A Global Perspective of Liberalizing World Textile and ... · 03 NOPEC 28 (2) Diao NY 11.12.02...

This article can be dowloaded from: http://www.nopecjournal.org/NOPEC_2002_a09.pdf Other articles from the Nordic Journal of Political Economy can be found at: http://www.nopecjournal.org

Nordic Journal of Political Economy

Volume 28 2002 Pages 127-145

A Global Perspective of Liberalizing World Textile

and Apparel Trade

Xinshen Diao Agapi Somwaru

Trends in world T&A trade flowIn the last four decades, world textile andapparel (T&A) trade increased from less than$6 billion in 1962 to 300 billion (in nominalterms) in 1999. Deflated by the U.S. GDPdeflator, world T&A trade increased by 11times in this period. Of the $300 billion ofT&A trade, slightly less than two-thirds istrade in apparel goods and the rest is trade intextile goods. Four decades ago, however, thevalue of world textile trade was twice that ofthe world apparel trade (figure 1). As worldtextile trade has increased by five times andtrade in apparel has grown more than 25

times, apparel trade has taken the lion’s shareof total world T&A trade.

In general, developing countries have acomparative advantage in textile and appareltrade. This advantage allows developingcountries to diversify their exports beyondtraditional primary commodities, whoseproduction may be restrained by naturalresources. As a leading, labor-intensive man-ufacturing sector, the textile and apparelindustry is often thought to represent the firstbase in a country’s economic growth anddevelopment. Moreover, unlike the primaryagricultural commodities that are often

Xinshen Diao and Agapi Somwaru*

A Global Perspective of Liberalizing World Textile

and Apparel Trade

* Research fellow, International Food Policy Research Institute, and Senior Economist, Economic ResearchService, USDA, respectively. We are grateful to the anonymous reviewers of the Journal and the participants ofthe conference "Globalization and Marginalization", June 9–11, 2001, Bergen, Norway, for the helpful com-ments and critical suggestions. Correspondence: Xinshen Diao, 2033 K Street, NW, Washington DC 20006.(O) 202-862-8113. Email: [email protected]

International trade in textile and apparel has been governed by quantitative restrictionsunder the Multi-Fiber Arrangement (MFA) and earlier agreements for more than 30 years.One of the major accomplishments of the Uruguay Round was the Agreement on Textilesand Clothing (ATC), which provides for the dismantling of these restrictions. Under theUruguay Round ATC, the MFA restrictions are to be phased out over a 10-year period andare scheduled to end by the year 2005. This study combines a data description of the trendsin world textile and apparel trade flow, an econometric analysis on the linkage betweentextile trade and growth, and an intertemporal, world CGE model to evaluate the possibleimpact of liberalizing world textile and apparel trade. As textile and apparel industry is animportant source of the growth, our study focuses on the effect on the developing countries.Key words: Textile trade and growth, ATC, Intertemporal general equilibrium

03 NOPEC 28 (2) Diao NY 11.12.02 12:07 Side 127

income inelastic, demand for textile andapparel commodities steadily grows in bothdeveloped and developing countries as coun-tries become wealthier. This implies that formany developing countries there is room forfuture expansion of their production andexport capacities.

In terms of the contribution to a country’seconomic development, many countries’experience shows that once export growthbegins in the textile and apparel sectors, othersteps in economic development follow. Thistransition has taken place in Korea, Taiwan,and now is happening in China, India, andmany other South and Southeast Asian, andLatin American countries. One reason is thatthere are strong linkages between the textileindustry and other economic sectors, bothagricultural and non-agricultural. Growth inthe textile sector benefits “upstream” agricul-tural or manufacturing sectors throughincreased demand for material inputs ormachinery and equipment. In addition, thetextile and apparel sectors depend on the pres-ence of many modern economic activities.Through developing export-oriented textileand apparel industries, a country acquiresother knowledge and skills such as marketing,advertising, transportation, and communica-tion. These advances highlight the impor-

tance of the textile and apparel industries to acountry’s development process.

International trade in textiles and apparelhas been governed by quantitative restrictionsunder the MFA and earlier agreements formore than 30 years. Restrictions in T&A tradecan be traced back even further to the late1950s when Japan imposed “voluntary exportrestraints” on cotton textiles destined for theUnited States (Spinanger, 1998). Startingwith cotton and selected exporters, the MFAeventually covered textiles of all fibers and reg-ulated exports from virtually all developingcountries. It is estimated that the MFA coversabout 15 percent of world textile trade and 40percent of world apparel trade (Cline, 1990).Quantitative restrictions on developing coun-try exports to developed countries wereimposed through bilateral arrangements sanc-tioned under the MFA, while developedcountries permitted unrestricted trade amongthemselves (see Trela and Whalley, 1990, fordetails on the development of the MFA).

The world trade regime that came togovern textiles was not entirely satisfying toanyone. Chafing under restrictions on exportgrowth, periodic closing of quotas, and inves-tigations of alleged quota circumvention,developing exporting countries railed againstinterference in one of their most promisingsources of export growth. In developed coun-tries, retailers resented the constraints onapparel sourcing, while domestic producers oftextiles and apparel faced rapidly risingimports despite the presumably restrictivesystem of quotas and above average tariffs.Quota-driven constraints on established, effi-cient exporting countries drove apparelproduction into lower-income developingcountries open to investment as the industrysought to stay ahead of importers’ regulatoryefforts. For their part, developing countriestypically imposed import barriers of theirown, barriers that were often even more

128 Xinshen Diao and Agapi Somwaru

0

50

100

150

200

250

300

350

62 66 70 74 78 82 86 90 94 98

ApparelTextile

All

Figure 1.World Textile and Apparel Exports(In Billions US Dollars)


restrictive than those imposed by developedcountries under the MFA, providing yetanother set of stumbling blocks to the globalexpression of comparative advantage. Underthese restrictions, industrial countries as agroup (ICs) accounted for the largest share ofworld total T&A exports until 1990, and wereovertaken by developing countries (LDCs)only in the post decade (figure 2).

When we consider textile and apparel astwo separate categories, the ICs exportedmore textile goods than that of LDCs until thelast decade, while exports of apparel goodsfrom LDCs exceeded the exports of the indus-trial country group in late 1970s. In the lastfive years, almost 70 percent of world apparelgoods were exported from the developingcountries (figure 3). The developing countrieshave gained market share in world appareltrade mostly due to the relative more rapidgrowth rate of their exports compared to ICsexports. Industrial countries’ apparel exportsactually grew quite rapidly and even fasterthan their textile exports in the last fourdecades (seven vs. three times). But develop-ing countries’ apparel exports increased bymore than 88 times, and many developingcountries have become major exporters inworld apparel markets.

While ICs are the destination of more than70 percent of world T&A trade (figure 4), themarket is not fully open to the developingcountries. Even though the developing coun-tries have gained considerable ground andincreased their market share in the last twodecades, the industrial countries are still trad-ing with other industrial countries to a largeextent. This is in particularly true for theEuropean Union, the world’s largest textileand apparel importer. While the share of theEU’s T&A imports is about 40–50 percent ofworld T&A trade, intra-EU’s trade accountsfor more than 50 percent of total EU’s T&Aimports, even in recent years. That is, roughly

A global perspective of liberalizing world textile and apparel trade 129

10

20

30

40

50

60

70

80

90

62 66 70 74 78 82 86 90 94

ICs

LDCs

10

20

30

40

50

60

70

62 66 70 74 78 82 86 90 94

Textile

Apparel

0

20

40

60

80

62 66 70 74 78 82 86 90 94

All ICsEUUS

Figure 2.Share in World Textile and Apparel Exports

Figure 3. Share in World Textile and ApparelExports for Developing Countries

Figure 4. Shares in World Textile and Apparel Imports


20 percent of world T&A exports are actuallytraded among the EU’s member countries. If we further take into account trade betweenthe U.S. and EU, and between the U.S. andCanada, intra-industrial country tradeaccounts for 50 percent of T&A market sharein the industrial countries (figure 5).

Economic growth and T&A tradeThe purpose of this study is to evaluate the pos-sible gains to developing countries if the MFAis dismantled. Before we proceed with the analy-sis, we first empirically investigate the impact oftrade in textile and apparel products on stan-dards of living. Many studies have empiricallyexamined the correlation between trade andincome. However, most of these studies look attrade in general, i.e., the relationship betweentotal trade (exports plus imports for all com-modities) and income level. While these stud-ies provide some insights, we want to specifi-cally investigate the linkage between trade intextiles and apparel and income.

We employ the UN COMTRAD databaseto develop a data set of textile and apparel tradefor 91 countries over 37 years (1962–98).Using GDP and population data for thesecountries, we conduct a time-series and cross-sectional estimation to analyze the relationshipbetween textile and apparel trade and income

growth. In order to compare our results withthose of other studies, we also include in theestimation total trade, agricultural trade, andtotal non-agricultural trade of the 91 countriesover a 37-year period. The estimated equationis adopted from Frankel and Romer (1999).We do not include the country geographicarea in the equation, as this type of effect isalready captured by the constant term in theestimated panel model. Moreover, due to dataconstraints, we are not able to develop the con-structed trade shares that Frankel and Romerhave estimated in their study, and hence we usethe actual trade shares.

The scale effects may distort the magni-tude of estimated coefficients, because, e.g.,shares for total trade are much larger than theshares for the textile and apparel trade. Toavoid the scale effects, we normalize the databefore we conduct the estimation. Specifi-cally, we use U.S. 1962’s data (including U.S.per capita GDP and all different trade sharesfor this year) as a reference, and divide all 91countries over 37 years data by the reference.

The endogeneity is another problem thatwe have to deal with, as the independentvariables are the trade shares which are theratios of imports plus exports over real GDP,while the depend variable is the GDP percapita. To avoid this problem, we use a two-step procedure in the estimation. In the firststep, each country’s trade shares are specifiedas functions of time (the instrument) and con-sequently the estimated trade shares (T̂it ) overtime are derived. We then estimate thefollowing equation using a time-series andcross-section (TSCSREG) procedure:

1nYit = α + βT̂ti + γ 1n Ait + εit

where Yit is the real income (GDP) per personof country i = 1, 2, …, 91, and time t = 1962,…, 98, T̂it (estimated) trade share, Ait popula-tion for country at time t, and εit the error term.


20

40

60

80

62 66 70 74 78 82 86 90 94

TextileApparelAll

Figure 5. Share of Industrial Countries' ImportsComing from Other Industrial Countries


The TSCSREG procedure is included inthe SAS software and it addresses both error-correction and fixed effect problems (SASInstitute, 1979). These methods are consistentwith the Generalized Method of Moments(GMM) (see Arellano and Bond, 1991).

The estimation results indicate statisticallysignificant positive relationships betweentotal trade and income, total nonagriculturaltrade and income, as well as textile and appareltrade and income. However, the estimatedparameter for the share of agricultural trade isnegative. The estimated parameter for theshare of total trade is 1.209 (table 1, column1) and this result is consistent with the esti-mation in Frankel and Romer (1999, table 3).The parameter estimate implies that anincrease in the share of total trade in GDP byone percentage point is associated with anincrease of 1.2 percent in income per person.When we distinguish total trade into twoaggregate categories: agricultural and non-agricultural trade, the estimated coefficientsvary considerably and the coefficient for agri-cultural trade share is negative (–1.54) andpositive for the non-agricultural trade (1.967,table 1, columns 2 and 3). As we mentioned

above, world agricultural trade grew muchmore slowly than world non-agriculturaltrade in the last four decades (15 vs. 55 times).This causes the share of agricultural trade inGDP to decline in the world (figure 6). Anegative coefficient on the share of agricul-tural trade is consistent with this fact, whichimplies that, in general, a decline in the shareof agricultural trade in GDP is associated withan increase in income. This result also impliesthat to increasing non-agricultural trade ismore crucial for developing countries to raisetheir living standards than depending mainlyon agricultural trade.

The positive coefficients are obtained forthe shares of textile and apparel trade. More-over, the coefficient for the share of appareltrade is significantly larger than that for theshares of textile trade and total nonagri-cultural trade. The results indicate that onepercentage increase in apparel trade shares isassociated with a 3.3 percent increase inincome per person (table 1, columns 6),which implies the importance of apparel tradein economic growth. Also, the data show thatgrowth in T&A trade, especially in appareltrade, is much more rapid than the growth in


Table 1. Trade and income

Parameter estimate (standard error in parentheses)

Variable/Statistic Total Agricultural Nonagricultural Textile Apparel Textile &trade trade trade trade trade Apparel trade(1) (2) (3) (4) (5) (6)

Intercept -4.923 -0751 -6.197 -4.909 -4.243 -7.260(0.164) (0.040) (0.264) (0.229) (0.185) (0.156)

Trade share 1.209 -1.540 1.967 0.702 3.281 1.272(0.016) (0.028) (0.085) (0.030) (0.065) (0.020)

Ln population -0.155 -0.439 1.900 -0.904 -1.141 -0.443(0.009) (0.017) (0.108) (0.047) (0.008) (0.010)

R-squared 0.72 0.72 0.85 0.71 0.87 0.84

Results are all statistically significant at the 1-percent level


GDP. World GDP per capita rose by morethan 2 times in the 37 years while T&A tradehas grown more than 50 times, mainly due thegrowth in apparel trade. Thus, the share ofT&A trade in world GDP rises to 1.5 percentin 1998, from 0.35 percent in 1962. (figure 7)

In sum, the regression results suggest thatT&A trade has a quantitatively large, robust,positive, and statistically significant effect onincome. Although we do not specificallyincorporate policy effects, the results suggestthat promoting trade will benefit economicgrowth. In the following sections, we analyzequantitatively how the MFA and other traderestriction policies affect the economy. Theregression results are used for the simulationanalysis later.

MFA phase-out processThe Uruguay Round’s ATC mandates the endof the quotas established under the MFA andalso the reciprocal termination of the restric-tions imposed by developing countries ontheir imports of textiles and clothing. By2005, restrictions that do not meet GATTstandards are supposed to be phased out, andthe strengthened dispute-settlement mecha-nism the Uruguay Round introduced to theWorld Trade Organization increases the likeli-

hood that the agreed liberalization will in factoccur.

The MFA phase-out is comprised of twoparts: a four-stage process eliminating exportrestraints contained in bilateral agreementspreviously negotiated on products coveredunder the MFA, and an increase in quotagrowth rates for products still under restric-tion during the transition period. The ATCalso deals with other non-MFA restraintmeasures relating to textiles and clothing.

The integration of the textile and clothingsector into the GATT is expressed as a per-centage of the total volume of imports in 1990of “covered” products. The four stages aredefined in the ATC as follows:

Stage 1 – On January 1, 1995, members shallintegrate products that account for at least 16percent of their total 1990 import volume;

Stage 2 – On January 1, 1998, they shallintegrate products that account for at least anadditional 17 percent of the total 1990 importvolume;

Stage 3 – On January 1, 2002, they shallintegrate products that account for at least anadditional 18 percent of the total 1990 importvolume;


0

5

10

15

20

25

62 66 70 74 78 82 86 90 94 98

Agr.

Nonagr.

0,0

0,2

0,4

0,6

0,8

1,0

1,2

1,4

1,6

1,8

62 66 70 74 78 82 86 90 94 98

TextileApparelAll

Figure 6. Share of Agricultural and NonagriculturalTrade in GDP in the World

Figure 7. Share of Textile and Apparel Trade inGDP in the World


Stage 4 – On January 1, 2005, all remainingATC restrictions are eliminated and the tex-tile and clothing sector is integrated into theGATT.

In addition to these minimum percentages,products from each of the four groups – topsand yarns, fabrics, made-up textiles, andclothing – must be included in each stage.However, the selection of products to inte-grate is determined by the importing country.Also, products not liberalized but under quotaor other restraint will have their quota growthrates increase during the first three stages of the phase-out period by 16, 25, and 27percent, respectively.

As successful as the countries were inachieving an agreement on T&A products,they did not succeed in bringing T&A prod-ucts under the jurisdiction of the GATTframework throughout the phase-out period.However, the selection of products to integrateis determined by the importing country, andthe products integrated into the GATT are notnecessarily those whose imports are restricted.Importing countries have largely chosen topostpone integration of the most heavily pro-tected products until 2005. United Nationestimates in Thomas and Whalley (1998) sug-gest that the major restraining countries ful-filled their obligations under the ATC in thefirst two stages without significantly eliminat-ing any MFA quota in place. As a result,modest trade opportunities for developingcountries may be available only after the thirdstage is in place in 2002. Even then, about halfof the 1990 import volume will remainrestricted until 2005 as specified in the ATC.Moreover, the ATC specifies nothing aboutwhat might or might not happen beyondthe year 2005 when the MFA is absent. The

language of the agreement speaks of textilesand apparel as a sector simply returning to nor-mal GATT/WTO disciplines. However, this

need not, and probably will not, mean that freetrade will prevail in T&A trade. There are fearsof new developed country dumping actions in this sector partially substituting for exist-ing restraints; there are also doubts openlyexpressed by developing countries as towhether the developed countries will have thepolitical will to actually implement their com-mitment. These doubts focus more heavily onthe North American than the Europeanimporters, but both are the subject of ques-tions (Whalley, 1999).

Perspectives of a post MFA worldGiven the fact that much of the trade liberal-ization due to the MFA phase-out will happenafter 2004, this study does not take intoaccount commitments that will be imple-mented in the first three stages. That is, we donot attempt to analyze specifically what willhappen in each stage in which the phase-outcommitments are implemented. Instead, wefocus on the potential outcome in a post MFAworld.

Several facts make the study difficult anddistinguish it from a standard trade liberaliza-tion case. First, regional trade in textile andapparel goods has grown rapidly in the twoworld’s largest industrial (importing) regions– the North America and the EU. In the last5 – 8 years, the U.S. has increased apparelimports from Mexico under NAFTA andfrom Jamaica and Dominican Republic underthe Caribbean Basin Initiative, while the EUhas increased imports from the Central Euro-pean countries. Most of this trade is free ofquota restraints and hence is growing rapidly.This regional trade growth has built a con-stituency among a subset of exporting coun-tries who now may find themselves morefavorable to the MFA, since these countriesnow have export markets partly protectedthrough MFA quotas from other suppliers.



Second, Korea, Taiwan, and Hong Kong,which were the largest apparel exportersamong developing countries 15–20 years ago,are now left with unused and unfilled MFAquotas for their apparel exports, while coun-tries like China, India, and Pakistan are facingtight quota restraints due to the fast growth intheir apparel exports. As the current quotasmay be not binding for some countries, itseems improper to use established quota levelsto quantitatively measure the restrictions thatthe exporting countries face due to the MFA.

Third, unlike most other non-tariffbarriers which allow importing countries tocapture the rent generated from restrictedimports, the MFA allows exporters to capturesome of these rents (Spinanger, 1998). Underthe MFA, importing countries (industrialcountries) “sell” to the exporting countries theright to continue to export given amounts ofT&A products. The exporting countries havethe opportunity to capture the rents ensuingfrom restricting supplies2. Thus, the domesticprices of imported T&A products in therestraining countries are not necessary higherthan the border prices in internationalmarkets, and hence, it is not proper to assumethat the gap between the domestic prices andthe border prices in the importing countries(i.e., the so called tariff equivalent rate) canrepresent a restraining country’s protectionlevel under the MFA. Modeling the MFArestriction as an export tax in the exportingcountries is also questionable, as in many ofthese countries, such as China and India,exports of T&A products are actually encour-aged by domestic policies.

To take into account these facts, we aggre-gate the world into 13 open economies, 9 developing economies and 4 industrial

economies. (1) China, (2) India, (3) otherSouth and Southeast Asian countries, (4)Middle-east countries, (5) former SovietUnion countries, and (6) Latin Americancountries (excluding Mexico and theCaribbean countries) represent the exportingcountries restrained by the MFA quotas in themodel, (7) North African and East Europeancountries, and (8) other African countriesrepresent the developing countries free fromrestraint in the EU market, and (9) Mexicoand Caribbean countries, represent the coun-tries free from restraint in the North Ameri-can markets. The industrial economies areincluded in the model as both majorimporters and exporters. Among them, tworepresent the restraining regions: (10) theNorth America (U.S. and Canada), and (11)the EU. The rest of the two industrialeconomies represent the non-restrainingregions: (12) Australia and New Zealand, and(13) Japan, Taiwan, Hong Kong, and Korea.As Taiwan, Hong Kong and Korea currentlyhave unused quotas in their apparel exports,we treat them as developed countries in themodel. Second, we try to model the MFAquotas as the impact on some developingcountries’ export efficiency. That is, the MFAdoes not create either a price gap betweendomestic and border prices or quota rents forthe restraining countries; instead, therestraints cause difficulty for some developingcountries to export their textile and apparelproducts to the restraining countries (NorthAmerica and the EU in the model), and hencelower the efficiency of their exports. In a postMFA world, exports of textile and apparelproducts become relatively easy for thedeveloping countries [included in (1) – (6)],and hence exports grow.


1. However, Krishna and Tan (1998) point out that market power by importers in industrial countries may meanthat quota rent is actually shared with the industrial country retailer.


The model and dataThe model used for the study belongs to thefamily of global general equilibrium modelswith multi-sector and multi-region setup andintertemporal feature. Such models have beenused widely to analyze the impact of regionalor global trade liberalization and structuraladjustment programs. The model developedfor this study draws in many ways upon therecent contributions by McKibbin (1993),Mercenier and Sampaio de Souza (1994),Mercenier and Yeldan (1997), and Diao andSomwaru (2000, 2001). The model focuseson the real side of economic activities, such asproduction, consumption, physical invest-ment, trade flows as well as real terms of inter-national borrowing and lending. It incorpo-rates considerable detail in these activities –both at sectoral levels and across regions.

In the model, a representative producer foreach sector of a region makes production and investment decisions to maximize an

intertemporal profit function or the value ofthe firm. In making production decisions, thefirms choose the levels of labor and inter-mediate inputs to produce a single sectoraloutput for each time period, taking intoaccount the price of outputs, the wage rate,the prices of intermediate inputs, and thestock of capital at each time period. Outputsare either sold in the domestic market orexported to foreign markets.

In making investment decisions, the firmshave to compare the costs of investment withthe expected future returns to capital, takinginto account the price of investment goodsand the interest rate in each time period.Firms are owned by households/consumersand investment is financed by undistributedprofits. In each time period, the firm’s profits,divn,i,t which is equivalent to the gross revenue,PXn,i,t minus labor costs, wn,tLn,i,t, inter-mediate input costs, ∑jPCn,j,tITDn,j,i,t , and

investment costs, PIn,i,t˙ In,i,t(1+φn,i)˙In,i,t ⁄ Kn,i,t

distributed to households. Investment raisesthe stock of capital with waste caused bycapital adjustment costs. Formally, the firm’sproblem can be described as follows:

max Vn,i,1 = ∑T

Rn,i,tdivn,i,t + {In,i,t,Ln,i,t ,ITDn,j1,i,t ,…,ITDn,jJ,i,t} t=1

(1 + rT)1-T

divn,i,T –––––––rT

divn,i,t ≡ Pn,i,t Xn,i,t – J

∑j

PCn,j,t ITDn,j,i,t –

In,i,twn,tLn,i,t – PIn,i,t In,i,t (1+φn,i –––– )Kn,i,t

s.t. Xn,i,t = f (Ln,i,t Kn,i,t ,ITDn,j1,i,t ,…,ITDn,jJ ,i,t )

Kn,i,t+1 = (1 – δn,i,)Kn,i,t +In,i,t

In the equation, Vn,i,1 represents the value offirm i in region n at the first time period;

Rn,i,t =t∏s=1

––––11 + rn,s

is the discount factor for the

future returns; Ln,i,t, Kn,i,t, and ITDn,j,i,t are,respectively, labor, capital and intermediateinputs in the production of Xn,i,t; In,i,t is quan-tity of new capital equipment built throughinvestments at time t; δn,i is a positive capitaldepreciation rate; and φn,iIn,i,t/Kn,i,t is adjust-ment cost per unit of capital investment. Themodel assumes full employment of labor andcapital, and both factors are mobile within aregion but cannot move across regions. Whilelabor supply is fixed along the entire timepath, capital accumulates over time along thetransition and becomes constant at the steadystate where the investment only covers thedepreciation and adjustment costs.

In each region the representative house-hold owns labor, the equity in domestic firms,and foreign bonds, and allocates income toconsumption and savings to maximize an



intertemporal utility function over an infinitehorizon:

1 tMax ∑∞

t=1(–––––) U (TCn,t)1 + ρ

subject to the following current budget con-straint:

SAVn,t = wn,t L̄n,t + ∑i divn,i,t +rn,t Bn,t–1–PTC

n,t TC n,t.

where ρ is the positive rate of time preference;TCn,t is aggregate consumption at time t;SAVn,t is household savings, Bn,t-1 is the stockof foreign assets, and rn,tBn,t-1 is interest earnedfrom ownership of foreign bonds. PTC

n,t is theconsumer price index, and TIn,,t is lump sumtransfer of government revenues from excisetaxes and tariffs. We assume no governmentsaving-investment behavior. “Government”spends all its tax revenues on consumption oras transfers to the households, and hence,public sector borrowing requirement is notexplicitly modeled. TCn,t, the instantaneousconsumption, is generated from the con-sumption of final goods by maximizing aCobb-Douglas function:

TCn,t = ∏i

C bn,j

n,i,t

subject to

∑PCn,i,tCn,i,t=PTCn,t TC n,t

where Cn,i,t is the final consumption for goodi, and the consumer shares, bn,i satisfy

0<bn,i<1, and ∑bn,i=1.

International trade flows are tracked by regionof origin and destination. The variable Mn,s,i,t

represents the trade flow of commodity i fromregion n to s at time t and is an endogenous vari-

able in the model. As the sectors are quite aggre-gate, a region can export and import a sameaggregate commodity, i.e., there exists imper-fect substitution relationship between the gooddomestically produced and imported.

International borrowing and lendingoccur in the model. When a country’s currentconsumption plus its investments are above itscurrent domestic income, the country experi-ences a trade deficit. If the reverse is true, thecountry experiences a trade surplus. If thecountry does not own net foreign assets thatcan generate income from abroad, the tradedeficit has to be financed by internationalborrowing (i.e., SAVn,t is negative). Onceinternational borrowing occurs, we observeforeign capital flowing into the country. Thecurrent period’s foreign borrowing becomes a net debt burden and either increases thecountry’s total outstanding debt or reduces itsforeign assets, i.e.,

FBORn,t = J

∑i

N

∑s

(PWs,n,i,t Ms,n,i,t –PWn,s,i,tMs,n,i,t)

SAVn,t = Bn,t –Bn,t–1=rn,t Bn,t–1+FBOR n,t

where a positive FBOR implies a surplus inthe region’s foreign trade.

The model is solved for the entire timepath simultaneously and the terminate periodof the model is assumed to be a steady stateequilibrium. The data are from the GTAPdatabase version 5, pre-release 3 (GTAP,2001), including data about trade flows in theworld and production and consumption ineach country/region in 1997. The originaldata set includes 66 countries/regions and 57aggregate sectors. For the purpose of thisstudy, we aggregate the data into 13 countries/regions (listed in the previous section) and 7sectors, including cotton, other crops, live-stock, processed food, textile, apparel, and anaggregated manufacturing and services sector.



T&A vs. the rest of economy – anillustration of input-output linkagesThe study focuses solely on the impact of theMFA phase-out and removal of other relatedtrade restrictions in world T&A trade.Historical trends discussed in the previoussections suggest that world T&A trade can beexpected to grow, and hence change in theworld market structure may continue regard-less of possible policy changes. The methodused in our study does not allow us to predictfuture trends, or the pace of structural changein the future. The method we use in thefollowing analysis is sometimes called a‘counter-factual’ analysis, i.e., we ask ques-tions such as given the trend in the growth andthe change in the market structure, how theMFA phase-out will add additional growth toworld T&A trade or will cause further changein the world market structure.

The MFA phase-out and other changes intrade policy can be expected to affect T&Atrade directly. Change in a country’s textileand apparel exports can also affect the coun-try’s domestic economy as well as the worldeconomy through input-output, supply-demand, and price linkages. This study triesto capture such linkages among economicactivities and hence to evaluate the generalequilibrium impact of the MFA phase-out onthe world economy. We first present a staticexample to illustrate how such linkages, e.g.,the input-output linkages, work when acountry’s exports of textiles and apparel rise.We start by assuming that the developingcountries and regions in (1) – (6) increasetheir textile and apparel exports by 5 percent.We further assume that prices in the worldand in each country/region remain un-changed, and no change in total world T&A trade. Hence, textile and apparelexports from the regions of North Americaand the EU are assumed to decline. With fixeddomestic demand, an increase in exports

needs to be supplied by an increase in pro-duction, which requires more intermediateinputs, labor, and capital. Increased use ofintermediate inputs may promote productionin the intermediate producing sectors, such ascotton and manufacturing goods. In addition,textile products are inputs of apparel produc-tion and hence apparel exports stimulate tex-tile production. The intermediate inputs maybe imported from abroad and hence exportscan stimulate imports, which allows inter-mediate exporting countries to expand pro-duction. Putting all these linkages together,increase in the exports of T&A productsresults in rises in GDP, total labor demand,and hence household income. Similarly, adecrease in the textile and apparel exports inNorth America and the EU causes GDP andincome to fall in these regions (table 2).

It is observed in table 2 that a 5 percentincrease in exports of T&A generates differ-ential effects among the countries. Theseeffects are static and mainly reflect the differ-ence in production and trade structures acrosscountries. In general, the more export-oriented a country’s textile and apparel indus-try, the larger the T&A share in a country’sGDP, or the more available labor supply, thestronger the response of the rest economy tothe exports of T&A. If we further take intoaccount some dynamic factors, such as invest-ment and capital accumulation, the higher acountry’s saving rate, or greater a country’saccess to foreign investment, the larger theeffect of trade on the economy.

The above illustration only considers thestatic input-output linkage while the possibleeffects of change in the structure of worldT&A trade on prices, consumer demand, andfirms’ investment behavior are ignored. In thefollowing analysis, we will take into accountthese factors in an intertemporal generalequilibrium framework.



General equilibrium perspective of MFAphase-outWe simulate the possible effect of MFA phase-out by improving the efficiency of textile andapparel exports from the countries/regionsrestrained by MFA (countries/regionsincluded in (1) – (6)). In the model, weemploy a CES function to capture the imper-fect substitution between exports and sales inthe domestic markets. Due to the MFAconstraint, firms in textile sector are actuallyforced to produce more for the domesticmarkets and less for the foreign markets.Thus, we assume that in the base run the firmsin the MFA restrained countries cannot reach

an efficient allocation between production forexports and production for the domesticmarkets. With MFA phase-out, we exo-genously increase the efficiency allocationcoefficient. Moreover, we assume other tradebarriers (represented by tariff equivalent rates)on textile and apparel imports are reduced by30–40 percent in all countries (includingdeveloping countries restrained by MFA), andthe reduced tariff equivalent rates are close toeach country’s average tariff rate for othermanufacturing imports. Integrating tradebarrier reductions in the developing countriesinto the simulation is based on the ATCrequirement, and also on the fact that while


Table 2. An illustration of input-output linkages between textile and apparel exports and the rest ofeconomy(% change from the base after a 5% increase in textile and apparel exports from developing countries)

China India Mid-east USA Other L.A. EU

Demand for laborTextile 2.11 1.28 2.63 -0.95 0.16 -1.97Apparel 2.49 2.25 2.13 -0.95 0.11 -1.78

Demand for intermediatesCotton 2.05 0.22 1.59 -0.80 0.13 -1.62Textile 1.61 1.20 1.65 -0.62 0.11 -1.29Apparel 0.38 1.32 0.67 -0.80 0.02 -1.04Manufacturing & services 0.13 0.13 0.07 -0.02 0.01 -0.04

OutputCotton 1.96 0.10 1.04 -0.57 0.11 -0.68Textile 2.11 1.28 2.63 -0.95 0.16 -1.97Apparel 2.49 2.25 2.13 -0.95 0.11 -1.78Manufacturing & services 0.08 0.06 0.03 -0.01 0.00 -0.02

ImportsCotton 1.96 0.11 1.23 -0.80 0.11 -1.58Textile 1.49 0.51 1.21 -0.46 0.08 -0.83Manufacturing & services 0.09 0.07 0.03 -0.01 0.00 -0.02

Macroeconomic effectsTotal labor demand 0.21 0.12 0.07 -0.02 0.01 -0.05Income 0.15 0.09 0.05 -0.02 0.01 -0.03GDP 0.19 0.10 0.06 -0.02 0.01 -0.04


many developing countries face the restraintsin their textile and apparel exports under theMFA, there exist high barriers on imports ofT&A goods in some of these countries, suchas China and India. Hence, there are alsoconflicts over textile and apparel trade amongthe developing countries. For this reason, weassume when the MFA is finally dismantled,other trade barriers (represented by tariffequivalents here) have to be also reduced inthe world. The econometric results discussedin the previous section concerning linkagebetween the share of T&A trade in GDP andgrowth in GDP are also incorporated into themodel as a function linking the total factorproductivity (TFP) with the trade share inGDP for each region.

1. More textile and apparel trade in the worldWorld T&A trade increases in the model dueto the MFA phase-out and tariff reductions.Comparing with the base, world T&A tradeincreases by 5–16 percent annually in thesimulated time period of 25 years. That is, ifworld T&A trade were expected to grow 8 per-cent annually in the next 25 years after 2005,then due to the MFA phase-out, the newannual growth rate in the model is about 8.5percent average. With this higher growth rate,trade level (not growth rate) would be 5–16

percent higher annually than in the base yearas indicate by the trend (figure 8). The gainsin world T&A trade are about $20 billion inthe early periods of the post-MFA and couldincrease to $200 billion in the long run (figure9). Consistent with the trend in the historicaldata, the model results show that worldapparel trade will increase twice as fast as tex-tile trade in the post MFA world (figure 10).

It is obvious that the increase in worldtrade is mainly due to more apparel exportsfrom developing countries, as their exportsbecome more efficient in the model. However,the model results also show that as a group,exports of T&A by the industrial countries,especially their textile exports, do not fall, butinstead rise by about 4–6 percent from thebase (figure 11). As we discussed in the aboveexample, textile and apparel sectors have


0200

400600800

100012001400

16001800

90 94 98 02 06 10 14 18

Data/trendModel result

0

40

80

120

160

200

05 09 13 17

4

8

12

16

20

1 5 9 13 17 21 25

TextileApparelAll

Figure 8. World T&A Trade – Data/Trend andModel Results (Billion US dollar)

Figure 9. Change in World T&A Trade –Difference between Trend and Model Results (Billion US dollar)

Figure 10. Increase in World Textile and Apparel Trade(Simulation results, % change from the base)


strong inter-linkages. When developing coun-tries increase their apparel exports, which aremainly labor-intensive products, theirdemand for industrial countries’ textile prod-ucts, which are often capital intensive, risesand hence industrial countries’ exports alsoincrease. This result indicates that protectivepolicies in the world T&A trade not onlyrestrain exports of developing countries, butalso limit the possible exports of industrialcountries and hence reduce overall worldtrade flow. While liberalization may causecompetition between rich and poor countriesin world T&A markets, it can also induceinterdependency among countries and henceenlarge world trade.

2. Countries’ market shares changeThe simulation results show that world mar-ket shares for some countries or regions changein the post MFA period. In total, developingcountries gain 4 percentage points of the worldT&A market from the industrial countries inthe base year. As the largest exporter among thedeveloping countries, China is observed tohave the most rapid growth in textile andapparel exports. This allows China to gainalmost 3 percentage points of the world T&A

market (figure 12). Following is the region ofthe other Asian countries, capturing more than2 percent more world market share (table 3).Exports in some developing countries/regionsthat are free from quota restraints under theMFA decline and in total, they lose about 20percent of their T&A markets (equivalent to2.3 percentage points of world total T&A markets) to those countries restrained by theMFA.

Changes in market shares may be under-estimated due to the regional aggregation inthe model. The aggregation causes tradeamong the countries that are aggregated intoa region to be ignored. In some regions, espe-cially in the EU, intra-regional trade is likelyto be replaced by imports from countries out-side the EU. For example, total EU exports ofT&A goods accounted for 30 percent of theworld T&A trade in the last decade, on aver-age. As about half of the exports are intra-EUtrade, which are not taken into account in themodel, the model only captures the exports ofthe EU to the rest of the world, which is about14.4 percent of world total textile and apparelexports in the base. This share falls to 12.7percent in the simulation, which implies thatas an exporter, EU loses about 13 percent of


-10

0

10

20

30

40

1 5 9 13 17 21 25

China

O.Asia

Mexico&Caribbean

E.Europe

India0

5

10

15

20

25

1 5 9 13 17 21 25

All-ICAll-LDCTextile-ICApparel-LDC

Figure 11. Increase in Textile and Apparel Exports(Simulation results, % change from the base)

Figure 12. Increase in Textile and Apparel Exports inSelected Regions (Simulation results % change from the base)


its markets outside the EU. However, if theintra-EU trade were included in the model,then the developing countries would not onlygain markets in the rest of the world, but alsowithin the EU.

Second, while the econometric results onthe linkage between T&A trade share in GDPand growth in GDP are incorporated in thesimulation, the model does not take intoaccount for the growth trend in T&A tradedue to population growth or technologicalchange. The historical data show that growthis a major reason for the structural change inthe world T&A market. In the early stage ofdevelopment, a country tends to export moreT&A products, especially high labor-inten-sive apparel products. When the countrybecomes wealthy and its labor cost increases itstarts to lose its comparative advantage in pro-ducing and exporting labor-intensive apparelgoods and shifts to other products which aremore capital or human-capital intensive.Dynamics of shifts in comparative advantageare observed in the process of economic devel-opment in countries such as Japan, Korea,Taiwan, and Hong Kong, all of which werelarge textile and apparel exporters in the past.While the model fails to capture such dynam-

ics in a country’s comparative advantage, themodel results seem to tell us that only whenthe MFA phase-out strongly affects thegrowth patterns and growth rates of develop-ing countries do such trade reforms have a sig-nificant impact on the market structure ofworld T&A trade.

3. Positive welfare effects among the countries From the world’s perspective, a more liberal-ized textile and apparel sector implies moreefficient allocation of resources and hencehigher global welfare. Moreover, in theintertemporal model with a linkage betweenTFP and trade-GDP share, a more liberalizedtextile and apparel sector also stimulatesinvestment, while a higher trade-GDP sharecauses TFP level to be higher. Thus, all coun-tries, not only those restrained by the MFA,but also the other ones, gain from the liberal-ization. We use the well-accepted equivalentvariation (often referred to as the willingnessto pay) to measure the social welfare gains orlosses in the post MFA world. In a static analy-sis, the welfare effects are often measured byusing the status-quo (pre-reform) prices as thebase, and addresses the question: what incomewould be equivalent to the change brought


Table 3. Market shares for selected countries/regions in world total textile and apparel exports

—— Simulation results ——

Base Year 5 Year 10 Year 15 Year 20

Developing countries 59.57 60.20 61.32 62.41 63.49(1) China 19.69 20.50 21.24 21.91 22.52(2) India 4.40 4.43 4.57 4.72 4.88(3) Other Asia 13.00 13.68 14.18 14.70 15.22(4) Middle East 5.03 5.22 5.39 5.57 5.76(7) E. Europe 6.50 6.02 5.80 5.59 5.38(9) Mexico and Caribbean 6.09 5.51 5.32 5.14 4.96

Industrial countries 40.43 39.80 38.68 37.59 36.51(10) N. America 6.61 6.31 6.13 5.95 5.77(11) EU 14.39 13.91 13.48 13.06 12.65


about by liberalizing world T&A trade (Var-ian, 1984)? We borrow this concept in ourdynamic analysis by evaluating the welfareeffects within each time period and the entiretime path by summing the discounted valueof this measure over time.

As expected, most countries whose textileand apparel exports are restrained by the MFAgain more post-MFA (table 4). However, aswe discussed in the previous section, due tothe difference in production and trade struc-ture among countries, the same level ofchange in textile and apparel exports can affectthe rest of the economy differentially acrosscountries. For this reason, the welfare gains canbe different among countries that benefitdirectly from the MFA phase-out. For exam-ple, textile and apparel exports increase morein the region of other Asian countries than thatin India (figure 12). However, from a welfarepoint of view, India gains more than the gainof the region of other Asian countries. One rea-son is that the T&A sector contributed more

to GDP in terms of the value added (table 5).For this reason, exports of T&A create moreemployment opportunities and hence GDPrises more in India than that in the region ofother Asian countries (figure 13).

For the developing countries free fromquota restraints under the MFA and industrial(importing) countries, the welfare effect ofMFA phase-out is ambiguous analytically.


3,0

4,0

5,0

6,0

7,0

1 5 9 13 17 21 25

China O.Asia

India Mid-east

Table 4. Welfare effect in the simulation

Year -5 Year -10 Year -15 Year -20Billion $ % Billion $ % Billion $ % Billion $ %

Developing countries(1) China 19.50 3.76 21.96 4.24 22.95 4.43 23.67 4.57(2) India 7.30 2.39 8.39 2.75 9.30 3.05 10.79 3.53(3) Other Asia 7.53 1.34 8.43 1.50 9.13 1.62 10.22 1.81(4) Middle East 9.24 1.65 10.52 1.88 11.62 2.08 13.43 2.40(5) Formal Soviet Union 15.33 3.32 17.48 3.79 19.01 4.12 20.88 4.52(6) O. Latin America 17.71 1.44 20.47 1.66 22.93 1.86 26.88 2.18(7) E. Europe 4.45 1.06 5.34 1.27 6.02 1.43 6.93 1.65(8) Africa 9.01 1.69 10.06 1.89 10.96 2.06 12.39 2.33(9) Mexico and Caribbean 1.93 0.50 2.33 0.60 2.70 0.69 3.29 0.84

Industrial countries(10) N. America 0.86 0.01 4.24 0.06 7.22 0.10 11.53 0.16(11) EU 5.60 0.08 9.98 0.14 13.87 0.20 19.39 0.28(12) Australia and New Zealand 0.33 0.09 0.56 0.16 0.77 0.21 1.08 0.30(13) Japan, Korea and Taiwan 4.33 0.12 7.11 0.20 9.56 0.27 13.11 0.37

Figure 13. Increase in GDP in Selected Regions (Simulation results % change from the base)


The direct effect of MFA phase-out may benegative for these countries’ welfare, as theirtextile and apparel exports and hence produc-tion may fall due to competition frompreviously restrained countries. However,consumers in these countries may benefitfrom lowered prices for T&A imports. Theoutput of the other sectors may rise byemploying resources released from the pro-tected T&A sector. Even within T&A sector,the indirect effect may be positive as theapparel exporting countries may increasedemand for imported textile products.

We observe a positive welfare effect ofMFA phase-out in the regions free from quotarestraints as well as in industrial importingcountries (table 4). Even though the compe-tition from other suppliers who used to berestrained by the quota in the EU and NorthAmerican markets causes the quota-freeregions’ exports to fall (figure 12), as world

apparel prices decline by about 5 percent postMFA, consumers in these countries are betteroff by consuming cheaper commodities.

In total, the world aggregate welfareincreases by $88 billion in the short-run andmore than $203 billion in the long run postMFA in the model, which are equivalent to0.38 and 0.88 percent of world total con-sumption in the short- and long-run, respec-tively. The developing countries as a groupbenefit more from a more liberalized worldT&A market, and total welfare gains for themare about $85.5 billion (equivalent to 1.7 per-cent of total consumption) in the short-runand $145.5 billion (2.9 percent of totalconsumption) in long-run. The developedcountries as a group also gain $2.6–$58billion in the short- and long-run, equivalentto 0.01 to 0.32 percent of total consumptionin all the developed countries in the short andlong run, respectively.


Table 5. Share of textile and apparel in GDP, 1997

T&A value-added1 Exports of T&A2

Developing countriesChina 5.26 5.38India 3.25 2.64Other Asia 3.02 3.89Middle East 1.36 1.68Former Soviet Union 1.32 0.43Other Latin America 3.29 0.16East Europe 2.86 3.10Other Africa 2.14 1.08Mexico and Caribbean 3.29 3.00

Industrial countriesNorth America 1.22 0.18European Union 1.06 0.40Australia and New Zealand 0.87 0.47Japan, Korea and Taiwan 1.20 0.86

1 GDP at factor cost2 GDP at expenditureData source: GTAP database version 5.


ConclusionsThis study focuses on the possible impact ofMFA phase-out on the world economy. Thestudy starts by analyzing trends in worldtextile and apparel trade. The developingcountries were a growing factor in world T&Atrade in recent decades. As 70 percent of worldT&A products are imported by the industrialcountries, a more open and freer market in theindustrial countries is an important conditionfor developing countries to maintain theirgrowth momentum. However, about 50percent of industrial countries’ markets arenot available for developing countries andintra-EU trade still accounts for half of totalEU imports of T&A products. The restraintsof MFA on the developing countries’ T&Atrade may partially explain this situation.

The relationship between trade in textileand apparel and the standard of living isempirically investigated in the study. Usingtrade data from 91 countries over 37 years, theregression results indicate a strong positivelinkage between trade in T&A and income perperson. This contrasts with the negative link-age between agricultural trade and incomeestimated by the study.

The prospect of a post MFA world isanalyzed by using an intertemporal generalequilibrium model. As the model frameworkdoes not allow us to predict the trend in worldT&A trade, the study focuses solely on thepossible impacts of the MFA phase-out onworld T&A trade and the world economy.The study finds that MFA phase-out wouldenlarge world trade of T&A and developingcountries will further gain market share inworld total exports. However, without evalu-ating the possible and differential impact ofMFA phase-out on the economic growthpattern and growth rate among countries, themodel fails to capture a significant change inworld T&A market structure.

Almost all countries, including both the

developing countries restrained by the MFAquotas and free from the MFA quotas, and theindustrial countries, gain in term of socialwelfare post-MFA in the model. Even thoughthe developing countries currently free fromMFA quota restrains may lose their marketshares, as world T&A prices are lowered byimproving the efficiency of world T&A tradepost-MFA, consumers are better off byconsuming cheap commodities.

ReferenceArellano M., and S. Bond 1991. “Some Tests of

Specification For Panel Data: Monte Carlo Evi-dence and an Application to Employment Equat-ions,” Review of Economic Studies, 58(2): 277-297.

Cline, W. 1990. The Future of World Trade in Textile andApparel, Institute for International Economics.

Diao, X., and A. Somwaru 2001. “A Dynamic Evalua-tion of the Effects of A Free Trade Area of theAmericas – An Intertemporal, Global GeneralEquilibrium Model,” Journal of Regional Integration16(1): 21-47.

Diao, X., and A. Somwaru 2000. “An Inquiry onGeneral Equilibrium Effects of MERCOSUR – AnIntertemporal World Model”, Journal of PolicyModeling 22(5): 557-588.

Frankel, J.A., and D. Romer 1999. “Does Trade CauseGrowth?” The American Economic Review 89(3):379 – 399.

Krishna, K.M. and L.H. Tan 1998. Rages and Riches:Implementing Apparel Quotas under the Multi-FiberArrangement. Ann Arbor: University of MichiganPress.

McKibbin, W.J. 1993. “Integrating macroeconometricand multi-Sector computable general equilibriummodels,” Brookings Discussion Papers in Inter-national Economics, No. 100.

Mercenier, J. And E. Yeldan 1997. “On Turkey’s tradepolicy. Is a customs union with EU enough?”European Economic Review, April 1997: 871-880.

Mercenier J. and M. Sampaïo de Souza 1994. “Struc-tural adjustment and growth in a highly indebtedmarket economy: Brazil,” in J. Mercenier and T.Srinivasan (eds.) Applied General EquilibriumAnalysis and Economic Development, Ann Arbor:University of Michigan Press.

Spinanger, D. 1998. “Textile Beyond the MFA Phase-



Out,” CSGR Working Paper No. 13/98, Center forthe Study of Globalization and Regionalization,The University of Warwick, July 1998.

SAS Institute Inc. 1979. SAS Technical Report S-106,TSCSREG: A SAS Procedure for the Analysis of Time-Series Cross-Section Data, Cary, NC: SAS InstituteInc.

Thomas, H., and J. Whalley (eds) 1998. “The Imple-mentation of the Uruguay Agreement on Textile andClothing: Some General Issues,” Uruguay RoundResults and the Emerging Trade Agenda: Quantita-tive-based Analyses from the Development Perspective,UNCTAD, United Nations.

Trela, I., and J. Whalley 1990a. “Global Effects ofDeveloped Country Trade Restrictions on Textilesand Apparel,” The Economic Journal 100 (Decem-ber): 1190-1205.

Trela, I., and J. Whalley 1990b. “Unraveling the Threadsof the MFA,” in Hamilton, Carl B. (ed.), TextilesTrade and the Developing Countries: Eliminating theMultifibre-Arrangement in the 1990’s, The WorldBank.

Varian, H.R. 1984. Microeconomic Analysis, SecondEdition, New York and London: W.W. Norton &Company.

Whalley, J. 1999. “Notes on Textile and Apparel in theNext Trade Round,” prepared for a conference onDeveloping Countries in the Next WTO TradeRound, Harvard University.



A Global Perspective of Liberalizing World Textile and ... · 03 NOPEC 28 (2) Diao NY 11.12.02...

Documents

Transcript of A Global Perspective of Liberalizing World Textile and ... · 03 NOPEC 28 (2) Diao NY 11.12.02...