – with an upsurge in FTAs and similar accords. This paper analyzes the impact of such changes on the textile and clothing industries, which are likely to be the most affected. First, the paper analyzes current trends in world textiles trade, comparing regional and intercontinental trends and examining how the division of labor between textiles and clothing has developed. Second, it surveys the main studies using computable general equilibrium models. Third, it briefly introduces the global trade analysis project (GTAP) model and simulates two scenarios of multilateral liberalization. The paper then compares the results of these two scenarios using different versions of the GTAP database. Qualitative results are quite significant, whereas others are more elusive. Finally, the paper compares two regional FTAs. This paper is an extended version of a study prepared with the cooperation of the Institut Français de la Mode.
I. INTRODUCTION For the last fifteen years, in contrast to the situation that prevailed in the 1960s and 1970s, world trade in textiles and clothing has displayed a strong tendency to develop along regional rather than intercontinental lines. This regionalization of trade stemmed first from regional trade accords such as the Caribbean Basin Initiative (CBI) and the North American Free Trade Agreement (NAFTA) in the Americas, and from the “association agreements” between Mediterranean or Eastern European countries and the European Union (EU). It was also the result of the Multifibre Agreement (MFA), and of evolving comparative advantage in Asia between Japan, the first wave of newly industrialized countries (NICs) comprising Hong Kong, South Korea and Taiwan, the second wave of NICs (China and South-East Asia), and more recently the third wave (Vietnam and South Asia). The dismantling of the Multifibre Arrangement by 2005 and the accession of China (the world’s biggest clothing exporter) to the WTO will profoundly alter international
I
N
T
E
G
R
A
T
I
O
N
&
T
R
A
D
E
1
competition in textiles and clothing. A simulation of these two components of trade liberalization, using the global trade analysis project (GTAP) version 5 system, makes it possible to identify the broad scale of these two shocks. The simulations are: (i) quota suppression except for China; and (ii) quota suppression including China. The results are compared using, alternatively, GTAP4 and GTAP5. Two regional simulations are also undertaken: the potential for a pan-EuroMediterranean agreement compared to a free trade agreement (FTA) between Latin American countries and NAFTA. Among developing countries, only large Asian nations are expected to gain from multilateral liberalization through the Agreement on Textiles and Clothing (ATC); their gains will have a positive effect on neighboring economies. Most other developing regions, including Latin America, Africa and South Asia (except India) are expected to lose. The welfare of developed countries improves as a result of greater access to large Asian markets and lower consumer prices for clothing. Regional agreements, by contrast, have a positive effect on Mediterranean and Latin American countries, but since they have little effect on global trade creation and thus on global welfare, their main effect is trade diversion. The question that arises is whether and how the computable general equilibrium model (CGE) can correctly capture the effect of trade liberalization on trade creation? Another question concerns poverty reduction: since the countries that gain most from multilateral liberalization are large Asian and middle income countries, should regional reciprocal trade agreements be pursued?
II. STRONG REGIONAL TRENDS Trade in textiles and clothing has tended to regionalize in the last fifteen years, in contrast to the globalization of commerce evident in the 1960s and 1970s. Considering a world aggregated into three major regions – Asia, the Americas, and the European and Mediterranean countries – Asia’s share of world clothing trade increased in that period by 16.1 percentage points (see Table 1), of which 11.2 originated within Asia. The figures for textiles are 16.1 points (coincidentally the same as for clothing), of which 13.7 are from within the region. Similar trends are evident in the Americas, whose share of world clothing trade increased by 10.3 percentage points and which saw an 11.8 point growth in inter-Americas trade. For textiles the numbers are 3.6 and 5.1 respectively, which means that the Americas are losing ground in other parts of the world. Europe is more difficult to analyze, since the collapse of the socialist bloc entailed a collapse in trade. What lies behind such phenomena? This paper argues that trade policy is at the heart of the vertical of division of labor analyzed below.
NAFTA AS AN EXAMPLE OF REGIONALIZATION Figures 1 and 2 show that NAFTA played a major role in the development of regional integration. In 1999 Mexico became the leading exporter to United States: exports totaled US$ 8 billion and the country accounted for 16.2% of US clothing imports; in 1993, just before the agreement came into effect, Mexico’s market share was only 4.8%. Mexico’s progress was matched by a similar increase in US textile exports to Mexico: in 1999 some 34% of total US exports went to Mexico (twice the level of 1996) at a value of US$ 2.4 billion (despite a stagnation in absolute terms). For Caribbean countries, success is less clear since
2
I
N
T
E
G
R
A
T
I
O
N
&
T
R
A
D
E
NAFTA came into force. Consequently, the US textile industry benefits from privileged access to the Mexican market, while Mexico benefits from greater access to the US clothing market. Asia’s share of the US clothing market fell from nearly 80% in 1983 to less than 50% in 1999, despite growing competitiveness (notably of Chinese firms). The vertical division of labor between the United States and Mexico clearly had a positive impact on the development of Mexican industries and employment, but it may have been detrimental to Asia.
EUROPEAN REGIONALISM Similar and more complex trends are evident in EU trade. Preferential trade agreements were concluded in the 1960s between the EU and Mediterranean countries and, since 1989, between the EU and Central and Eastern European countries (CEECs). In the latter case the goal was to help the CEECs make the transition to a market economy. In the textile and clothing industry it took the form of quota allocations under “passive processing trade” (PPT). The quotas were progressively extended and eventually disappeared. This quantitative mechanism was designed to manage intra-European trade because some feared it could have been disruptive at the start of the process, since CEEC prices were not governed by the market. PPT quotas on clothing made from European textiles favored the EU’s neighboring countries, since European importers were encouraged to work with producers nearby who enjoyed the benefits of shorter delivery times and lower transport costs. This regional division of labor helped West European business maintain its upstream industry and improve its revealed comparative advantage in textiles. In 1988, therefore, for every US$ 100 worth of clothes imported from the Mediterranean Basin and Central Europe, the EU exported US$ 55 to these countries in textiles and yarns (Figures 3 and 4).
DIVISION OF LABOR IN ASIA Asian developing countries’ share of world exports stagnated after 1984 (see Figure 5) following a long period in which it had increased sharply. Regional agreements are still very limited in Asia. Japan has not been party to the Multifibre Agreement, and hence the decline in the Japanese textile and clothing industries has been markedly swift in the past ten years. The South Korean and Taiwanese textile industries, as well as the Chinese clothing industry, have benefited from this decline. This division of labor is in line with the traditional determinants of comparative advantage. High levels of industrial employment in Japan, and a strong surplus in the external current account, may have eased the declining industries’ structural adjustment problems. However, the decade-long banking crisis and China’s accession to the WTO are spurring trade conflicts between Japan and China, and Japan has tried to impose new quantitative restrictions on some Chinese imports.
FUTURE OF REGIONAL INTEGRATION The elimination of all discriminatory measures by 2005 under the terms of the ATC agreements will profoundly change the rules of the game for textile and clothing companies. Although some experts remain skeptical (Spinanger [1998 and 1999]), they believe that there are many ways to avert a full phasing out of discriminatory practices (such as the use of safeguard clauses). This study assumes that the quota system will totally disappear and will not be replaced by other barriers. Do multilateral agreements lessen or even eliminate the rationale for a regional agreement? A priori the answer should be in the affirmative, but in reality governments tend to make ever greater efforts to negotiate regional
I
N
T
E
G
R
A
T
I
O
N
&
T
R
A
D
E
3
agreements. Should they do so simply to give work to their civil servants and economists? The rationale for regional agreements is clear in areas such as services and foreign direct investment, where multilateral negotiations are proceeding slowly, but it is also apparent in industries wherein the remaining trade barriers are significant.
III. PREVIOUS CGE ESTIMATES Analysis of the effects of trade liberalization on the textiles and clothing sector requires a method that takes account of the interdependencies between countries and sectors. General equilibrium models are important in such an endeavor and many simulations are available. We began by comparing some of the more interesting results achieved using different models and databases (see also Whalley [1999]). The results are summarized in Table 2. The results seem to be very diverse in quantitative terms, for three main reasons: (i) differing databases for different years; (ii) diverging estimates of the level of various trade barriers; and (iii) the models have varying sensitivities to the removal of trade barriers. Qualitatively, however, the effects are quite similar. In terms of world welfare gains, the effects vary from US$ 7 to US$ 37 billion. Gains from quota elimination are generally greater than those caused by tariff reductions. In most cases the simulations reveal that the large Asian countries (especially China) are the main winners in terms of production and exports. Among developed regions, European countries and the United States are the main winners: the positive effect for consumers offsets the negative effect for producers.1 More generally the most significant results are less quantitative than qualitative.
IV. EFFECTS OF THE ATC AND CHINA’S ACCESSION TO THE WTO The simulations are based on version 5 of the GTAP dataset, which has 66 regions and uses 1997 data. We use the standard modeling framework (CGE) proposed by the GTAP model 2 to facilitate comparison between our results and previous findings. Hence the model is static and uses constant return to scale and perfect competition. Our aggregation of the GTAP database has 13 geographic zones and 7 sectors, including textiles and clothing. The geographical breakdown chosen reflects the vertical division of labor mentioned above. The Euro-centered zone covers the EU, the Mediterranean countries, the CEECs and the Commonwealth of Independent States (CIS); Asia covers Japan, China, NIC1 (the four dragons) and NIC2 (the four tigers), India and South Asia; the Americas comprise NAFTA and Latin America (the latter excluding Mexico); the remaining countries, such as sub-Saharan countries and Oceania are classified with the rest of the world (ROW). We first consider two multilateral scenarios: elimination of quotas (ATC) and China’s accession to the WTO. We separate the two issues in order to compare the magnitude of these events. Scenario 1 (SC1) is based on quota elimination excepting China as stipulated by the ATC. Scenario 2 (SC2) adds the impact of quota elimination including China.
EFFECTS OF QUOTA ELIMINATION WITHOUT CHINA (SC1) The ATC is a transitional agreement that should lead to WTO members’ complete elimination of quotas by January 1, 2005. The ATC has two main measures: first, the progressive integration of products covered by the quotas into the general regime of the GATT; and second, the liberalization of restrictions – in other words, a progressive rise in the growth of import quotas set out in the bilateral agreements concluded in the MFA framework.
4
I
N
T
E
G
R
A
T
I
O
N
&
T
R
A
D
E
In the GTAP, quotas are modeled using export tax equivalents (see Table 1). That is, the economic effect of quotas is to give leading exporters a rent and to increase the price of imported goods for consumers (Figures 6, 7, and 8).
MAIN RESULTS OF SC1 In welfare terms, India records the biggest gains (over US$ 6 billion), far ahead of second-placed NAFTA (about US$ 3 billion). India’s gains spring from export growth in clothing (a 350% increase) and a slight decline in textile exports.3 Exports from South Asia and South East Asia (NIC2) increase but only by 19% and 30% respectively in clothing, and by 17% and 10% in textiles. Quota elimination improves the welfare of developing Asia by boosting exports. World exports grow little in value terms. For clothing the increase is 7%. Lower prices translate into higher export volumes but not enough to offset negative value effects. This result is not consistent with previous findings that trade in clothing is one of the most dynamic components of world trade: on average, trade in clothing has grown by 7% a year in constant dollars during the last 30 years. Eliminating quotas would merely have the effect of one year’s growth. NAFTA’s welfare gains stem from advantages to consumers that offset what producers lose through an increase in imports and a decline in exports. Nonetheless, constant budgetary shares do not allow for growth in the value of consumption even if there is an increase in the import content and in the quantities consumed. Latin America and other developing countries lose in terms of welfare and in terms of exports. Latin American countries are the main losers, with a fall of 26% in their clothing exports and a loss of around one third in NAFTA (the second loser in intra-NAFTA trade), in both cases to the main benefit of India.
EFFECTS OF QUOTA ELIMINATION WITH CHINA (SC2) Before there was any quota elimination, and before it had any WTO status, China became the world’s leading exporter of clothing (17.5% of world exports in 1999, compared to second-ranked Italy’s share of 6.9%). China is also the leading target of antidumping measures. Several studies have examined the consequences of China’s accession to the WTO (Walmsley and Hertel [2000], Lemoine [2000a and b], and Yang [1999]). It has been pointed out (Finger and Harrison [1994]) that there is some paradox in the fact that China is the most successful country and also the main “victim” of protectionism. China’s accession to the WTO has two dimensions: the opening up of industrialized country markets to Chinese products (in this case, quota elimination) and the opening up of the Chinese market to products from other WTO members (Lemoine [2000a]). Unsurprisingly, the impact of quota elimination is strongest when applied to Chinese exports. Overall, the welfare gains are more than twice the level apparent under SC1: almost US$ 10 billion compared to less than US$ 4 billion. The main winners are NAFTA and the EU: NAFTA’s gains under SC1 increase by a third, and EU gains (which were around US$ 1 billion under SC1) become US$ 1.5 billion. More troubling is that most developing countries are losing in welfare terms. Except for India, these countries are worse off than before the elimination of quotas. India’s gains of US$ 3 billion under SC1 increase to a little more than US$ 4 billion. China doubles its exports; the growth in Indian exports falls from 350% to 250%; Latin American exports fall by 50% in clothing; and NAFTA exports decline by 30%.
I
N
T
E
G
R
A
T
I
O
N
&
T
R
A
D
E
5
A few countries are successful in textile exports: Japan (5%), China (9%), NIC1 (4%), NIC2 (8%) and South Asia (22%), but this does not offset the losses in clothing exports. Total world clothing exports increase by 13%, twice the growth under SC1. Latin America is still the main loser with a decline of 40% in the markets of North and South America. China’s exports do not increase relative to developing countries, whose protection remains constant.
V. COMPARING THE GTAP4 AND GTAP5 SIMULATIONS GTAP is an evolving and open project with regular dataset updates and improvements in calibrating procedures, as well as in modeling world trade interdependencies. The protection dataset is of particular interest, especially the measure of ATC export tax equivalents (ETE), since it has benefited from the work of Francois and Spinanger [2000]. Thus we decided to compare two different simulations of the same scenarios using the two most recent and different versions of the GTAP. For SC1, the world welfare gains are similar for the two versions of GTAP (see Figure 9) but the distributions are somewhat different. Indian welfare gains are 3.4 times higher in GTAP5 than in GTAP4. This is a direct consequence of the new estimates of the ETE (see Table 3), which relative to the EU increase from 6.1 in GTAP4 to 12 in GTAP5 for textiles, and from 9.2 to 15.2 for clothing. Relative to NAFTA the increases are small for textiles (from 8.9 to 9.6) but higher for clothing (from 25 to 33.8). These changes, however, do not fully explain the new distribution of welfare gains and losses. With higher ETEs, bigger gains might be expected for India’s partners. In fact, almost all partners lose except South Asia, which recorded a similar increase in ETE between the two versions of GTAP. For the other regions there is a decline in the evolution of ETE, and for that reason they may lose in terms of market share relative to the EU and NAFTA. The increase in ETE might spring from a real increase in the effectiveness and level of the quotas between 1995 and 1997. The increase of quotas or their economic cost for India and South Asia cannot stem from changes in the real world but probably from a change in the evaluating process. Since documents on these issues are not yet available, we can only speculate (Francois, Glismann and Spinanger [2000]). Whatever the reason, this does not explain some other changes in the results, especially the decline in welfare gains for developed countries such as the EU and NAFTA. These regions do benefit from a fall in the costs of their imports, which were higher than the costs of a decline in production. In GTAP5 the drop in production must be greater than the fall in the import cost, although this is also speculation. As regards the SC2 comparison (Figure 10), the inclusion of China significantly increases world welfare in GTAP5 relative to GTAP4: about US$ 4 billion more than the US$ 6 billion of GTAP4. The changes are dramatic for China, whose welfare grows by US$ 5 billion. ETE estimates are also several times higher in GTAP5 than in GTAP4. For developed countries such as the EU and NAFTA, the welfare effect is positive but modest. Japan is still losing. All developing countries lose, and even more so, in GTAP5 than in GTAP4. As mentioned earlier, the losses stem from the decline of export share, mostly in clothing (See Figure 8). Latin America’s losses double.
VI. TWO REGIONAL FTAS: EU-MEDITERRANEAN BASIN (SC3) AND LATIN AMERICA–NAFTA (SC4) Having simulated multilateral developments, this paper turns to regional issues. Scenario 3 (SC3) compares the effects of an EU-Mediterranean FTA, and Scenario 4 (SC4)
6
I
N
T
E
G
R
A
T
I
O
N
&
T
R
A
D
E
focuses on a Latin America-NAFTA4 FTA that is limited to the textile and clothing industries. Both cases include the results of SC2 – that is, quotas are eliminated (Figures 11 and 12). As pointed out by Bouët et al [2001] there is a broad array of agreements between the EU and its traditional partners, both between close European partners and former colonies from among the African, Pacific and Caribbean (ACP) countries. These accords, especially those with developing countries, were asymmetrical: the EU granted non-reciprocal access to developing countries. In most cases the outcome of this policy was negative, since the market share of these “preferred” countries in Europe continued to decline in the 1970s and 1980s. There were a few exceptions in North Africa. Following the Marrakesh agreement, all bilateral accords should be based on reciprocity. Before the 1990s many developing countries maintained very high levels of protection and were marked by low transparency (no bound rates) and a substantial degree of corruption, making it very expensive to acquire foreign inputs or equipment. Preferential, non-reciprocal agreements should be forbidden in the coming years, a circumstance that has raised fears of further losses for poor countries, especially in the southern Mediterranean (Fontagné and Péridy [1997], Bensidoun and Chevallier [1996], Chaponnière [2000]) (Figure 13). The relative success of the CEECs and the Mediterranean countries in the 1990s shows that freeing trade with developed European countries helped the formers’ development. Hence further regional liberalization might improve conditions in a system of global liberalization. In principle there are no tariffs between the EU and most Mediterranean countries, but estimates using GTAP5 reveal a remaining tariff equivalent of 10.5% for textiles and 12.6% for clothing. The situation is different in the Americas. Until 1985 the United States was reluctant to sign a regional agreement and favored multinational negotiations. Latin American countries, by contrast, engaged in a number of regional projects that had few significant outcomes until the 1990s. The main reason for this is the difficulty involved in reducing trade barriers with close partners while retaining high barriers towards third countries. Liberalization has to proceed in parallel on the regional and multilateral levels. Once liberalization has begun, the process of regional negotiations gains credibility. Two years after the signing of NAFTA, which marked a significant change in US trade policy, the countries of the Americas began negotiating a Free Trade Area of the Americas (FTAA) that is scheduled to start in 2005 and to be fully operational within ten years. We compared the effects of the elimination of trade barriers between the EU and the Mediterranean countries using SC2 as a reference (quota elimination as stipulated in the ATC) with SC3 (see Figure 9) and between NAFTA and Latin American countries with SC4 (see Figures 9 and 10). These scenarios suggested strong positive effects on the Mediterranean countries (more than US$ 3.5 billion in welfare gains under SC3) and less impact on Latin American countries (losses of US$ 1.3 billion). The differences between the two regions are that levels of protection are much higher in the Mediterranean than in Latin America, and that the specialization of Mediterranean countries is much more intense in clothing. In both cases, world welfare increases by about US$ 5 billion. Mediterranean countries increase their clothing exports by more than US$ 11 billion (120%) and by about US$ 2.4 billion in textiles, while the EU increases both its textiles exports and its clothing imports. The division of labor between the two areas increases. Similar conditions prevail in Latin America.
I
N
T
E
G
R
A
T
I
O
N
&
T
R
A
D
E
7
VII. CONCLUSION Among developing countries, only large Asian nations are expected to gain from multilateral liberalization (the ATC). Their gains will have a positive effect on neighboring economies while most other developing areas, including Latin America, Africa and South Asia (except India) are expected to lose. The gains for developed countries stemming from greater access to large Asian markets are also modest. By contrast, regional agreements have a positive effect on Mediterranean and Latin American countries but, since they have minor effects on global trade creation and therefore on global welfare, the main effect is trade diversion. The question is whether and how the CGE can correctly capture the trade liberalization effect on trade creation. Openness to trade in textiles and clothing has had and will have positive effects on a few large developing countries but it has somewhat negative effects on small and very poor countries. In that context there are grounds for arguing that some kind of preferential treatment based on reciprocity should be maintained.
8
I
N
T
E
G
R
A
T
I
O
N
&
T
R
A
D
E
Notes
1 More limited estimates for the United States alone can be found in Cline [1987], Hufbauer, Berliner and Elliot [1986] and USITC [1999]. 2
For more information, see http://www.gtap.agecon.purdue.edu.
3
For an analysis of the Indian case, see Kathuria and Bhardwaj [1998] and Kathuria, et al [1999]. 4
Latin America includes the Caribbean countries and excludes Mexico, which is a member of NAFTA.
I
N
T
E
G
R
A
T
I
O
N
&
T
R
A
D
E
9
Table 1
LONG TERM VARIATIONS IN WORLD TRADE SHARE AMERICAS
EURAFRICA
ASIA
WORLD
Americas
CLOTHING 2000-1970
11,8
-0,8
-0,1
10,3
Eurafrica
-2,4
-24,5
0,3
-26,7
Asia
1,4
3,6
11,2
16,1
World
11,0
-21,5
11,5
0,0
TEXTILES 2000-1970 Americas
5,1
-1,1
-0,1
3,6
Eurafrica
-3,7
-16,8
-0,7
-21,5
Asia
-0,5
3,0
13,7
16,1
World
0,9
-14,3
14,0
0,0
Source: CHELEM-CEPII [2001].
Table 2
COMPARING THE MAIN RESULTS OF CGEMS USED TO SIMULATE QUOTA ELIMINATION AUTHORS DATA
MAIN MODELING HYPOTHESIS
Trela and Whalley [1990] Data 1986 7 textile and clothing quotas 7 textile and clothing tariffs
10
WELFARE GAINS (US$ BILLIONS)
US$ 23 billion world gain. No consumer discrimination between US$ 8 billion gains for developing imported goods and local goods. countries; most of these are winners.
Harrison; Rutherford and Tarr [1995;1997]
Static model, constant return.
US$ 14 billion, of which –2 for developing countries.
Data 1992 (GTAP 2)
Pseudo-dynamic model.
Hertel; Martin; Yanagishima and Dimaranan [1995] Data 1992 (GTAP 2)
World economy estimated year 2005 before liberalization.
US$ 20 billion gains, of which +4 for developing countries. China the only winner.
US$ 37 billion gains from quota, US$ 13 billion for tariff reduction. Large Asian countries are the main winners.
US$ 28 billion gains for the world. US$ 11 billion for developing countries. Large Asian countries are the main winners.
Yang [1997] Data 1992 (GTAP 2)
Static model with perfect competition.
Yang [1994] Data 1986
Static model with perfect competition. Quota abolition only.
US$ 7 billion gains for the world. US$ 3.7 billion for developing countries. NICs are the main winners.
Francois, Glissman and Spinanger [2000] Data 1995 (GTAP4)
Base year projected for 1997.
25 billion Euros gains for EU consumers.
I
N
T
E
G
R
A
T
I
O
N
&
T
R
A
D
E
Table 3
ETE (EXPORT TAX EQUIVALENT) GTAP 4 VS GTAP 5
TEXTILES EU
CLOTHING EU
GTAP4
GTAP5
GTAP4
GTAP5
2 Med. Basin
-1.1
-1.7
2 Med. Basin
-1.1
-1.2
3 East Europ.
0
0
3 East Europ.
0
0
-0.2
-0.2
4 CIS
-0.1
-0.1
5 Japan
4 C IS
0
-3.1
5 Japan
0
-3.1
6 China
-1.1
-12
6 China
-4.7
-15
7 NIC1
-1.5
-0.9
7 NIC1
-4.2
-3.9
8 NIC2
-2.6
-6
8 NIC2
-6
-6.1
-9.2
-15.2
-2
-7.1 -0.3
9 India
-6.1
-12
9 India
10 South Asia
-2.7
-8.1
10 South Asia
11 NAFTA
-0.5
0
11 NAFTA
-0.3
12 Lat. Am.
-2.4
4.3
12 Lat. Am.
-0.1
-2
13 ROW
3.3
2.9
13 ROW
-1.3
0.2
TEXTILES NAFTA GTAP4
CLOTHING NAFTA GTAP4
GTAP5
1 EU
GTAP5 1 EU
-0.9
0
2 Med. Basin
2 Med. Basin
-0.7
-4
3 East Europ.
3 East Europ.
-0.1
-5
4 CIS
4 CIS
-0.1
-3.1
5 Japan
5 Japan
0
-3.1
6 China
6 China
-18.8
-32.7
7 NIC1
7 NIC1
-9
-6.6
8 NIC2
8 NIC2
-13
-8.8
9 India
9 India
-25
-33.8
10 South Asia
10 South Asia
-5
-8.2
12 Lat. Am.
12 Lat. Am.
-0.5
-0.4
13 ROW
13 ROW
1.7
1.6
Source: Global Trade Analysis Project (GTAP).
I
N
T
E
G
R
A
T
I
O
N
&
T
R
A
D
E
11
Figure 1 AMERICA AND ASIA (EXCLUDING JAPAN) SHARE
OF THE
US IMPORT MARKET FOR CLOTHING
80.0
Asia (excluding Japan) 70.0
60.0
50.0
40.0
30.0
Rest of Latin America plus Mexico 20.0
Rest of Latin America
Mexico
10.0
1984
1992
0.0
Source: CEPII, Database CHELEM, CD-ROM 2003.
12
I
N
T
E
G
R
A
T
I
O
N
&
T
R
A
D
E
Figure 2 STRUCTURE OF AMERICAN
TEXTILE
EXPORTS
60.0
50.0
Latin America 40.0
30.0
20.0
Asia 10.0
Europe Periphery
0.0
Source: CEPII, Database CHELEM, CD-ROM 2003.
I
N
T
E
G
R
A
T
I
O
N
&
T
R
A
D
E
13
Figure 3 SHARE
OF
NEIGHBOURING COUNTRIES AND ASIA IN
THE
EUROPEAN CLOTHING IMPORTS
30.0
25.0
Asia
20.0
EU neighbouring countries
15.0
Southern Mediterranean countries 10.0
Central Europe
5.0
0.0
Source: CEPII, Database CHELEM, CD-ROM 2003.
14
I
N
T
E
G
R
A
T
I
O
N
&
T
R
A
D
E
Figure 4 STRUCTURE OF EUROPEAN TEXTILE EXPORTS
25.0
20.0
EU Periphery
15.0
Central Europe 10.0
Mediterranean Basin 5.0
Asia
Latin America 0.0
Source: CEPII, Database CHELEM, CD-ROM 2003.
I
N
T
E
G
R
A
T
I
O
N
&
T
R
A
D
E
15
Figure 5 DEVELOPING COUNTRIES
SHARE OF
WORLD EXPORTS
IN
CLOTHING
80
70
60
50 Rest of Developing Economies
40 Rest of Developing Asia
30
China 20
NIE 1 10
0
Source: CEPII, Database CHELEM, CD-ROM 2003.
16
I
N
T
E
G
R
A
T
I
O
N
&
T
R
A
D
E
Source: Author's calculations using GTAP model.
I
N
T
E
G
R
A
T
I
O
N
&
T
R
A
D
E
17
-2,000.0
0.0
2,000.0
4,000.0
6,000.0
8,000.0
10,000.0
Figure 6
SC1
SC2
-50.0
0.0
50.0
100.0
150.0
200.0
250.0
300.0
Clothing
Textile
(%)
(US$ millions) 350.0
EXPORT CHANGES SC1
WELFARE GAINS SC1 & SC2
Figure 7
Source: Author's calculations using GTAP model.
18
I
N
T
E
G
R
A
T
I
O
N
&
T
R
A
D
E
-50
0
50
100
150
-1,000.0
0
1,000.0
2,000.0
3,000.0
4,000.0
5,000.0
250
200
6,000.0
(%)
EXPORT CHANGES SC2
Figure 9
300
Figure 8
GTAP4
GTAP5
WELFARE GAINS SCENARIO 1 COMPARING GTAP4 AND GTAP5
Source: Author's calculations using GTAP model.
I
N
T
E
G
R
A
T
I
O
N
&
T
R
A
D
E
19
-4,000.0
-2,000.0
2,000.0
4,000.0
6,000.0
8,000.0
10,000.0
12,000.0
0
GTAP4
GTAP5
COMPARISON OF SIMULATIONS FROM SCENARIO 2 USING GTAP4 AND GTAP5
Figure 10
-4,000.0
-2,000.0
0.0
2,000.0
4,000.0
6,000.0
8,000.0
10,000.0
12,000.0
SC3
SC4
WELFARE GAINS COMPARING TWO REGIONAL FTA EU-MEDITERRANEAN COUNTRIES (SC3) AND FTA (SC4)
Figure 11
Source: Author's calculations using GTAP model.
20
I
N
T
E
G
R
A
T
I
O
N
&
T
R
A
D
E
-100
-50
0
50
100
150
200
250
300
Figure 12
SC3
SC4
(%)
EXPORT CHANGES IN CLOTHING EXPORTS
+16,5%
-20
-10
0
10
20
30
40
50
Figure 13 (%)
EXPORT CHANGES IN TEXTILE EXPORTS
Bibliography BENSIDOUN, ISABELLE AND AGNÈS CHEVALLIER. “Europe-Méditerranée: le pari de l’ouverture”, Economica, CEPII collection, 1996. BOUËT, ANTOINE. “La mesure des protections commerciales nationales”, CEPII Working Paper, 15. November, 2000. BOÜET, ANTOINE AND E STELLE DHONT-PELTRAULT. “How is Trade Protectionism to be Measured?”, la Lettre du CEPII. November, 2000. BOUËT, ANTOINE ET AL. “Market Access Maps: A Bilateral and Disaggregated Measure of Market Access”, CEPII Working Paper 18. 2001. CENTRE D'ETUDES PROSPECTIVES ET D'INFORMATIONS INTERNATIONALES. CEPII - Database CHELEM, CD-ROM, 2001 and 2003.
CLINE, WILLIAM R. “The Future of World Trade in Textiles and Apparel”. Washington, D.C.: Institute For International Economics. 1987. CHAPONNIÈRE, JEAN-RAPHAËL. “Textile-habillement: les défis pour les pays méditerranéens et d’Europe Centrale”, Revue Elargissement, DREE. Paris: Ministère de l’économie, des finances et de l’industrie. 2000. FINGER, MICHAEL J. AND ANN HARRISON. “The MFA Paradox: More Protection and More Trade?”, NBER, Working Paper 4751, May 1994. FINGER, MICHAEL J. AND LUDGER SCHUKNECHT. “Market Access Advances and Retreats: The Uruguay Round and Beyond”. World Bank and WTO, 1999. FONTAGNÉ, LIONEL AND NICOLAS PÉRIDY. “The EU and the Maghreb”. Paris: OECD Development Center. 1997. FRANCOIS, JOSEPH; HANS GLISMANN AND DEAN SPINANGER. “The Cost of EU Trade Protection in Textiles and Clothing”. Kiel Institute of World Economics, August 2000. FRANCOIS, JOSEPH AND DEAN SPINANGER. "Hong Kong's Textile and Clothing Industry: the Impact of Quotas, the UR and China's WTO Accession. Kiel Institute of World Economics. 2000. GAULIER, G UILLAUME. “Discrimination commerciale: une mesure à partir des flux bilatéraux”, CEPII Working Paper, 4-2001.
I
N
T
E
G
R
A
T
I
O
N
&
T
R
A
D
E
21
HARRISON, GLENN; THOMAS RUTHERFORD AND DAVID TARR. “Quantifying the Uruguay Round”, in W. Marin and A. Winters (eds.) The Uruguay Round and Developing Countries. World Bank Discussion Papers 307, October 1995. HERTEL, THOMAS W.; WILL MARTIN; KOJI YANAGISHIMA AND BETINA DIMARANAN. “Liberalising Manufactures Trade in a Changing World Economy”, in W. Marin and A. Winters (eds.) The Uruguay Round and Developing Countries. World Bank Discussion Papers 307, October 1995. HUFBAUER, GARY C.; DIANE E. BERLINER AND KIMBERLY A. ELLIOT. Trade Protection in the United States: 31 Case Studies. Washington, D.C.: Institute for International Economics. 1986. HUFBAUER, GARY C. AND KIMBERLY A. ELLIOT. “Measuring the Costs of Protection in the United States”. Washington, D. C.: Institute For International Economics, January 1994. KATHURIA, SANJAY AND A. BHARDWAJ. “Export Quotas and Policy Constraints in the Indian Textile and Garment Industries”. World Bank, November 1998. KATHURIA, SANJAY ET AL. “Implications of MFA Abolition for South Asian Countries”. World Bank, December 1999. LEMOINE, FRANÇOISE. “Pourquoi la Chine veut-elle adhérer à l’OMC?”, La lettre du CEPII, 189, April 2000a. (Also available in English at CEPII web site: http://www.cepii.fr). _____________. “FDI and the Opening Up of China's Economy”, CEPII Working Paper 11-2000b. MERCER MANAGEMENT CONSULTING, SNC. “Impact de l’Uruguay Round sur le secteur européen du textile et de l’habillement." Report prepared for the Commission of the European Communities, Directorate General for Industry. August 1995. NATHAN ASSOCIATED, INC. “Analysis of the Effects of Chinese Quota Phase-out in 2005 on Production and Employment in the US Textile and Apparel Industries”, presented to the American Textile Manufacturers Institute. September 1999. NIMAN, WESLEY AND JOHN BEGHIN. “Eco-Labelling and International Trade in Textiles”, Center for Agriculture and Rural Development, June 1999. L’OBSERVATOIRE EUROPÉEN DU TEXTILE ET DE L’HABILLEMENT. “The EU Textile and Clothing Sector 2000, a Factual Report”. Brussels, May 2000. SPINANGER, DEAN. “Textiles Beyond the MFA Phase-Out.” Center for the Study of Globalisation and Regionalism, July 1998.
22
I
N
T
E
G
R
A
T
I
O
N
&
T
R
A
D
E
SPINANGER, DEAN. “Faking Liberalization and Finagling Protectionism: The ATV At Its Best”, World Bank, June 1999. THONGPAKDE, NATTAPONG AND WISARN PUPPHAVESA. “Returning Textiles and Clothing to GATT Disciplines”, World Bank, July 1999. Data Tables 3.1/4.1 / 3.4/3.5 / 3.2/4.3/4.4. TRELA, IRENE AND JOHN WHALLEY. “Global Effects of Developed Country Trade Restrictions on Textiles and Apparel”, Economic Journal 100, pp. 1190-1205. 1990. ___________. “Internal Quota Allocation Schemes and the Costs of the MFA”, CSIER. Mimeo, 1993. US INTERNATIONAL TRADE COMMISSION - USITC. “The Economic Effects of Significant US Import Restraints”, Second Update 1999. WALMSLEY, TERRIE AND THOMAS HERTEL. “China’s Accession to the WTO: Timing is Everything”. Center for Global Trade Analysis, Purdue University, September 2000. WHALLEY, JOHN. “Notes on Textiles and Apparel in the Next Trade Round”. Paper presented at the Center for International Development at Harvard University, November 1999. YANG, YONGZHEN. “The Impact of the MFA Phase Out on World Clothing and Textile Markets”, Journal of Development Studies 30. July 1994. ___________. “China’s Textile and Clothing Exports: Changing International Comparative Advantage and its Policy Implications”. Australia Pacific Press, Australia National University, CEP99-3, 1999. YANG, YONGZHEN; WILL MARTIN AND KOJI YANAGISHIMA. “Evaluating the Benefits of Abolishing the MFA in the Uruguay Round Package” in Thomas Hertel. (ed) Global Trade Analysis: Modelling and Applications. Cambridge University Press, 1997.
I
N
T
E
G
R
A
T
I
O
N
&
T
R
A
D
E
23
24
I
N
T
E
G
R
A
T
I
O
N
&
T
R
A
D
E
