NBER WORKING PAPER SERIES HOW CLOSE IS ASIA TO ALREADY BEING A TRADE BLOC? Chunding Li John Whalley

Transcription

1 NBER WORKING PAPER SERIES HOW CLOSE IS ASIA TO ALREADY BEING A TRADE BLOC? Chundng L John Whalley Workng Paper NATIONAL BUREAU OF ECONOMIC RESEARCH 1050 Massachusetts Avenue Cambrdge, MA August 2014 We are grateful to the Ontaro Research Fund for fnancal support and to semnar groups at the Xamen Unversty, Bejng Normal Unversty, Qnghua Unversty, and Unversty of Internatonal Busness and Economcs for comments. The vews expressed heren are those of the authors and do not necessarly reflect the vews of the Natonal Bureau of Economc Research. NBER workng papers are crculated for dscusson and comment purposes. They have not been peerrevewed or been subject to the revew by the NBER Board of Drectors that accompanes offcal NBER publcatons by Chundng L and John Whalley. All rghts reserved. Short sectons of text, not to exceed two paragraphs, may be quoted wthout explct permsson provded that full credt, ncludng notce, s gven to the source.

2 How Close s Asa to Already Beng A Trade Bloc? Chundng L and John Whalley NBER Workng Paper No August 2014 JEL No. D58,D61,F15 ABSTRACT FTA blateral and regonal negotatons n Asa have developed quckly n the past decade movng Asa ever closer to an economc unon. Unlke Europe wth the EU and the 1997 treaty of Rome and the 1993 NAFTA n North Amercan, Asan economc ntegraton does not nvolve a comprehensve trade treaty, but an acceleratng process of buldng one blateral agreement on another. For countres n Asa there s negotaton of a Chna-Japan-Korea agreement, a Chna-Inda agreement, a Trans-Pacfc Partnershp (TPP) agreement, and a Regonal Comprehensve Economc Partnershp (RCEP). Ths paper uses a ffteen-country global general equlbrum model wth trade costs to numercally calculate Debreu dstance measures between the present stuaton and potental full Asa ntegraton n the form of a trade bloc. Our results reveal that these large Asa economes can be close to full ntegraton f they act tmely n agreements through negotaton. All Asa countres wll gan from Asa trade bloc arrangements except when the Asa FTA can only elmnate tarffs. These countres gan wll ncrease as blateral non-tarff elmnaton deepens. Larger countres wll gan more than small countres. Asa FTA, Asa Unon and RCEP wll beneft member countres more than ASEAN+3. Global free trade wll beneft all countres the most. Chundng L Insttute of World Economcs and Poltcs Chnese Academy of Socal Scences No.5 Janguomennedaje Bejng, PRC Postcode: chundngl@gmal.com John Whalley Department of Economcs Socal Scence Centre Unversty of Western Ontaro London, ON N6A 5C2 CANADA and NBER jwhalley@uwo.ca

3 1. Introducton In Asa formal FTA arrangements have developed more slowly than n North Amercan and Europe. Untl now, there s no fully comprehensve FTA n Asa, but nstead a number of lesser arrangements. At present, some mportant negotatons are n progress, ncludng the Regonal Comprehensve Economc Partnershp (RCEP), the Trans-Pacfc Partnershp (TPP), the ASEAN Plus Three (APT), an Chna-Japan-South Korea FTA, and others. In the future, Asa may thus probably move to some form of comprehensve trade bloc. Under these crcumstances, t s helpful to numercally explore how close Asa s already to beng a trade bloc as well as what wll be the effects of potental Asan blocs lke arrangements on ndvdual economes. Here we utlze the Debreu (1951, Econometrca) coeffcent as a measure of closeness to beng a formal tradng bloc. Lttle research has used ths measure n applcaton, and there are only a few papers n whch ths measure plays a role n the theoretcal structure, ncludng Raa (2008), Ahlhem et al (1988), Anderson and Neary (1996), Brown and Srnvasan (2007). Rezman, Whalley and Zhang (2006) also construct dstance measures to estmate dstance of economes from free trade, but do not use Debreu s dstance coeffcent. Computatonal general equlbrum methodology has been wdely used to explore free trade agreement (FTA) effects. The Hcks (1943) welfare varaton measures are the most common ndex used for analyzng FTA effects. Ths paper thus njects the Debreu dstance nto explorng FTA effects. 3

4 Lterature on Asa trade blocs s sparse and mostly analytcal, such as Shno (2012), Fukunaga and Isono (2013), Wllams (2012), Lews (2011), Ezell and Atknson (2011), and Stubbs (2002). Some earler research numercally explores the effects of regonal free trade agreements (see the survey by Lloyd and MacLaren (2004)), but few papers try to capture the potental effects of recent FTA developments n Asa. Petr et al (2011), Itakura and Lee (2012), Kawa and Wgnaraja (2008) are the exceptons. Our paper departs from these n usng a Debreu dstance ndcator to explore the closeness of Asa trade arrangements to beng formal trade blocs. We also use Hcks (1943) compensaton and equvalent varaton measures to compare these results n senstvty analyss. Our global general equlbrum model has 15 countres, whch are Chna, the US, the EU, Japan, South Korea, Canada, Mexco, Inda, AN (Australa and New Zealand), CP (Chle and Peru), BMSV (Brune, Malaysa, Sngapore and Vetnam), CILMPT (Camboda, Indonesa, Laos, Myanmar, Phlppne, and Thaland), ODDC (other developed countres), ODC (other developng countres), and the Rest of World (ROW). Each country produces two goods (Manufaturng goods and non-manufacturng goods) wth two factors of producton (Captal and Labour). The model uses an exogenous trade mbalance structure and ncludes trade costs as trade barrers. We use a trade cost calculaton method that recognzes lmtatons of data by utlzng an estmaton method that follows Wong (2012) and Novy (2008). Our calculaton results show that the dstances of present stuaton to Asa FTA, Asa Unon, RCEP are nearly the same, but the dstance to ASEAN+3 s nearer, whch 4

5 means ASEAN+3 wll beneft nvolved countres less than other Asa FTAs. Dstances to global free trade are farther than all regonal Asa FTAs, whch means that the global free trade agreement wll gan all countres more than regonal free trade arrangements. All countres n Asa trade bloc wll gan from FTA arrangements when tarff and non-tarff can both been elmnated. But f the FTA arrangements can only remove tarff, some of bg countres wll be hurt. All countres gan wll ncrease when non-tarff barrers remove more. In the meanwhle, large countres wll gan less than small countres. 5

6 2. Asa FTAs and Ther Developments Asa has not formed comprehensve free trade agreement, but multcounty Asan FTAs have been the subject of dscusson partcularly after the 2008 global fnancal crss. These nclude ASEAN+3, ASEAN+6 and RCEP (Regonal Comprehensve Economc Partnershp). A comprehensve Asa FTA may come nto beng n the future, and result n a formal Asan bloc lke NAFTA (North Amerca Free Trade Agreement) and EU (Europe Unon), but for now, Asan trade ntegraton rests on the accommodaton of overlappng sub Asan FTAs rather than blocs n Europe and North Amercan, whch reflect the desgn of an overachevng arrangement n the absence of pror agreements. In contrast to Afrca, the Amercas and Western Europe, before 1992 Asa had no regonal or blateral free trade agreements (FTAs). In January 1993, the ASEAN Free Trade Area (AFTA) came nto force over the last decade; Asa has seen a rapd ncrease n FTAs. Accordng to the Asan Development Bank (ADB) FTA statstcs as of 2013, there are 257 FTAs n the Asa Pacfc Area (compared wth just 50 n 2000), wth 132 concluded, 75 under negotaton and 50 proposed. Wthn these, 189 are blateral, and 68 are plurlateral. The leadng countres nvolved n Asa FTAs are Sngapore (37), Inda (34), Korea (32), Chna (27), Pakstan (27), Thaland (26), Japan (26) and Malaysa (26) 1. Among Asa FTAs, larger agreements such as ASEAN act as a hub. For example: ASEAN s own FTA s complemented by FTAs between ASEAN and other economes 1 See Asa Development Bank Asa Regonal Integraton Center statstcs, 6

7 (Chna, Inda, Japan, Korea); FTAs between ndvdual ASEAN countres and other countres and comprehensve economc partnershp agreements for East Asa. In Asa area, there are about 30 major FTAs n effect 2 (see Table 1). Table 1: Major FTAs In Effect In the Asa Regon FTAs Date In Effect Laos-Thaland June, 1991 AFTA (ASEAN) January, 1992 Japan-Sngapore November, 2002 ASEAN-Chna January, 2004 Thaland-Inda September, 2004 Sngapore-Inda August, 2005 Sngapore-South Korea March, 2006 Japan-Malaysa July, 2006 ASEAN-South Korea June, 2007 Japan-Thaland November, 2007 Japan-Indonesa July, 2008 Japan-Brune July, 2008 ASEAN-Japan December, 2008 Japan-Phlppne December, 2008 Sngapore-Chna January, 2009 Japan-Vetnam October, 2009 ASEAN-Inda January, 2010 South Korea-Inda January, 2010 Malaysa-Inda July, 2011 Japan-Inda August, 2011 Source: Shno (2012). Among FTAs stll under negotaton n Asa regon, the most promnent arrangements are ASEAN Plus Three (APT) and the Regonal Comprehensve Economc Partnershp (RCEP). The proposed Trans-Pacfc Partnershp (TPP) also nvolves the Asa Pacfc regon, although we do not nclude t n the analyss of ths paper snce t exclusvely ncludes non-asan countres. The ASEAN Free Trade Area (AFTA) s a trade agreement nvolvng the Assocaton of Southeast Asan Natons sgned n January 1992 n Sngapore. When the AFTA was orgnally sgned, ASEAN had sx members, namely Brune, Indonesa, Malaysa, Phlppnes, Sngapore and Thaland. Vetnam jonted n 1995, Laos and 2 See K. Shno, Overvew of Free Trade Agreements n Asa, BRC Research Report No.9, 2012, IDE-JETRO. 7

8 Myanmar n 1997 and Camboda n AFTA now comprses ten countres of ASEAN. ASEAN Plus Three (APT) coordnates cooperaton between ASEAN and the three East Asan natons of Chna, Japan, and South Korea. The APT s the latest example of East Asan regonal cooperaton. It reflects earler proposals for an Asan Free Trade Bloc, such as Korea s call for an Asan Common Market n 1970 and Japan s 1988 suggeston for an Asan Network. The frst APT s leaders meetng was held n 1996, untl now t has 13 summts; the latest one n October The Regonal Comprehensve Economc Partnershp (RCEP) members nclude the 10 ASEAN countres and prospectve 6 FTA partners: Australa, Chna, Inda, Japan, Korea and New Zealand. The RCEP negotaton ams to conclude by the end of 2015 and cover more than 3 bllon people. The area has a combned GDP of about $17 trllon, and accounts for about 40% of world trade. The dea of such a new trade treaty was frst mooted at the 19th ASEAN Summt n November 2011, when leaders of the ten ASEAN member states adopted general prncples for broadenng and deepenng ASEAN s engagement wth ts FTA Partners, and sgnaled ASEAN s commtment to play a central role n the emergng regonal economc archtecture. RCEP s frst round mnsteral meetng was held n Brune n August 2013, and the second round of RCEP negotatons was conducted on September n Brsbane, Australa. The 16 economc mnsters of Asean+6 have agreed to fnalze the Regonal Comprehensve Economc Partnershp by 2015, when the ASEAN 8

9 Economc Communty takes full effect 3. The am of ths paper s to numercally calculate how close Asa s already surrogate trade bloc, pror to the concluson of a RCEP negotaton. FTA development n Asa area has receved consderable attenton. But the Asan area ncludes a lot of large countres and stll does not have a hgh standard FTA. In the meanwhle, Asa countres all seem eager to form a FTA to promote regonal ntegraton after the global fnancal crss n The ssue of an Asa trade bloc may well become the most mportant FTA development n the near future and how close the Asan economes are already towards ths goal s a key element n the debate. 3 See P. Pratruangkra, Economc mnsters agree to establsh ASEAN+6 FTA by 2015, The Naton webste,

10 3. Debreu s Coeffcent and Dstance Measures To measure the remanng dstance for the Asan closeness to a comprehensve FTA we use a calbrated numercal general equlbrum model n whch we ncorporate a modfed form of Debreu the resource utlzaton coeffcent of neffcency. 3.1 The Debreu Coeffcent of Resource Utlzaton Debreu (1951) proposes a measure of the neffcency of the allocaton of resources n an economy by calculatng how much less resources could attan the same level of satsfacton to consumers. The rato of how much less resources can show the neffcency level of the allocaton and we take ths as the measure of dstance between a mult-country equlbrum under a potental Asan trade arrangements and what a comprehensve agreement would permt. To expost the Debreu measure, consder an economy whch comprses m consumers wth preference relatonshps and observed consumpton vectors X R (=1,, m), 0 l where l s the number of commodtes. Y l R s the set of possble nput vectors (net quanttes of commodtes consumed by the whole producton sector), ncludng the observed one 0 y. A combnaton of consumpton vectors and an nput vector s feasble f the total sum (the economy-wde net consumpton), does not exceed the vector of utlzable physcal resources, Z R. 0 l 0 Z s assumed to be at least equal to the sum of the observed consumpton and nput vectors, ensurng the feasblty of the latter. The set of net consumpton vectors that are at least as good as the observed ones s 0 X X X, 1,, m Y (1) where B stands for the better set. The mnmal resources requred to attan the same levels 10

11 of satsfacton that come wth 0 X belong to B mn, the South-western edge or subset of elements z that are mnmal wth respect to. Assume that preferences are convex and contnuous, and that producton possbltes form a convex and closed set, then a supportng prce vector pz ( ) 0 exsts such that Z ' B and whch mples pzz ( ) ' pzz ( ) (2) Where p 0 mples all components are postve. The Debreu coeffcent of resource utlzaton s defned by 0 mn max ( ) / ( ) (3) z pzz pzz z B and the coeffcent measures the dstance from the set of mnmally requred physcal resources, z mn B, to the utlzable physcal resources, 0 z, n the metrc of the supportng prces. Debreu shows that the dstance or the max n equaton (2) s attaned by z z B 0 mn (4) In words, the Debreu coeffcent of resource utlzaton s the smallest fracton of the actually avalable resources that would permt the achevement of the levels of satsfacton that come wth 0 X. Coeffcent s a number between zero and one, the latter ndcatng full effcency 4 (Raa, 2008). 3.2 Debreu Dstance Measures Used n Smulaton In ths paper we use the Debreu s dstance coeffcent to calculate the dstance from present trade agreement n Asa to an Asa trade bloc. Dstance n our paper s measured by the shrnkage n the endowment of Asa such that a full trade bloc scenaro could be attaned. We use a numercal global general equlbrum model to calculate the dstance for 4 See: Raa, T.T. Debreu s coeffcent of Resource Utlzaton, the Solow Resdual, and TFP: the Connecton by Leontef Preferences. Journal of Productvty Analyss, 30, pp

12 present FTAs to a possble comprehensve Asan trade bloc. We compute two knds of dstance, one s ndvdual country dstance, and the other s whole Asa bloc dstance. If we assume an N country world, each countres denoted as (1,, m). In the base case (present status), a global general equlbrum has a consumpton set ( C,, C N ), utlty * * 1 set ( U,, U N ), endowment set * * 1 ( E,, E N ), and factor demand set * * 1 ( K,, K N ). Under * * 1 an Asa bloc scenaro, the global general equlbrum has a consumpton set denoted as ( C1,, C N ), utlty set ( U1,, U N ), endowment set ( E1,, E N ), and factor demand set ( K,, K N ). 1 For the dstance we know to an Asa bloc, we denote the dstance measure as. Under the Debreu s resource utlzaton concept, we defne as mn st.. (1) GE, (2), K E, ( country n the bloc) U U ( country n the bloc * (3),, ) (5) where s the dstance of the present stuaton to that under an Asa bloc. The utlty of all countres n the bloc wll be not less than n ther base case stuaton. Fgure 1(II) shows ths concept of dstance a sngle country. In the base case, the equlbrum consumpton pont s B; n the Asa bloc scenaro, the utlty possblty fronter s UPF 1. If we keep the utlty n the new scenaro as not less than n base case, the utlty possblty fronter can shrnk to UPF 2, and s the rato between UPF 1 and UPF 2. For the dstance of an ndvdual country to Asa bloc, we can agan denote the dstance measure as, and defne as 12

13 mn st.. (1) GE, (2) K E, (3) U U * (6) where s the dstance of country from the present RTA stuaton to an Asa RTA bloc. It shows the mnmum endowment rato that keeps the utlty of country n the bloc not less than n the base case. Fgure 1(I) shows the dstance measured for ndvdual countres. We keep the foregn country s consumpton fxed, and own country consumpton moves from pont B to pont A n utlty space, and the own country s utlty possblty fronter changes from UPF 1 to UPF 2, the rato s the dstance measure for country. Fgure 1: Debreu Dstance Measure Relatve to Full Asa FTA Foregn Goods Foregn Goods UPF 2 UPF 2 B A B UPF 1 UPF 1 O O A λ Own Goods λ Own Goods (I) Indvdual Country (II) Whole If the dstance measure s larger than 1, t mples that the trade bloc arrangement cannot mprove welfare of the whole trade bloc of countres and not for ndvdual countres, whch means the trade bloc has negatve effects. If the dstance measure s smaller than 1, t mples that the trade bloc arrangement decreases welfares for the whole trade bloc and for ndvdual countres. The nearer the dstance measure to 1, then the dstance between present stuaton and an Asa trade bloc s nearer. A small dstance value means the trade bloc effects are sgnfcant and the present stuaton to trade bloc s large. 13

14 We thus use a revsed Debreu dstance ndcator, whch equals D (1 ) 100 D (1 ) 100 (7) where D or D are separately the Debreu dstance ndcators for the whole FTA groups and ndvdual countres. When D or D are postve, t means that FTA arrangements have postve effects for the whole group or ndvdual countres and they can keep ther orgnal welfare level by decreasng D or D percent of ther endowment; When D or D are negatve, ths means FTA arrangements have negatve effects for the whole group and for ndvdual countres. If they want to keep ther orgnal welfare level they must ncrease D or D percent of ther endowment. 14

15 4. Model Structure and Trade Cost Calculaton We use a global general equlbrum model wth an exogenous trade mbalance and trade barrers based on trade costs to calculate Debreu dstance measures to assess how far close s already to a surrogate trade bloc. 4.1 An Exogenous Trade Imbalance GE Model wth Trade Costs Our sngle perod global general equlbrum model has ffteen countres and each country produces two goods wth two factors. These ffteen countres are Chna, the US, the EU, Japan, Korea, Canada, Mexco, Inda, AN (Australa and New Zealand), CP (Chle and Peru), BMSV (Brune, Malaysa, Sngapore, and Vetnam ), CILMPT (Camboda, Indonesa, Laos, Myanmar, Phlppne, and Thaland), ODDC (other developed countres, ncludng Swtzerland, Norway, Israel, and Iceland), ODC (other developng countres, ncludng Brazl, Russan, Egypt, Argentna, and South Afrca), and ROW (the rest of the world). The two goods are manufactured goods and non-manufactured goods. We use agrcultural and servce sectors as producers of non-manufactured goods and assume these cannot be traded between countres. The two factors n each country are labor and captal, whch are treated as ntersectorally moble but nternatonally mmoble. We nclude trade mbalances n the model, wth the trade mbalance for each country assumed to be fxed and exogenously determned. On the producton sde of the model, we assume CES technology for producton of each good n each country 15

16 l l l 1 1 l l l l l l l l l 1 Q [ ( L ) (1 )( K ) ], country, l goods (8) where l Q s the output of the lth ndustry (ncludng tradable goods and non-tradable goods) n country, l L and K l are the labor and captal nputs n sector l, l s the scale parameter, l s the dstrbuton parameter and substtuton. Frst order condtons mply the factor nput demand equatons, l s the elastcty of factor K l l l L (1 ) l l l Q l w (1 ) l 1 [ [ ] (1 )] l l K w (9) l l l l l l (1 ) 1 l l K Q w L [ (1 )[ ] ] l l L (1 ) w (10) where K w and L w are the prces of captal and labor n country. On the consumpton sde, we use the Armngton assumpton of product heterogenety across countres, and use a nested CES utlty functon for each country ( M, NM ) [ 1 ( M ) 2 ( NM ) ] U X X X X, country (11) where NM X denotes the consumpton of non-manufacturng goods n country, M X denotes the consumpton of composte Armngton manufacturng goods n country. Addtonally 1 and 2 are share parameters and s the top level elastcty of substtuton n consumpton. The composte of manufactorng goods s defned as another reflectng the country from whch goods come. We assume ths level 2 composte consumpton s also of CES form and represented as, 16

17 1 ' 1 ' M ' M ' ' 1 j j j X [ x ], j country (12) where M x j s the consumpton of manufactory goods from country j n country. If j ths mples that ths country consumes ts domestcally produced tradable goods. j s the share parameter for country ' j ' s manufactorng goods consumed n country. s the elastcty of substtuton n level 2 preferences n country. Fg. 2 Nestng Structure of Producton and Consumpton Functons Producton Functon (CES) Consumpton Functon (Nested CES) Manufacturng and Non -Manufacturng Goods Consumpton Level 1 Non-Manufacturng Goods Manufacturng Goods Labor Captal Level 2 Chna US ROW Source: Compled by authors. The utlty optmzaton problem above yelds X E ( ) [ ( ) ( ) ] M 1 M M 1 NM 1 P 1 P 2 pc (13) X E ( ) [ ( ) ( ) ] NM 2 NM M 1 NM 1 pc 1 P 2 pc (14) where M P and NM pc are the separate consumpton prces of composte manufacturng goods and non-manufacturng goods n country. E s the total consumpton expendture of country. For the composte manufacturng goods whch enter the second level preferences and come from dfferent countres, the country specfc demands are 17

18 x ( X P ) M M M j j ' M M pcj j pcj j ' (1 ) ( ) [ ( ) ] (15) where M pc j s the consumpton prce n country of manufacturng goods produced n country j, X P s the total expendture on manufacturng goods n country. The M M consumpton prce for the composte of manufacturng goods s 15 M M j pcj j1 P 1 ' ' (1 ) 1 [ ( ) ] (16) and the total consumpton expendture of country s E P X pc X (17) M N NM NM Equlbrum n the model characterzed by market clearng prces for goods and factors n each country such that Q M x j M j (18) l K K, L L l l l (19) Later we dscuss the non-manufacturng goods market clearng condton. A zero proft condton must also be satsfed n each ndustry n each country, such that l l K l L l pq w K w L l M, NM (20) We ntroduce trade costs for parwse trade between countres (Anderson and Wncoop, 2003). Trade costs nclude not only mport tarffs but also other non-tarff barrers such as transportaton costs, language barrers, and nsttutonal barrers. We dvde trade costs nto two parts n our model; mport tarffs and non-tarff trade costs. We denote the mport tarff n 18

19 country as t, and non-tarff trade costs as N j (ad volume tarff-equvalent non-tarff trade costs for country mported from country j ). Ths yelds the followng relaton between consumpton prces and producton prces n country for country j ' s exports. M M pc (1 t N ) p (21) j j j We assume that trade costs are covered by mportng country. Import tarffs generate revenues R, whch are gven by R p x t (22) M M j j j, j Non-tarff trade barrers generate no revenue, and mporters need to use real resources to cover the costs nvolved. In the model, we assume that these resource costs are denomnated n terms of domestc non-manufacturng goods. We ncorporate ths resource usng feature through use of non-manufacturng goods equal n value terms to the cost of barrers. We assume reduced non-tarff trade costs (ncludng transportaton cost) wll thus occur under trade lberalzaton as an ncrease n non-manufacturng goods consumpton NR by the representatve consumer n mportng countres. The representatve consumer s ncome n country s gven by w K w L R I (23) K L and the demand-supply equalty nvolvng non-manufacturng goods becomes Q NM NR NM X (24) p NM where 19

20 NR p x N (25) M M j j j j, j The Asa bloc thus reduces both mport tarffs and non-tarff trade costs between member countres whch wll nfluence trade n the whole world. We assume an exogenously determned fxed trade mbalance, denoted as S, whch wll be postve when n trade surplus and negatve when n trade defct. Trade equlbrum wll nfluence ndvdual country s budget constrant. In the equlbrum, we have I E S (26) whch means that one country s total ncome equals ts total consumpton expendture plus ts surplus (trade mbalance), f one country has trade surplus then ts ncome wll more than consumpton expendture, but f one country has trade defct than ts ncome wll be less than consumpton expendture. 4.2 Trade Cost Calculaton Methodology Trade costs capture not only polcy barrers (Tarffs and Non-tarff barrers) but also transportaton costs (freght and tme costs) as well as communcaton and other nformaton costs, enforcement costs, foregn exchange costs, legal and regulatory costs and local dstrbuton costs. Fgure 3 reports the structure of representatve trade costs used by Anderson and Wncoop (2004) to llustrate conceptually what s nvolved. We calculate trade costs followng the approaches n Novy (2008) and Wong (2012). Ther method s to take the rato of blateral trade flows over local trade, scaled to some parameter values, and then use a measure that capture all barrers. Some have argued that ths 20

21 measure s consstent wth the gravty equaton and robust across a varety of trade models (Novy, 2008; Wong, 2012). Fg. 3 Representatve Trade Costs Trade Costs Transport Costs Border Related Trade Barrers Retal and Wholesale Dstrbuton Costs Freght Costs Transt Costs Polcy Barrers Language Barrer Currency Barrer Informaton Costs Securty Barrer Tarffs Non-tarff Barrers Source: Anderson and Wncoop (2004) and De (2006). The measure of trade barrer used here s based on the gravty equaton derved from Chaney s (2008) model of heterogeneous frms wth blateral fxed costs of exportng. Trade barrers can take two forms n the model, a varable trade barrer r and a fxed cost of exportng F r. The varable trade barrer r s an ceberg cost. In order to delver one unt of good to from r, r 1 unt of good has to be delvered. The gravty equaton supported by ths model s: X r Y ( 1) Yr wr r 1 ( ) Fr (27) Y Where X r s mport of country from country r. Y, Y r and Y are the economc szes of both countres and the total world, w r s labor costs, r s varable trade costs and F r s the fxed cost of exportng. The Pareto parameter governs the dstrbuton of frm productvtes. s the elastcty of substtuton n preferences. s a remoteness measure 21

22 for the mportng country whch captures trade dverson effects. The mechansm s that the further away s from the rest of the world, the more lkely that r could export more to due to less competton from thrd party countres n the mporter country. Ths has a smlar nterpretaton to the multlateral resstance term n Anderson and Wncoop (2003). We relate data on trade flows to unobservable trade barrers by takng ratos of blateral trade flows of two regons over local purchases of each of two countres: X X X X F F ( 1) 1 ( ) ( ) (28) F F r r r r r r rr rr rr Ths equaton reveals the relatonshp between observable trade data and unobservable trade barrers and elmnates the need to worry about the omsson of unspecfed or unobserved trade barrers. If the fxed costs of exportng are not blaterally dfferentated ( Fr Fr ) or are constant across locatons ( Fr F ), fxed costs drop out of ths measure and measured trade costs are nterpreted as varable trade costs, as n models wthout fxed export costs such as Eaton and Kortum (2002), Anderson and Wncoop (2003). For smplcty, we normalze own trade costs to 1,.e. 1 and F 1. Defnng the geometrc average of trade costs between the country par and r as t r 1 2 Xr Xr ( ) (29) X X rr We then get a measure of the average blateral trade barrer between country and r : X ( ) X rr r ( ) ( r r ) ( FrFr ) Xr Xr t (30) Data for ths equaton are relatvely easy to obtan, and so we have a comprehensve 22

23 measure of trade barrers, and the ad valorem tarff-equvalent blateral average trade cost between country and r can be wrtten as t r 1 2 X Xrr tr 1 ( ) 1 (31) X X r r Usng the trade costs equaton above, we can calculate actual trade costs between blateral country pars n our general equlbrum model. In the calculaton, X r and X r are separately exports and mports between countres and r. Due to market clearng, ntranatonal trade X or X rr can be rewrtten as total ncome mnus total exports (see equaton (8) n Anderson and Wncoop(2003)), X y X (32) Where X s the total exports, defned as the sum of all exports from country, whch s X X (33) r, r r In the trade cost calculaton, all trade data are from the UN Comtrade database. For y, GDP data are not sutable because they are based on value added, whereas the trade data are reported as gross shpments. In addton, GDP data nclude servces that are not covered by the trade data (Novy, 2008). It s hard to get ths ncome data accordng to such a defnton, so here we use GDP data mnus total servce value added. We get GDP data from the World Bank database, and the servce share of GDP data from World Development Indcators (WDI) of World Bank database, we then calculate results for GDP mnus servces. We take the value of to be 8.3 as n Eaton and Kortum (2002). We only use trade cost data for 2011 n our numercal general equlbrum model, calculaton results are shown n Table 2. 23

24 Table 2: Ad Valorem Tarff-Equvalent Trade Costs Between Countres n 2011 (Unt: %) Country US EU Chna Japan Korea Canada Mexco Inda AN CP BMSV CILMPT ODDC ODC ROW US EU Chna Japan Korea Canada Mexco Inda AN CP BMSV CILMPT ODDC ODC ROW Notes: (1) (1) BMSV denotes Brune+Malaysa+Sngapore+Vetnam, AN denotes Australa+New Zealand, CP denotes Chle+Peru, CILMPT denotes Camboda+Indonesa+Laos+Malaysa+Phlppne+Thaland. (2) We see group countres as a whole to calculate trade costs. Source: Calculated by authors. 24

25 5. Data and Parameters Calbraton We use 2011 as our base year n buldng a global benchmark general equlbrum dataset for use n calbraton and smulaton followng the methods set out n Shoven and Whalley (1992). There are ffteen countres n our model, and country group data s obtaned by addng ndvdual country data together. ROW data s obtaned from total world values mnus values for the other fourteen countres. For the two goods, we assume secondary ndustry (manufacturng) reflects manufactured goods, and prmary and tertary ndustres (agrculture, extractve ndustres, and servces) yeld non-manufactured goods. For the two factor nputs, captal and labor, we use total labor ncome (wage) to denote labor values for nputs by sector. All data are n bllon US dollars. We adjust some of the data for mutual consstency for calbraton purposes. All data except for the EU are from the World Bank database (World Development Indcate). We use agrculture and servce share of GDP data and GDP data to yeld producton data of manufacturng goods and non-manufacturng goods, and use captal/gdp ratos to yeld captal and labor nput n producton. The EU data are from EU statstcs, the currency unt s Euro, and we use annual average exchange rates to change them nto US dollar currency unts. These data are lsted n Table 3. Trade data between each par of countres are from the UN Comtrade database. We use ndvdual country total export and mport values to ndrectly yeld exports to and mports from the ROW, and add ndvdual country trade data to yeld country 25

26 group s trade data. Usng producton and trade data, we can then calculate each country s consumpton values. The trade data we use are lsted n Table 4. Table 3: Base Year Data Used For Calbraton and Smulaton (2011 Data) Country GDP T-G NT-G Balance Captal Labor T-G NT-G T-G NT-G USA EU Chna Japan Korea Canada Mexco Inda AN CP BMSV CILMPT ODDC ODC ROW Note: (1) Unts for producton, captal, labor, nsde money and endowments are all bllon US$, and labor here denotes factor ncome (wage). (2) AN denotes Australa+New Zealand, CP denotes Chle+Peru, BMSV denotes Brune+Malaysa+Sngapore+Vetnam, CILMPT denotes Camboda+Indonesa+Laos+Myanmar+Phlppne+Thaland. (3) ODDC denotes other developed countres, ncludng Swtzerland, Norway, Israel, Iceland; ODC denotes other developng countres ncludng Brazl, Russa, Egypt, Argentna, and South Afrca. (4) T-G denotes tradable goods producton; NT-G denotes non-tradable goods producton. (5) We add countres together to generate AN, CP, BMSV values. (6) We use world values mnus all ndvdual countres to generate ROW values. Sources: EU data from EU statstcs, and the currency unt s Euro, we use annual average exchange rate to change them nto US dollar currency unt; Other countres data are all calculated from WDI of World Bank database. Trade costs have two parts, mport tarffs and all other non-tarff barrers. We obtan each country s mport tarff data from WTO Statstcs Database. For ROW, we use world average tarff rates to denote these values. Import tarffs data are lsted n Table 5. We then can get non-tarff barrers by usng trade costs mnus mport tarffs. Table 5: Import Tarffs for Countres n 2011 (Unt: %) Country USA EU Chna Japan Korea Canada Mexco Inda Tarff Country AN CP BMSV CILMPT ODDC ODC ROW / Tarff / Notes: (1) Import tarffs here are smple average MFN appled tarff rates. (2) We use the average ndvdual country s mport tarff to get country groups mport tarff. (3) AN denotes Australa+New Zealand, CP denotes Chle+Peru, BMSV denotes Brune+Malaysa+Sngapore+Vetnam, CILMPT denotes Camboda+Indonesa+Laos+Myanmar+Phlppne+Thaland. (4) We use mport tarff of the world to denote the tarff for the ROW. Source: WTO Statstcs Database. 26

27 There are no avalable estmates of elastctes for ndvdual countres on the demand and producton sdes of the model. Many of the estmates of domestc and mport goods substtuton elastcty are around 2 (Betna et al, 2006), so we set all these elastctes n our model to 2 (Whalley and Wang, 2010). We perform senstvty analyss around these elastctes. Wth these data, we calbrate the model parameters. When used n model soluton these regenerate the benchmark data as an equlbrum for the model. Then, usng these parameters we can form a numercal global general equlbrum system, and can use ths system to calculate Debreu dstance measures for Asa trade arrangements. 27

28 Exporter Countres Table 4:Trade between Countres n 2011 (Unt: Bllon USD) Importer USA EU Chna Japan Korea Canada Mexco Inda AN CP BMSV CILMPT ODDC ODC ROW USA EU Chna Japan Korea Canada Mexco Inda AN CP BMSV CILMPT ODDC ODC ROW Notes: (1) AN denotes Australa+New Zealand, CP denotes Chle+Peru, BMSV denotes Brune+Malaysa+Sngapore+Vetnam, CILMPT denotes Camboda+Indonesa+Laos+Myanmar+Phlppne+Thaland. (2) We get trade data of AN, CP, and BMSV by addng separate country s trade data together, and these do not nclude nner trade between these group countres. (3)We get the ROW trade data by deductng from each country s total export, total mport and total world trade value. Sources: Unted Natons (UN) Comtrade database and WTO Statstcs. 28

29 6. Smulaton Results We use our 15-country global general equlbrum model to numercally calculate Debreu dstances of present arrangements from varous forms of an Asa trade bloc. We calculate dstances for ndvdual countres that are nvolved n the trade bloc and also dstances for whole trade bloc of countres. From these dstances, we can assess how close Asa s already to a surrogate trade bloc, and analyze the effects of dfferent Asa trade bloc arrangements. The man large countres n the Asa regon nclude Chna, Japan, Korea, Inda, and ASEAN; and they are near each other n geography. Therefore, a typcal Asa bloc should nclude these four countres and one country group. In the meanwhle, an Asa trade bloc may have two dfferent types. One s lke NAFTA, and n ths all member countres do not have tarffs between each other but each has dfferent tarff levels for outsde countres. We call ths type as an Asa FTA. The other s lke the EU, all member countres do not have tarff between each other but they also have the same tarff level to outsde countres. We call ths type as Asa Unon. ASEAN Plus Three (APT) and RCEP are presently mportant FTA arrangement negotatons and they may develop to an Asa FTA, we also take account of these two trade blocs. We are also nterested n the dstance to global free trade. Therefore, the paper wll analyze fve dfferent scenaros: Asa FTA, Asa Unon, ASEAN Plus Three, RCEP, and global free trade. We use trade costs n our global general equlbrum model, to nclude both tarff and non-tarff barrers. We assume that all these Asa FTA arrangements wll remove 29

30 the whole of tarffs and partal non-tarff barrers. For non-tarff barrer removal, we use fve dfferent assumptons, whch are 0% non-tarff barrer elmnaton (tarff only), 20% non-tarff barrer elmnaton, 40% non-tarff barrer elmnaton, 60% non-tarff barrer elmnaton, and 80% non-tarff barrer elmnaton. We separately calculate Debreu dstances under ths fve dfferent alternatves. For the dstance value, t shows how much less resources under Asa trade bloc scenaro are needed to attan the same level of utlty satsfacton as under the present scenaro. The dstance ndcator s the shrnkage percent for the total endowment. If the dstance ndcator s negatve, ths means an ncreased percent rato of total endowment are needed to keep the orgnal welfare level, so that FTA arrangements have negatve effects. 6.1 Dstance Measures to An Asa FTA Asa FTA s the scenaro that Chna, Japan, South Korea, Inda, and ASEAN form a trade bloc. ASEAN has 10 countres; we separate them nto two country groups of BMSV and CILMPT n our numercal general equlbrum model. We calculate dstances for the present stuaton to an Asa FTA. Table 6, Fgure 4 and Fgure 5 lst these results. We fnd that nearly all the results are postve except the whole and Chna under only tarff elmnaton case. Ths reveals that almost all FTA member countres wll gan. But f Asa FTA only remove tarff between countres and has no effect to non-tarff removal, Chna wll be hurt. The reason s that Chna s mport tarff s comparatvely hgh, and Asan countres are not Chna s man trade partner, under 30

31 such crcumstance, when only tarffs removed between members, trade condton effects wll make Chna lose. Country/Trade Cost Table 6: Dstance Measures for the Present Stuaton to Asa FTA (%) Tarff Only T+20%NT T+40%NT T+60%NT T+80%NT Chna Japan Korea Inda BMSV CILMPT WHOLE Note: (1) Tarff only denotes all tarff removed between FTA member countres, T+20% NT denotes all tarff and 20% non-tarff removed between FTA member countres, others are the same for T+40% NT, T+60% NT and T+80% NT. (2) CILMPT denotes countres group of Camboda+Indonesa+Laos+Myanmar+Phlppne+Thaland, BMSV denotes countres group of Brune+Malaysa+Sngapore+Vetnam, NT denotes non-tarff barrer. Source: Calculated by authors. Chna Japan Dstance(%) %NT 20%NT 40%NT 60%NT 80%NT Dstance(%) %NT 20%NT 40%NT 60%NT 80%NT Korea Inda Dstance(%) %NT 20%NT 40%NT 60%NT 80%NT Dstance(%) %NT 20%NT 40%NT 60%NT 80%NT BMSV CILMPT Dstance(%) %NT 20%NT 40%NT 60%NT 80%NT Dstance(%) %NT 20%NT 40%NT 60%NT 80%NT Fg. 4 Dstances of Indvdual Countres to Asa FTA Source: Calculated by Authors. As the non-tarff barrers are removed more, the dstance value becomes bgger, whch means that all ndvdual countres wll gan more as non-tarff barrers elmnate more. For ndvdual countres, BMSV wll gan the most, then CILMPT, 31

32 Inda, Korea, Japan, and the least s Chna. Chna s low beneft s also because of Chna s hgh mport tarff and Asa countes are not Chna s man trade partner. Dstance to Asa FTA Under T+40%NT Elmnaton Case Dstance(%) BMSV CILMPT Chna Inda Japan Korea Fg. 5 Dstances to Asa FTA under T+40%NT Case Source: Calculated by Authors. For the specfc values, we take T+80%NT elmnaton case as an example. The Asa FTA countres as a whole can shrnk 1.6% of ther endowment to obtan utltes for each country n the trade bloc not less than ther former level. 6.2 Dstances to Asa Unon Asa Unon s the scenaro of Chna, Japan, South Korea, and ASEAN formng a free trade area and they have the same tarff level to outsde countres n the meanwhle. The Asa Unon s same outsde tarff can choose dfferent levels, we assume t to separately choose Chna tarff level, Japan tarff level and Inda tarff level. Table 7 and Fgure 6 show all dstance calculaton results. We fnd the results are nearly the same as Asa FTA, when Asa Unon only elmnate tarffs, Chna wll be hurt under Chna tarff level and Japan tarff level. All 32

33 of other member countres n all other stuatons wll gan from the Asa Unon arrangement. In the meanwhle, the more non-tarffs are removed the more the gan of member countres. Comparatvely, ASEAN countres gan the most, and then Inda, Korea, Japan and Chna. These show that comparatvely larger countres gan less from FTAs. Country/Trade Cost Table 7: The Dstance of Present Stuaton to Asa Unon (%) Tarff Only T+20%NT T+40%NT T+60%NT T+80%NT Chna Tarff Level Chna Japan Korea Inda BMSV CILMPT WHOLE Japan Tarff Level Chna Japan Korea Inda BMSV CILMPT WHOLE Inda Tarff Level Chna Japan Korea Inda BMSV CILMPT WHOLE Note: (1) Tarff only denotes all tarff removed between FTA member countres, T+20% NT denotes all tarff and 20% non-tarff removed between FTA member countres, others are the same for T+40% NT, T+60% NT and T+80% NT. (2) CILMPT denotes countres group of Camboda+Indonesa+Laos+Myanmar+Phlppne+Thaland, BMSV denotes countres group of Brune+Malaysa+Sngapore+Vetnam, NT denotes non-tarff barrer. Source: Calculated by authors. Compared wth Asa FTA arrangement, we fnd that most countres wll gan more n Asa Unon than n Asa FTA under Chna tarff level case and Inda tarff level case. But f the Asa Unon takes the Japan tarff level as ther outsde tarff 33

34 value, most countres wll gan less than n Asa FTA scenaro. Chna Tarff Level Japan Tarff Level Inda Tarff Level BMSV BMSV BMSV CILMPT CILMPT CILMPT Chna Chna Chna Inda Inda Inda Japan Japan Japan Korea Korea Korea WHOLE WHOLE WHOLE Dstance(%) Dstance(%) Dstance(%) Fg. 6 Dstances of Whole to Asa Unon under T+40% NT Elmnaton Case 6.3 Dstances to ASEAN+3 Source: Calculated by Authors. ASEAN Plus Three (APT) s the scenaro of Chna, Japan and South Korea formng a free trade area. Although the present ASEAN+3 s just a forum, but t s possble to form an FTA, so we nclude ths scenaro for analyss. Table 8 and Fgure 7 show the dstance calculaton results. Country/Trade Cost Table 8: The Dstance of Present Stuaton to ASEAN+3 (%) Tarff Only T+20%NT T+40%NT T+60%NT T+80%NT Chna Japan Korea BMSV CILMPT WHOLE Note: (1) Tarff only denotes all tarff removed between FTA member countres, T+20% NT denotes all tarff and 20% non-tarff removed between FTA member countres, others are the same for T+40% NT, T+60% NT and T+80% NT. (2) CILMPT denotes countres group of Camboda+Indonesa+Laos+Myanmar+Phlppne+Thaland, BMSV denotes countres group of Brune+Malaysa+Sngapore+Vetnam, NT denotes non-tarff barrer. Source: Calculated by authors. The same as n Asa FTA and Asa Unon scenaros, all member countres wll 34

35 gan n all stuatons except Chna under only tarff elmnaton scenaro. All countres and the whole wll gan more as non-tarff removes more. Comparatvely, ASEAN countres wll gan the most, and then do Korea, Japan and Chna. Dstance of WHOLE to APT Dstance(%) T+0% NT T+20% NT T+40% NT T+60% NT T+80% NT Fg. 7 Dstances of Whole to ASEAN+3 Source: Calculated by Authors. Compared wth other Asa trade bloc arrangements, ASEAN Plus Three wll beneft member countres less than Asa FTA and Asa Unon. In specfc dstance values, the whole trade bloc countres can shrnk 1.1% of ther endowment to obtan utltes not less than ther present level. 6.4 Dstances to RCEP RCEP nvolves countres of Chna, Japan, Korea, Inda, Australa and New Zealand. RCEP s amng to reach an agreement by the end of 2015; t may become an mportant FTA n Asa. Table 9 and Fgure 8 show all dstance calculaton results. The effects are nearly the same as Asa FTA. When RCEP just elmnates tarff between member countres, Chna and the trade bloc as a whole wll be hurt whch means RCEP cannot beneft Chna and all countres n the bloc as a whole. Under all 35

36 other stuatons, all countres wll gan, so RCEP should negotate manly on non-tarff barrers. As non-tarff barrers remove more, all member countres wll beneft more from t. Comparng dstances of ndvdual countres, Chna s the nearest country, and ASEAN s the most far country group, whch means that Chna wll gan the least and ASEAN the most. Country/Trade Cost Table 9: The Dstance of Present Stuaton to RCEP (%) Tarff Only T+20%NT T+40%NT T+60%NT T+80%NT Chna Japan Korea Inda AN BMSV CILMPT WHOLE Note: (1) Tarff only denotes all tarff removed between FTA member countres, T+20% NT denotes all tarff and 20% non-tarff removed between FTA member countres, others are the same for T+40% NT, T+60% NT and T+80% NT. (2) CILMPT denotes countres group of Camboda+Indonesa+Laos+Myanmar+Phlppne+Thaland, BMSV denotes countres group of Brune+Malaysa+Sngapore+Vetnam, AN denotes countres group of Australa+New Zealand, NT denotes non-tarff barrer. Source: Calculated by authors. Dstance of Whole to RCEP T+0% NT T+20% NT T+40% NT T+60% NT T+80% NT Dstance(%) Fg. 8 Dstances of Whole to RCEP Source: Calculated by Authors. Comparng the effects of RCEP wth other FTA arrangements, ts postve effects on member countres s nearly the same as Asa FTA and Asa Unon, and the effects 36

37 are stronger than ASEAN Plus Three (APT). 6.5 Dstances to Global Free Trade Dstances to global free trade (GFT) s an nterestng topc, and we can compare Asa trade bloc effects wth global free trade effects. We calculate dstances of each ndvdual county n our model and dstances of all countres as a whole. Table 10 and Fgure 8 report all these results. Country/Trade Cost Table 10: The Dstance of Present Stuaton to Global Free Trade (%) Tarff Only T+20%NT T+40%NT T+60%NT T+80%NT US EU Chna Japan Korea Canada Mexco Inda AN CP BMSV CILMPT ODDC ODC ROW WHOLE Note: (1) Tarff only denotes all tarff removed between FTA member countres, T+20% NT denotes all tarff and 20% non-tarff removed between FTA member countres, others are the same for T+40% NT, T+60% NT and T+80% NT. (2) CILMPT denotes countres group of Camboda+Indonesa+Laos+Myanmar+Phlppne+Thaland, BMSV denotes countres group of Brune+Malaysa+Sngapore+Vetnam, AN denotes countres group of Australa+New Zealand, ODDC denotes other developed countres, ODC denotes other developng countres, NT denotes non-tarff barrer. Source: Calculated by authors. All countres n all stuatons except only tarff elmnaton case wll gan from global free trade. When the global free trade s just tarff elmnaton, the US, the EU, Chna, Japan, Mexco, CILMPT and other developng countres (ODC) wll lose, ths may be caused by trade condton effects. Ths proves that global trade lberalzaton n the future should manly focus on non-tarff elmnaton. Another mplcaton of the 37